Psychology of Bilingualism The Cognitive and Emotional World of Bilinguals by Alfredo Ardila, Anna B. Cieślicka, Roberto R. Heredia, Mónica Roselli

The Bilingual Mind and Brain Book Series 5
Alfredo Ardila
Anna B. Cieślicka
Roberto R. Heredia
Mónica Roselli Editors
Psychology
of
Bilingualism
The Cognitive and Emotional World of
Bilinguals
The Bilingual Mind and Brain Book Series
Volume 5
Series editors
Roberto R. Heredia, Department of Psychology and Communication,
Texas A&M International University, Laredo, TX, USA
Anna B. Cieślicka, Department of Psychology and Communication,
More information about this series at http://www.springer.com/series/13841

Alfredo Ardila • Anna B. Cieślicka
Roberto R. Heredia • Mónica Roselli
Editors
Psychology of Bilingualism
The Cognitive and Emotional World
of Bilinguals
Editors
Alfredo Ardila Anna B. Cieślicka
Communication Sciences and Disorders Department of Psychology and
Florida International University Communication
Miami, FL, USA Texas A&M International University
Laredo, TX, USA
Roberto R. Heredia
Department of Psychology and Mónica Roselli
Communication Department of Psychology
Texas A&M International University Florida Atlantic University
Laredo, TX, USA Davie, FL, USA
The Bilingual Mind and Brain Book Series

ISBN 978-3-319-64097-6 ISBN 978-3-319-64099-0 (eBook)
DOI 10.1007/978-3-319-64099-0
Library of Congress Control Number: 2017955224
© Springer International Publishing AG 2017

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology
now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the
editors give a warranty, express or implied, with respect to the material contained herein or for any errors
or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims
in published maps and institutional affiliations.
Printed on acid-free paper
This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
It is with great joy that we present The Psychology of Bilingualism: The Cognitive
World of Bilinguals to students, teachers of bilingualism, and the multilingual scien-
tific community. The goal of this volume is to provide an interdisciplinary perspec-
tive of bilingualism from fields such as educational psychology, cognitive linguistics
and cognitive psychology, developmental psychology, and clinical and personality
psychology. The book looks at the bilingual as an individual exploring his/her inter-
nal world across a wide range of dimensions, such as emotional word processing,
the interaction of personality traits and bilingualism, language effects on the mind,
cognitive effects of bilingualism, as well as the psychopathology of bilingualism.
To our knowledge, no other published book has integrated these perspectives to look
at the bilingual mind. The Psychology of Bilingualism: The Cognitive World of
Bilinguals is intended as a professional reference by the beginning and seasoned
bilingual researcher, as well as by communication disorder practitioners, school
teachers/counselors interested in bilingual language processing and representation,
and also diverse types of professionals potentially interested in this book, including
clinical and counseling psychologists.
The book is divided into four parts. The first part includes three chapters. The
chapter Linguistic Relativity in Conceptual Metaphors by Huang and Tse analyzes
the influence of language on human cognition; that is, to what extent human lan-
guage influences cognition. The authors first present the linguistic relativity hypoth-
esis, the Conceptual Metaphor Theory, and the relationship among bilingualism,
conceptual metaphor, and embodiment. Then, they focus on space–time metaphor,
a type of conceptual metaphor, to depict the impact of the bilinguals’ spatial lan-
guage on their mental representation of time. Finally, it is concluded that despite the
mixed evidence in the literature, most of the current findings support the linguistic
relativity hypothesis that spatial language does have an impact on bilingual’s mental
representation of time. The second chapter, Inner Speech in Bilinguals: The Example
of Calculation Abilities by Ardila and Rosselli, argues that the internal language
in bilinguals can be the first (L1) or second (L2) language depending on diverse
variables. As a specific example of the use of inner language, it is shown that inner
language can be L1 or L2, depending on the context in which a specific mathemati-
v
vi Preface
cal ability was learned. In Emotional Associations of Words in L1 and L2 in

Bilinguals, Rosselli, Vélez-Uribe, and Ardila argue that in early—simultaneous—
proficient bilinguals, the processing of emotion words is identical in the two lan-
guages, whereas emotional dissociations are more likely in late sequential
non-proficient bilinguals.
Part II includes four chapters. It starts with Cieślicka’s chapter on Bilingual
Figurative Language Processing, which provides a state-of-the-art theoretical and
empirical account of how bilingual speakers comprehend different types of figura-
tive language such as idiomatic expressions (let the cat out of the bag), metaphors
(cats are potato chips), proverbs (curiosity killed the cat), and irony. The chapter
discusses major themes that have emerged in the bilingual figurative language
research, such as the storage and retrieval format of figurative expressions, cerebral
differences in L1 vs. L2 figurative processing, and the role of L1 knowledge in the
course of L2 figurative processing. The following chapter, The Cost of Being
Bilingual: The Example of Verbal Fluency and Switching, by Altarriba and Kazanas
points out that researchers have noted consistent cognitive costs in lexical access
among bilingual participants. Bilingual deficits on fluency measures are observed
across the life span, though recent findings would argue that deficits are associated
with diverse variables, including language proficiency and everyday switching fre-
quency. This chapter outlines these findings, as well as the current debate surrounding
the bilingual advantage: a stronger executive function, as measured with behavioral
and neuroimaging methods. In Bilingual Episodic Memory Processes, Heredia and
García provide a general overview of some of the major theoretical approaches
to encoding (e.g., levels of processing, the generation, distributed-practice, and
repetition effects) and retrieval (e.g., encoding specificity principle and transfer appro-
priate processing) in bilingual episodic memory. As pointed out by Heredia and
García, an overview of the published findings suggests that, like the classic monolin-
gual general findings, bilingual retrieval is enhanced by factors such as deep process-
ing, elaboration, distinctiveness, and spaced practice repetitions, as opposed to
shallow processing and massed repetitions. Moreover, the authors conclude that
there is a propensity for between-language conditions (i.e., translations) to enhance
memory retrieval, relative to within-language conditions (i.e., monolingual condi-
tions), suggesting that translations are more distinctive and require more mental
effort and mental elaboration. The final chapter, Active–Passive Bilingualism and the
Functional Distance between L1 and L2 as Viewed within one Unifying Theoretical
Framework, by Sharwood Smith examines the way in which a particular theoretical
perspective—The Modular Online Growth and Use of Language (MOGUL) frame-
work—sheds light on two quite different phenomena: (1) active–passive bilingual-
ism and (2) the functional distance between L1 and L2. It is shown that the two
phenomena under discussion, when set in a context that includes the mind as a whole,
can be most satisfactorily explained as examples of the operation of general princi-
ples of cognition as well as of principles that are specifically linguistic.
The third part includes two chapters. In Aging in Bilinguals: Normal and
Abnormal, Lerman and Obler point out that in bilinguals, changes in language abili-
ties across the adult life span are not necessarily parallel in both languages. During
Preface vii
normal aging, difficulties with language skills such as lexical retrieval are further
confounded in bilinguals by differences in language proficiency and dominance,
age of acquisition and language use, as well as the types of assessments and stimuli
used to test them. In abnormal aging, such as dementia, these changes in language
use and abilities become highly variable and often more extreme than in normal
aging. Ardila’s chapter, Dissociated Language Disorders in Bilinguals, illustrates
that bilingual individuals can sometimes present language disorders that are not
completely equivalent in their two languages. This situation can be observed in both
acquired and developmental language disorders. Dissociated language disorders
can be found not only in oral language (acquired aphasia and developmental dys-
phasia) but also in written language (acquired alexia and developmental dyslexia).
These dissociations, Ardila argues, suggest a nonequivalent brain organization of
L1 and L2.
In the final part, the chapter Psychopathology and Bilingualism by Terrazas-
Carrillo examines the literature documenting the relationship of various forms of
psychopathology to bilingualism. It is observed that L2 acquisition impacts the
same psychological domains; thus, psychopathology and bilingualism may have a
shared influence in the cognitive, emotional, and behavioral domains. Evidence of
differential symptomology observed across languages is presented. This chapter
also includes a discussion of the methodological and theoretical shortcomings pres-
ent in much of the literature exploring the mechanisms and influence of bilingual-
ism and psychopathology. The chapter Personality Traits in Bilinguals by Rosselli,
Vélez-Uribe, and Ardila analyzes the effect of language of test administration on the
results of personality inventories using a sample of Spanish–English bilinguals; it is
concluded that different personality profiles are found when personality inventories
are administered in L1 and L2. The final chapter, Cognitive Abilities in Bilinguals
when Tested in L1 and L2, by Pham, Castro-Olivo, Chun, and Goforth, examines
the discrepancies in cognitive test scores when bilinguals are tested in L1 and L2,
or using a bilingual approach. That means that the profile of abilities is not coinci-
dental across languages. This observation has crucial consequences not only from
the educational point of view but also for understanding cognition in bilinguals and
multilinguals.
We wish to thank the many people who made The Psychology of Bilingualism:
The Cognitive World of Bilinguals a reality. First, we thank Morgan Ryan of Springer
for catching the vision of this unique work. Also, we would like to express our grati-
tude to the contributors who have accepted our invitation to participate in this volume
and share their expertise. We thank them for their patience and cooperation throughout
the process and for graciously accepting our editorial suggestions.
Miami, FL, USA Alfredo Ardila

Laredo, TX, USA Anna B. Cieślicka
Laredo, TX, USA Roberto R. Heredia
Davie, FL, USA Mónica Roselli
Acknowledgements
As series editors of Bilingual Mind and Brain, we would like to thank Alfredo
Ardila and Mónica Roselli for their vision and original idea of The Psychology of
Bilingualism: The Cognitive World of Bilinguals. We thank them for sharing their
vision and for inviting us as coeditors and contributors to this volume. We thank
Morgan Ryan of Springer for her continual support and encouragement and always
responding to our queries in a timely fashion, as well as for her trust in our abilities
to make the best editorial decisions. Also, we would like to express our gratitude to
our contributors who kindly agreed to be part of this project and helped us capture
the unique vision of this volume by sharing their expert knowledge.
I, Roberto, feel grateful for all the support of my colleagues and dear friends.
In particular, I am thankful to my coauthor, coeditor, colleague, and dear friend,
Anucha Cieślicka, for her faith in my work and for her willingness to always stop
whatever she is doing to listen to my crazy ideas (spoken or written) and provide me
with genuine and objective feedback. I am grateful to my undergraduate and graduate
students who continuously inspire me and indirectly contribute to my work. I must
say that the idea of the Heredia and García’s chapter, in this volume, originated as
I lectured on long-term memory and episodic memory in cognitive psychology and
learning and memory. I am also grateful to my beloved Michelle, for her love,
patience, and support, and Andrea my teenage daughter! I dedicate this book to my
father Eliseo, my mother Esperanza, and a wonderful human being that I had the
opportunity to meet, and who is now brewing her own beer in heaven, rest in peace,
Mama Tamara!
I, Ania, want to thank all my colleagues, friends, and past and current collabora-
tors—it is through many inspiring discussions and exchanging research ideas with
them that my own research path has been shaped and is constantly developing.
Special thanks to Roberto Heredia, my dear colleague and friend, for his never-
ending resourcefulness in coming up with research and project ideas, for his superb
professionalism in approaching editorship work, and for making our teamwork such
an exhilarating and fulfilling experience. My students deserve a special recognition
for being a constant source of inspiration by asking questions and helping me make
sure I never stop formulating new questions either. Finally, I need to recognize my
ix
x Acknowledgements
cat rescues, both those that are gone (Kacper, Bambi, Chmurka, Limpuś, Lolek,
Magus, and Szarik) and those still with me (Daktyl, Pisia, Tofi, Minia, Marusia,
Puszek, Sówka, Grusia, and Rudy) whose unconditional love and purring company
have made long hours behind my desk a most gratifying experience. Over and over,
they keep proving Aldous Huxley’s famous quote, If you want to write, keep cats, to
be very true! I dedicate this book to my beloved parents: my father Jerzy and my
mother Tamara Cieślicka.
Roberto R. Heredia and Anna B. Cieślicka

Texas A&M International University
Laredo, TX, USA
Contents
Part I The Internal World of the Bilingual Person

1 Linguistic Relativity in Conceptual Metaphors�� 3
Yanli Huang and Chi-Shing Tse
2 Inner Speech in Bilinguals: The Example of Calculation Abilities �� 27
Alfredo Ardila and Mónica Roselli
3 Emotional Associations of Words in L1 and L2 in Bilinguals�� 39
Mónica Roselli, Idaly Vélez-Uribe, and Alfredo Ardila
Part II Bilingual Language Representation

4 Bilingual Figurative Language Processing�� 75
Anna B. Cieślicka
5 The Cost of Being Bilingual: The Example of Verbal
Fluency and Switching�� 119
Jeanette Altarriba and Stephanie A. Kazanas
6 Bilingual Episodic Memory Processes�� 137
Roberto R. Heredia and Elva A. García
7 Active–Passive Bilingualism and the Functional
Distance Between L1 and L2 as Viewed Within One Unifying
Theoretical Framework�� 157
Michael Sharwood Smith
Part III Bilingualism Across the Life-Span

8 Aging in Bilinguals: Normal and Abnormal �� 189
Aviva Lerman and Loraine K. Obler
9 Dissociated Language Disorders in Bilinguals�� 211
Alfredo Ardila
xi
xii Contents
Part IV Bilingual Cognitive and Personality Dimensions

10 Psychopathology and Bilingualism�� 227
Elizabeth Terrazas-Carrillo
11 Personality Traits in Bilinguals�� 259
12 Cognitive Abilities in Bilinguals in L1 and L2�� 269
Andy V. Pham, Sara Castro-Olivo, Heejung Chun,
and Anisa N. Goforth
Index�� 293
Contributors
Jeanette Altarriba University at Albany, State University of New York, Albany,

NY, USA
Alfredo Ardila Florida International University, Miami, FL, USA
Sara Castro-Olivo Texas A&M University, College Station, TX, USA
Heejung Chun New Mexico State University, Las Cruces, NM, USA
Elva A. García Texas A&M International University, Laredo, TX, USA
Anna B. Cieślicka Texas A&M International University, Laredo, TX, USA
Anisa N. Goforth University of Montana, Missoula, MT, USA
Roberto R. Heredia Texas A&M International University, Laredo, TX, USA
Yanli Huang University of Hong Kong, Pokfulam, Hong Kong
Stephanie A. Kazanas Tennessee Technological University, Cookeville, TN, USA
Aviva Lerman The Graduate Center of the City University of New York, New
York, NY, USA
Loraine K. Obler The Graduate Center of the City University of New York, New
York, NY, USA
Andy V. Pham Florida International University, Miami, FL, USA
Mónica Roselli Florida Atlantic University, Davie, FL, USA
Michael Sharwood Smith The University of Edinburgh, Edinburgh, UK
Heriot-Watt University, Edinburgh, UK
Elizabeth Terrazas-Carrillo Texas A&M International University, Laredo, TX, USA
xiii
xiv Contributors
Chi-Shing Tse The Chinese University of Hong Kong, Kong, Shatin, NT,
Hong Kong
Idaly Vélez-Uribe Florida Atlantic University, Davie, FL, USA
About the Authors
Alfredo Ardila received his Ph.D. from the Moscow State University in the field
of neuropsychology. Currently, he is Professor at the Florida International University.
His primary research interests include brain organization of cognition, historical
origin of human cognition, aphasia, and bilingualism. He has received several
awards and honors, such as the Latin American award in neurosciences.
Anna B. Cieślicka received her Ph.D. from Adam Mickiewicz University in

Poznań, Poland, in the field of psycholinguistics. She is currently Associate
Professor at Texas A&M International University (TAMIU) and Director of the MS
in Psychology Graduate Program. Her research focuses mainly on the psycholin-
guistics of second/foreign language acquisition and processing, bilingual lexicon,
figurative language, and neuropsychology of bilingualism. She is recipient of
TAMIU’s Teacher of the Year and Scholar of the Year Awards and Alpha Delta
Kappa Golden Apple Award for teaching excellence in higher education. She is
cofounder and coeditor of The Bilingual Mind and Brain Book Series, published by
Springer.
Roberto R. Heredia is Regents Professor at Texas A&M International University

(TAMIU); his research interests include bilingual lexical processes, evolutionary
psychology, figurative language processes, sentence processing, the neuroscience of
bilingualism and cognition, bilingual memory, and information processing. He is a
former Chair of the Behavioral Sciences department at TAMIU. He is cofounder
and coeditor of The Bilingual Mind and Brain Book Series, published by Springer.
Mónica Roselli received her Ph.D. in biomedical sciences (neuropsychology)

from the National Autonomous University of Mexico. She is currently Assistant
Chair and Professor in the Department of Psychology at Florida Atlantic University,
and heads the Neuropsychology Laboratory at FAU. She is an investigator of the
Florida Alzheimer Disease Research Center. Her research interests are in the areas
of bilingualism, neuropsychological assessment, neurocognitive development,
cross-cultural neuropsychology, and geriatric neuropsychology.
xv
Part I
The Internal World of the Bilingual Person
Chapter 1
Linguistic Relativity in Conceptual Metaphors
Yanli Huang and Chi-Shing Tse
Contents
Linguistic Relativity Hypothesis 3
Conceptual Metaphor Theory 6
Linguistic Relativity on Conceptual Metaphors 9
A Case Study of Linguistic Relativity: Space–Time Metaphors 10
Conclusion 18
References 22
Linguistic Relativity Hypothesis
Language is a communicative tool to convey information, such as to describe things,

to share ideas with others, and to greet friends, which is a system of signs governed
by grammar. Language is used almost everywhere and every time and we cannot
imagine what our life would be like without language. There are around 7000 dis-
tinct languages in the world and they differ from one another in numerous ways
(Boroditsky, 2011). Does this mean that people speaking different languages differ
Y. Huang (*)
Faculty of Education, Centre for Advancement of Chinese Language Education and Research,
University of Hong Kong, Pokfulam, Hong Kong
e-mail: yanlihuang1010@gmail.com
C.-S. Tse
Department of Educational Psychology, The Chinese University of Hong Kong,
Kong, Shatin, NT, Hong Kong
Centre for Learning Sciences and Technologies, The Chinese University of Hong Kong,
Kong, Shatin, NT, Hong Kong
e-mail: cstse@cuhk.edu.hk
© Springer International Publishing AG 2017 3

A. Ardila et al. (eds.), Psychology of Bilingualism, The Bilingual Mind and
Brain Book Series, DOI 10.1007/978-3-319-64099-0_1
4 Y. Huang and C.-S. Tse
in their ways of perception and thinking of the world? Whether or not language
could influence cognition has still been debated in psychology and linguistics.
This question begins to be increasingly investigated since the writings by
Edward Sapir and Benjamin Whorf in 1940s, known as the Sapir-Whorf hypoth-
esis, the most widely known hypothesis on the language–cognition relationship.
Two views originate from the Sapir-Whorf hypothesis: linguistic determinism
and linguistic relativity (Hardin & Banaji, 1993; Hunt & Agnoli, 1991; Kay &
Kempton, 1984). Linguistic determinism refers to the view that the structure of
a language strongly influences or fully determines the way its native speakers
perceive the world. This strong view has largely been disconfirmed (e.g., Au,
1983; Berlin & Kay, 1969; Heider, 1972; Heider & Olivier, 1972). Although
linguistic determinism gained little empirical support, some researchers have
still enthusiastically considered the weaker view, linguistic relativity, which pos-
tulates that the structural differences between languages are generally paral-
leled by non-linguistic cognitive differences in the native speakers of the two
languages. Casasanto (2008) proposed two important questions, do people think
in language? (i.e., are people’s thoughts determined by the categories made
available by their language) and does language shape thought? (i.e., do differ-
ences among languages cause differences in the thoughts of their speakers), to
point out the key difference between the strong (linguistic determinism) and
weak (linguistic relativity) views of the Sapir-Whorf hypothesis. Casasanto also
pointed out the relationship between the two views. If people think in language,
it has to be the case that language shapes thought. However, if people do not
think in language, it is still possible that language shapes thought. In this chap-
ter, we focus on the latter view; that is, whether or not language can shape
thought. The linguistic relativity hypothesis claims that language does not
strictly determine thought, but can shape speakers’ cognition.
Several studies have been conducted to test the linguistic relativity hypothesis in
various domains, such as color, number, space, time, and musical pitch. Although
the physiological basis of these concepts may be the same for all humans, they
express these concepts in different ways in their own languages. Given the diversity
of languages, researchers have attempted to test the impact of language on cognition
through cross-linguistic studies. Taking color perception as an example, Berlin and
Kay (1969) reported that there are different numbers of basic color terms in differ-
ent languages. While some languages only have two basic color terms (black and
white, e.g., Dugum Dani; Heider, 1970), others may have as many as 12 basic color
terms (e.g., Russian and Turkish; Ozgen & Davies, 1998). However, whether or not
the color categorization is universal in mental representation remains debatable.
Universalists hold that color perception is innate and universal. Berlin and Kay
(1969) argued that, despite the diverse number of basic color terms across lan-
guages, they do so in a systematic manner (see for example, the hierarchy of basic
color terms in Berlin and Kay). Studies with prelinguistic infants have demonstrated
categorical perception (i.e., discrimination) of color (e.g., Franklin & Davies, 2004),
suggesting that color perception is not coded linguistically, supporting the universal
1 Linguistic Relativity 5
view. However, proponents of the linguistic relativity hypothesis argue that lan-
guage plays a role in shaping color perception after people acquire terms that are
used to express specific colors.
The linguistic relativity hypothesis suggests that the categorical perception
of color varies as language varies, as supported by the findings of cross-lin-
guistic studies. For example, Roberson and his colleagues (e.g., Roberson &
Davidoff, 2000; Roberson, Davies, & Davidoff, 2000) examined the relation-
ship between language and categorical perception of color in two populations
whose languages coded color category differently. The Berinmo language (in
Papua New Guinea) uses one term to express blue and green colors, whereas
English uses two terms. In addition, Berinmo distinguishes the nol–wor colors
(wor refers to the leaves ready to fall from a tree and covers a range of yellow,
orange, brown, and khaki; and nol means live and covers green, yellow-green,
blue, and purple), but English does not. In a variety of tasks, native English
speakers have shown categorical perception for stimuli across the green–blue
boundary, but not for stimuli across nol–wor boundary. The reverse pattern has
been found with Berinmo speakers. In addition, native English speakers show
a faster discrimination between blue and green than among arbitrary colors
within the green category, and faster yellow–green discrimination than nol–
wor discrimination. Conversely, the nol–wor discrimination is easier for native
Berinmo speakers than yellow–green discrimination. Similar findings have
been observed in other cross-linguistic studies, such as the categorical percep-
tion between lighter blue and darker blue, which are considered as two color
categories (goluboy and siniy) in Russian but only one (blue) in English
(Winawer et al., 2007). Native Russian speakers’ color discrimination was
faster when colors belonged to different linguistic categories in Russian (e.g.,
one color belonging to goluboy and the other one belonging to siniy) than when
colors came from the same category (e.g., both were goluboy or siniy). These
results showed that language indeed influences color discrimination, thus pro-
viding support for the linguistic relativity hypothesis.
Further, the categorical perception of color has been found to be disrupted by
verbal, but not visual, interference, indicating the role of language in online
color processing (Gilbert, Regier, Kay, & Ivry, 2006; Winawer et al., 2007).
Franklin et al. (2008) showed that categorical perception was stronger in the
right visual field, which is more involved in the language-dominated left hemi-
sphere (see also Gilbert et al., 2006; Roberson, Pak, & Hanley, 2008). In addi-
tion, neuroimaging studies showed that easy-to-name color squares evoked
stronger activation (as compared to difficult-to-name color squares) in the infe-
rior parietal lobule and left posterior superior temporal gyrus, which are respon-
sible for linguistic processing, indicating the neural basis of the linguistic
relativity hypothesis.
For the concepts of color, like other concrete concepts, they directly have physi-
cal referents in the real world (i.e., spectrum of light), so the physiological basis is
likely common for all humans. In this case, language is often found to have an influ-
ence on color perception. On the other hand, abstract concepts (e.g., time) do not
have physical referents in the real world, such that we do not directly interact with
them. How do people represent abstract concepts? Does language also have an
impact on the processing of abstract concepts?
Conceptual Metaphor Theory
Unlike concrete concepts, abstract concepts do not have physical referents in the
real world, such as time, valence, number, pitch, and power. How abstract concepts
are represented in human mind is a critical question in cognitive psychology. The
Conceptual Metaphor Theory postulates that abstract concepts can be represented
in terms of concrete concepts via conceptual metaphors (Lakoff & Johnson, 1980,
1999). Specifically, they can be structured and grounded on sensorimotor informa-
tion originating from the representation of concrete concepts. Take spatial-valence
conceptual metaphor (good is up/bad is down) as an example. Valence (positive or
negative) may not be only described linguistically in spatial terms (e.g., I am feeling
down to express sad mood), but also understood and represented by such spatial
information (see, e.g., Crawford, Margolies, Drake, & Murphy, 2006; Gozli, Chow,
Chasteen, & Pratt, 2013; Horstmann, 2010; Meier & Robinson, 2004; Ostinelli,
Luna, & Ringberg, 2013; Weger, Meier, Robinson, & Inhoff, 2007, for evidence).
For space–time conceptual metaphor, time can be described in terms of spatial
information (e.g., the meeting was moved forward/backward 2 h; the concert lasted
for a long/short time). The space–time relationship also helps structuring and repre-
senting time through the metaphoric association between space and time (see, e.g.,
Boroditsky, 2000; Boroditsky & Gaby, 2010; Brown, 2012; Casasanto & Boroditsky,
2008; Ulrich & Maienborn, 2010; Weger & Pratt, 2008, for evidence).
Conceptual Metaphors and Embodiment
Embodied cognition theories have different forms (Barsalou, 2008), such as

grounded cognition and situated cognition (see Wilson, 2002, for a theoretical dis-
cussion). Here we focus on the general notion of embodied cognition—the human
mind is embodied and grounded in bodily states and in the brain’s modality-specific
systems (Niedenthal, Barsalou, Winkielman, Krauth-Gruber, & Ric, 2005). This
account emphasizes the roles of bodily states, simulation, and situated environment
in cognition (Barsalou, 1999, 2003, 2008; Hauk, Johnsrude, & Pulvermuller, 2004;
Smith, 2005; Smith & Semin, 2004).
Concrete concepts that inherently have physical referents in the real world can
be understood and represented in the sensorimotor system via direct physical expe-
rience and interaction with them. For abstract concepts, however, the Conceptual
Metaphor Theory postulates that they can be grounded in embodied and situated
knowledge via metaphoric association (see also Gibbs, 1994). With metaphoric
association, bodily experience (e.g., sitting erect) and perceptual characteristics

with concrete concepts (e.g., vertical spatial information) make it possible that
mental representations of abstract concepts (e.g., positive valence) are grounded in
bodily and sensorimotor experiences (i.e., good-up/bad-down conceptual meta-
phor). Similarly, spatial information can be used to understand and represent time
perception (i.e., space–time conceptual metaphor; Boroditsky, 2000; Gentner,
Imai, & Boroditsky, 2002) and container information can be used to represent and
process category information (Boot & Pecher, 2010). In brief, abstract concepts
can be represented and grounded in embodied experience and sensorimotor system
through conceptual metaphors (Casasanto, 2009; Gibbs, 1994; Gibbs, Costa Lima,
& Francozo, 2004).
Linguistic Metaphors and Conceptual Metaphors
Linguistic metaphors are specific linguistic expressions, surface manifestations of

conceptual metaphors. Conceptual metaphors are systematic mappings across two
conceptual domains, from source domains to target domains, such as good is up,
categories are containers, love is a journey, and so on (Lakoff, 1993; Lakoff &
Johnson, 1980). Similarly, Kövecses (2005) posits that metaphors exist in language
because they exist in thought, and thus linguistic metaphors are expressions of
metaphoric concepts in the conceptual system. In other words, metaphoric expres-
sions in language make conceptual metaphor manifest, and we can use these lin-
guistic metaphors to test the inner mechanism that links concrete and abstract
concepts. However, Casasanto (2009) argued that some conceptual metaphors
could not be predicted based on linguistic metaphors. Casasanto adopted a concep-
tual judgment task (i.e., judging the similarity in meaning of pairs of abstract
nouns) and a perceptual judgment task (i.e., judging the similarity in visual appear-
ance of pairs of unfamiliar faces) to test the metaphor similarity is proximity. This
metaphor is often linguistically expressed in daily language, such as The two
shades of blue are not identical, but they are close and The opposing candidates’
stances on this issue couldn’t be farther apart. Results showed that in conceptual
judgments, closer stimuli were judged more similar with each other, which was
consistent with and could be predicted by the similarity is proximity linguistic
metaphor. However, in perceptual judgments, closer stimuli were judged less simi-
lar, which was contrary to the prediction of the similarity is proximity linguistic
metaphor. Therefore, Casasanto concluded that linguistic metaphors might only be
a subset of conceptual metaphors that appeared to structure the mental representa-
tions, and suggested that it was important to use an experimental method, in con-
junction with the linguistic corpus analyses, to test whether a linguistic metaphor
might reflect an underlying conceptual metaphor.
Even though much evidence on conceptual metaphors comes from linguistic
studies, increasing evidence from non-linguistic tasks suggested that conceptual
metaphors not only reflect linguistic convention, but also represent how people
think. For example, Boroditsky and Ramscar (2002) tested the relationship between
space and time, showing that thinking about time is dependent on the representa-
tions of a more concrete concept like space with direct sensorimotor experience. For
instance, in one study, after an object-moving spatial prime (e.g., imagining how to
maneuver the chair to the X) or an ego-moving spatial prime (e.g., imagining how
to draw an arrow indicating the path of motion), participants were asked to answer
an ambiguous temporal question, Next Wednesday’s meeting has been moved for-
ward two days. What day is the meeting now that it has been rescheduled? Results
showed that following by object-moving spatial prime participants were more likely
to think of time as coming toward them (an object/time-moving-consistent manner)
and thus answer Monday; whereas participants receiving the ego-moving spatial
prime were more likely to think of themselves as moving through time (an ego-
moving-consistent manner) and answer Friday. Further, Boroditsky and Ramscar
also investigated the influence of spatial experience (e.g., standing in a lunch line)
and spatial thinking (e.g., thinking about their journey on a train) on people’s think-
ing about time. Results showed that people changed how they think about time
when engaging in such kind of spatial experience or spatial thinking. Specifically,
participants who waited further along in line (i.e., experiencing more forward spa-
tial motion) were more likely to think of themselves as moving through time
(answering Friday to the ambiguous question). Participants, who just got on or
would get off the train within 5 min, were more likely to take the ego-moving per-
spective on time than those who were in the middle of their journey. Taken together,
people’s thinking and understanding of the abstract concept of time is largely depen-
dent on more concrete concept of space, even in non-linguistic tasks.
Based on the Conceptual Metaphor Theory, thinking about abstract concepts is
grounded on representations of more concrete concepts with direct physical experi-
ence through metaphoric association. Several theorists (Boroditsky & Ramscar,
2002; Piaget & Inhelder, 1972) assume that such kind of cross-domain mappings
originated from the co-occurrences in sensorimotor experience (i.e., co-activation).
For example, upright posture is often correlated with positive status whereas bent
posture with negative status. Space and time are closely linked such that the longer
the movement is (distance in space), the longer the time duration is (Casasanto et al.,
2004). Importantly, these co-occurrences in sensorimotor experiences can be built on
and reflected by linguistic expressions. That is, people use linguistic metaphors to
express the inner mappings from more experience-based concrete domains to more
abstract domains, such as I am feeling up/down to express the emotional status and
spatial terms (such as front/back and longer/shorter) to express the characteristics of
time. The space–time conceptual metaphor has been supported by several studies
(e.g., Boroditsky, 2001; Casasanto, 2008; Dolscheid, Shayan, Majid, & Casasanto,
2013), although inconsistent results have also been reported (e.g., Chen, 2007;
January & Kako, 2007; Tse & Altarriba, 2008). In the following section, we provide
a literature review on this issue to explore whether language plays a role in under-
standing and representing abstract concepts in the conceptual metaphor literatures.
Linguistic Relativity on Conceptual Metaphors
As predicted by the linguistic relativity hypothesis, people who speak different lin-
guistic metaphors think of target concepts differently. For example, Dolscheid et al.
(2013) adopted a pair of psychophysical tasks (i.e., pitch reproduction tasks) with
non-linguistic stimuli and responses to test the mental representation of musical
pitch in native Dutch speakers and native Farsi speakers. Native Dutch speakers
often talked about pitches as high (hoog) or low (laag) (i.e., height-pitch meta-
phors), whereas native Farsi speakers often called pitches as thin (nāzok) and thick
(koloft) (i.e., thickness-pitch metaphors). In one task (i.e., height interference), par-
ticipants first listened to the tones with different pitches while seeing lines having
different heights. They then needed to reproduce the pitch by singing it. Similar
procedure was used in the other task (i.e., thickness interference) except that partici-
pants saw lines having different thickness when listening to the tones. Even though
spatial information (height or thickness) was not relevant to pitch reproduction,
native Dutch speakers incorporated height information but ignored the thickness
information, whereas native Farsi speakers incorporated thickness information but
ignored the height information when they performed in the pitch reproduction task.
These findings suggested that people using different linguistic metaphors represent
abstract concepts differently, even in non-linguistic tasks. Moreover, after native
Dutch speakers were trained to describe pitches as thin or thick as Farsi speakers do,
they showed a similar pattern of thickness interference to that demonstrated by
native Farsi speakers. The training effect that using linguistic metaphors could
change the mental representation of abstract concepts further showed the causal
relationship between language and musical pitch estimation. In addition, Dolscheid,
Hunnius, Casasanto, and Majid (2014) investigated the space–pitch conceptual met-
aphor among prelinguistic 4-month-old infants using a preferential-looking para-
digm in which congruent trials (e.g., high pitch with taller line) were preferred and
looked at longer than incongruent trials (e.g., high pitch with shorter line) (Walker
et al., 2010). Results revealed that infants showed sensitivity to both height-pitch
and thickness-pitch metaphoric associations, looking longer at metaphorically con-
gruent stimuli than metaphorically incongruent stimuli, suggesting that language
may build on pre-existing mappings (i.e., metaphoric associations), which can be
shaped through strengthening one pre-existing mapping and weakening the other
one (see also Shayan, Ozturk, Bowerman, & Majid, 2014).
Space–time conceptual metaphor has also been extensively investigated in lin-
guistic relativity research on whether language influences or shapes the mental rep-
resentation of time (e.g., Boroditsky, 2000; Boroditsky, Fuhrman, & McCormick,
2011; Casasanto, 2008; Chan & Bergen, 2005). For example, in English, horizontal
spatial terms are more often used to talk about time (e.g., moving meetings forward
and pushing deadline backward; Boroditsky, 2001). In Chinese, both horizontal and
vertical spatial terms are used to describe time (Scott, 1989). For instance, horizon-
tal terms, qian [前] (front) and hou [後] (back), are used to express earlier and later
time points, e.g., huiyi ti qian [會議提前] (moving the meeting forward) and bi ye
zhihou [畢業之後] (after graduation). Also, vertical spatial terms, shang [上] (up)
and xia [下] (down), can be used in the same way to describe time, e.g., shanggeyue
[上個月] (last month) and xi yinian [下一年] (next year). Given the different space–
time linguistic metaphors, native Chinese speakers are more likely to represent time
both horizontally (left-to-right) and vertically (top-to-bottom), whereas native
English speakers are more likely to think about time horizontally (left-to-right). In
the following section, we focused on the space–time conceptual metaphor and pro-
vided a comprehensive review of the role of language in the mental representation
of time (i.e., linguistic relativity in space–time conceptual metaphor) and the role of
reading/writing direction in time representation.
A Case Study of Linguistic Relativity: Space–Time Metaphors
Space–Time Metaphors Across Languages
Across languages in the world, time can be described in various ways, such as time-
lines in one-dimensional space (direction and distance) or containers in three-
dimensional space (amount). For instance, English speakers predominantly use
distance metaphors to express time duration (e.g., a long time), while Greek speakers
use amount metaphors much more frequently (i.e., much time) (Casasanto, 2008).
For the timelines, the main characteristics are direction and distance. The distance of
timeline refers to the duration of time, while the direction of timeline is about the
relations between two time points (i.e., one is earlier or later than the other). To
describe the relations, people have to choose a frame of reference, which is a coordi-
nated system to describe relations between the movements of objects or events
(Bender, Beller, & Bennardo, 2010). People speaking different languages use differ-
ent frames of reference and talk about time differently. Most cultural groups have
been found to spatialize time along the egocentric coordinates, including front/back
(e.g., English, Chinese, and Spanish), left/right (e.g., English, Chinese, and Spanish),
and up/down (e.g., Chinese) axes. However, not all cultures rely to the same extent
on egocentric coordinates. Some groups prefer allocentric coordinates, such as car-
dinal direction (east, west, north, and south; e.g., west/east for Thaayorre speakers)
and environment-based direction (e.g., uphill/downhill for Mexico speakers and river
directions for Mian speakers) (Núñez, Cooperrider, Doan, & Wassmann, 2012).
Time can also be described as stationary from ego-moving perspective, in which
ego or the observer move along the timeline to a time point, or as moving from the
time-moving perspective, in which time moves in a stationary environment. In the
latter case, the timeline is just like a river or conveyor belt on which an event is mov-
ing. In the ambiguous target question, next Wednesday’s meeting has been moved
forward two days. What day is the meeting now that it has been rescheduled?, the
answer differs depending on time-moving perspective (i.e., from Wednesday to
Monday) or ego-moving perspective (i.e., from Wednesday to Friday) (Boroditsky,
2000; Gentner et al., 2002). The way that time is described as moving or stationary
differs across languages. Even though both time- and ego-moving metaphors are
available in English and Chinese languages, some researchers found that time-
moving metaphors are used more frequently in Chinese (Dong, 2004; Lai &
Boroditsky, 2013). Lai and Boroditsky showed that Chinese–English bilinguals
were less likely to take an ego-moving perspective when tested in English than
English monolinguals, but more likely to take an ego-moving perspective when
tested in Chinese than Chinese monolinguals. These suggested that bilinguals’ sec-
ond language (L2) could influence the time conceptualization in their first language
(L1) and their L1 could also influence the time conceptualization in their L2 (see
also Lai, Rodriguez, & Narasimhan, 2014).
Overall, while mapping time onto space may be universal, specific ways of
spatializing time vary across languages and cultures. People with different lan-
guages use different space–time metaphors to describe time. The cross-linguistic
differences provide us with a test bed to investigate whether or not people who
express time differently may also think about time differently (i.e., the linguistic
relativity hypothesis).
oes Language Shape the Representation of Time: Empirical

D
Evidence
One-Dimensional Timeline: Egocentric-Based
As mentioned above, both horizontal and vertical spatial terms are used to describe
time in Chinese, but only horizontal spatial terms are frequently used to talk about
time in English. Boroditsky (2001) employed a spatial priming paradigm to exam-
ine whether native Chinese speakers and native English speakers who speak differ-
ent spatial metaphors might represent time differently. In this paradigm, a spatial
prime consisting of two objects was aligned horizontally or vertically with a state-
ment and participants were asked to judge whether the statement correctly depicted
the spatial relationship of the two objects (e.g., horizontal spatial prime: the black
worm is ahead of the white worm; vertical spatial prime: the black ball is above the
white ball). Following the spatial prime, a target statement appeared and partici-
pants were instructed to judge whether it correctly described a temporal relationship
(e.g., March comes before April). Results showed that native English speakers’ tem-
poral judgments were faster after horizontal primes than after vertical primes;
whereas native Chinese speakers showed the opposite pattern (i.e., faster after verti-
cal primes than after horizontal primes). This suggested that the mental timeline
was more likely to be mapped onto a horizontal axis for English speakers and onto
a vertical axis for Chinese speakers. In other words, different space–time metaphors
spoken by native English and native Chinese speakers led to different mental repre-
sentation of time. It should be noted that in this study, all Chinese speakers had
Mandarin Chinese as their L1, which was their only language until 6 years of age,
and English as their L2 with a mean onset age of acquisition of 12.8. A vertical bias
in thinking about time was observed for native Chinese speakers, even though they
had English as their L2 and completed the task in English, showing the strong
impact of native language in shaping speakers’ time representation. In addition,
Boroditsky observed a positive correlation between the extent that Chinese–English
bilinguals think about time vertically and the ages that they began to learn English.
The later they began to learn English, the more likely they thought about time verti-
cally, showing that English, despite not being their L1, also had an impact on the
bilinguals’ mental representation of time. More importantly, Boroditsky tested the
causal relationship between spatial language and time representation further by
training native English speakers (i.e., English monolinguals) to use vertical terms to
talk about time. After the training, results of native English speakers looked more
like those of Chinese speakers, compared to the results of untrained English speak-
ers, supporting the causal role of spatial language in temporal thinking. Taken
together, different language use of spatial metaphors could shape the way of think-
ing and representing time.
However, several subsequent studies reported inconsistent findings (Chen, 2007;
January & Kako, 2007; Tse & Altarriba, 2008). Boroditsky’s (2001) findings were
based on the assumption that native English speakers primarily use horizontal spatial
metaphor, whereas native Chinese speaker also use the vertical spatial metaphor,
which is much more frequently used than the horizontal spatial metaphor. However,
based on the linguistic analyses, Chen (2007) reported that native Chinese speakers
actually use horizontal metaphor more frequently than vertical metaphor, contradict-
ing Boroditsky’s major assumption. Further, by using Boroditsky’s paradigm, Chen
failed to replicate the original results and found that Chinese and English speakers
did not think about time differently (see also January & Kako, 2007; Tse & Altarriba,
2008), which was not consistent with the patterns in their linguistic experience.
These findings were not in line with the linguistic relativity hypothesis.
The controversy attracted further studies to test the representation of time across
cultures and languages, and several paradigms were developed accordingly. Chan
and Bergen (2005) employed an arrangement task, in which participants saw several
pictures depicting the developmental stages of a natural entity (e.g., egg/small
chicken/big chicken) and they were asked to spatially arrange the temporal sequence
shown in these pictures from the earliest to the latest stage. Chinese–English bilin-
guals arranged the pictures in both horizontal arrays (left-to-right) and vertical arrays
(top-to-bottom), whereas English monolinguals only arranged in left-to-right pat-
tern. Likewise, when the task was to locate events in three-dimensional space by
pointing to locations around the body, Chinese–English bilinguals were more likely
to arrange time vertically than English monolinguals (Fuhrman et al., 2011). Further,
Fuhrman et al. also revealed that Chinese–English bilinguals with more proficient
Chinese were more likely to arrange temporal sequence vertically. They were more
likely to arrange temporal sequence vertically when tested in Chinese than in English.
Boroditsky et al. (2011) adopted a non-linguistic implicit association task (e.g.,
a spatial-numerical association of response codes [SNARC] paradigm) to detect the
differences in time representation between native English and Chinese speakers (see
also Miles, Tan, Noble, Lumsden, & Macrae, 2011). These English speakers did not
have any exposure to Chinese and the native Chinese speakers had English as their
L2 (i.e., English monolinguals and Chinese–English bilinguals). In this task, par-
ticipants saw two pictures of an event that appeared one by one; then they needed to
judge whether the second picture occurred earlier or later than the first one (e.g., the
whole banana in the first picture and the half-peeled banana in the second one). The
critical manipulation was on the key assignment. In the horizontal assignment, half
of the participants were instructed to press a left key for earlier responses and a right
key for later responses (canonical condition), and the other half were given a
reversed key assignment (non-canonical condition). In the vertical assignment, half
of them were instructed to press a top key for earlier responses and a bottom key for
later responses (canonical condition), and the other half were given a reversed key
assignment (non-canonical condition). Response times were faster in the canonical
condition than the non-canonical condition, which was called the spatial-temporal
association of response codes (STARC) effect. Boroditsky et al. demonstrated a
Chinese–English difference in temporal thinking. Chinese–English bilinguals
showed a larger vertical STARC effect than English monolinguals did, which was
consistent with space–time metaphor usage in language that vertical terms were
used more often in Chinese than in English. Fuhrman et al. (2011, Experiment 1)
further used the STARC task with a three-dimensional space, including the sagittal
(front/back), transverse (left/right), and vertical (up/down) axes. Results showed
that both English monolinguals and Chinese–English bilinguals showed a horizon-
tal (left-to-right) mental timeline, but only Chinese–English bilinguals showed a
vertical (top-to-bottom) mental timeline. However, inconsistent findings were also
observed in the STARC task. Chen and O’Seaghdha (2013) showed a vertical
STARC effect for Chinese speakers from Taiwan, but not for the Chinese speakers
from mainland China. They claimed that this might be due to direction of text print-
ing, which was both vertical and horizontal in Taiwan but only horizontal in main-
land China. In other words, the findings that participants represent time differently
could be confounded with other factors, such as text reading direction, so it is not
clear whether the differences in time representations could be attributed to the dif-
ferences in language usage (i.e., the linguistic relativity hypothesis) and/or other
factors such as reading/writing directions.
One-Dimensional Timeline: Allocentric-Based
Besides the egocentric spatial referents, some other languages involve allocentric
coordinates to describe space, such as absolute cardinal directions and environmental-
based directions (e.g., uphill/downhill and river directions). These ways to express
space are also used to talk about time in these languages.
KuukThaayorre is the language spoken by the residents of Pormpuraaw, a remote
Australian aboriginal community. It relies on the absolute cardinal direction terms
to describe spatial relations, and time is found to be represented along the absolute
east-to-west axis, with earlier times mapped to eastward and later times to westward
direction (Boroditsky & Gaby, 2010; Gaby, 2012). Boroditsky and Gaby (2010)
adopted the temporal arrangement task described above and a dot-drawing task to
elicit spatial representation of time for American and Pormpuraawan participants. In
the temporal arrangement task, participants were asked to arrange pictures to depict
a temporal order from earliest to latest. In the dot-drawing task, participants were
first told that the dot appeared in front of them represented today, and then asked to
draw a dot representing tomorrow and yesterday. In both tasks, participants were
rotated either 90°or 180°, such that participants faced to different cardinal direction
when they did the arrangement or drawing task. Both tasks showed that American
participants arranged time from left to right, regardless of their cardinal directions
that they faced. In contrast, Pormpuraawan participants took into account their fac-
ing directions in the temporal arrangement: they were more likely to arrange time
from left to right when facing south and from right to left when facing north, and
they were more likely to arrange time as coming towards them when facing east and
moving away when facing west. Gaby (2012) further examined the role of language
in shaping spatial representations of time by testing two groups of Pormpuraawans
who differed in their language fluency in KuukThaayorre, yet shared the same
social/environmental context and reading/writing direction. Specifically,
Pormpuraawans who spoke KuukThaayorre tended to arrange time from east to
west, whereas Pormpuraawans who could speak English only arranged time from
left to right.
Yupno speakers, an indigenous group from the mountains of Papua New Guinea,
often use allocentric topographic (uphill/downhill) terms to describe spatial rela-
tions. Núñez et al. (2012) found that Yupno speakers spontaneously construed time
spatially in terms of allocentric topography. Past is construed as downhill and pres-
ent as co-located with the speaker and future as uphill. Further, they found the time
was represented not as the linear timelines but as a particular geometry that reflected
the local terrain. These findings reflected the effects of linguistic, cultural, and envi-
ronmental factors on people’s time representation, which were in line with the
embodiment view. Similarly, speakers of Mayan language Tzeltal mainly relied on
allocentric frames of reference, the uphill/downhill slope of the land, to describe
spatial relations (Levinson, 2003), though they also used an egocentric frame of
reference (front/back, but not left/right). However, by using the temporal arrange-
ment task and time-pointing task, Brown (2012) found that this kind of allocentric
frame of reference of uphill/downhill was not the only or the dominant metaphors
to process or think about time. With a variety of directional bases, participants dis-
played inconsistency in time representation across trials, such as timelines pointing
downhill, from west to east, or from right to left, suggesting that the absolute frame
of reference in expressing time was not necessarily used to represent time.
On the other hand, in the language of Mian of Papua New Guinea, there is no
word for left or right meaning or term for cardinal direction. Instead, this language
relies on the topographic environment-based absolute reference for spatial relations.
Specifically, Mian employs an absolute frame of reference, which is associated
with the lay of the Hak and SEK rivers, running from east to west, roughly parallel
near the Mianmin village, so the terms (i.e., met upriver, tab downriver, and tām
sideways of the river) are used to describe horizontal dimension. However, the ways
to express spatial information were not used to describe time in this language.
Fedden and Boroditsky (2012) tested how Mianmin represent time in a temporal
arrangement task and revealed a variety of representational strategies for organizing
time, relative pattern: left-to-right axis (4 participants) and back-front axis (2 par-
ticipants) and absolute pattern: east-west axis (3 participants), suggesting that Mian
speakers do not have a standardized spatialization for time. It should be noted that
English becomes increasingly important for Mianmin and their 3–12 years old chil-
dren are now taught almost entirely in English. Fedden and Boroditsky found a
significant predictor (the number of years of formal education) for temporal arrange-
ment pattern: with more exposure to formal education, the relative pattern (left-to-
right) was reinforced and absolute pattern (east-to-west) was weakened.
One-/Three-Dimensional Space–Time Metaphor
Besides the one-dimensional space, language could also describe time as three-
dimensional, such as saving time and a lot of time. For instance, native Greek speak-
ers more often talk about time in a three-dimensional way such as amount metaphors
(e.g., much time) than in a one-dimensional way such as distance metaphors (e.g., a
long time). In contrast, native English speakers use distance metaphors more often
than amount metaphors. Casasanto et al. (2004) tested whether the different meta-
phors used by native Greek vs. native English speakers in describing time could
trigger their thinking about time differently. To avoid potential linguistic effects
brought by the task that required participant to process in language, Casasanto et al.
adopted a non-linguistic task (duration reproduction) to test whether the difference
in space–time metaphors in two languages influenced native Greek and native
English speakers’ mental representations of time. In this task, participants were
asked to estimate the duration of events along with distance-related or amount-
related distracting information. In the distance interference condition, participants
were presented a growing line across the screen and then asked to reproduce the
duration of the growing event. In the amount interference condition, participants
were instructed to watch a schematic drawing of a container being filled up with
liquid and then asked to reproduce the duration of the filling event. Casasanto et al.
showed that the duration reproduction task was more affected by distance interfer-
ence than by amount interference for native English speakers, whereas the duration
reproduction task was more affected by amount interference than by distance inter-
ference for native Greek speakers. Moreover, Casasanto et al. also tested the causal
relationship. Native English speakers were randomly assigned into distance and
amount metaphor training groups. Those who were trained by using the amount
metaphor to describe temporal duration showed a larger influence by amount inter-
ference, like native Greek speakers, supporting the role of linguistic experience in
shaping non-linguistic representations of time (see also Casasanto, 2008).
Ego-Moving/Time-Moving Perspectives
Based on the linguistic analyses in Lai and Boroditsky (2013), Chinese monolin-
guals are more likely to take the time-moving perspective than English monolin-
guals. Moreover, Chinese–English bilinguals took more time-moving perspective
when tested in English than English monolinguals and took more ego-moving per-
spective when tested in Chinese than Chinese monolinguals. This showed that the
usage of conceptual metaphor in one language could influence temporal judgments
in the ambiguous question tasks. Furthermore, Lai and Boroditsky employed the
three-dimensional pointing paradigm to test the role of metaphor use in modulating
how people represent time. They found that Chinese–English bilinguals were more
likely to represent time on the front-back array when understanding front-back met-
aphors and more likely to represent time on the up-down array when understanding
up-down metaphors.
From the evidence discussed above, even though some inconsistent findings
were observed in various paradigms and across a variety of languages, the role of
language in time representation is quite clear, given that people who speak about
time differently also represent time differently and that a causal relationship between
language and time representation was obtained in several training studies. However,
there are still many questions unresolved. For example, does language play a deter-
mining role or just modulate the mental representation of time? What is the origin
of such kinds of space–time conceptual metaphors? What is the role of embodiment
in processing conceptual metaphors, and how do the effects of language and embodi-
ment interact? In the next sections, we discuss an embodied factor of reading/writ-
ing systems across cultures to provide some potential answers for some of these
questions and then propose some potential directions for future research.
xperience of Reading/Writing Directions and Space–Time

E
Conceptual Metaphor
Even though the aforementioned evidence supported the influence of language on

the time representation, only few of them attempted to exclude the possible con-
founding factors, like the experience of reading/writing direction (Bergen & Lau,
2012; Chan & Bergen, 2005; Chen, Friederich, & Shu, 2013; Fuhrman & Boroditsky,
2007, 2010; Ouellet, Santiago, Israeli, & Gabay, 2010). Casasanto (2012) argues
that language is only a part, even if important, of the way that experience shapes
our way we think about time, which suggests that the role of language can be attrib-
uted to what the experience of language is grounded on, not the superficial signals.
However, the two are not always consistent. For example, studies have shown that
time is represented on the left-right axis, although there is no corresponding space–
time metaphor in that language. Hence, it is important for the researchers to distin-
guish the role of language itself from the experience of language use (e.g., reading/
writing direction) in time representation.
Fuhrman and Boroditsky (2007) compared two groups of participants who had
opposite reading/writing directions. English speakers read and write from left to
right, whereas Hebrew speakers have the right-to-left reading/writing direction. By
doing the STARC task, English speakers were faster to make earlier responses on
the left keys than on the right keys, whereas the opposite pattern was observed for
Hebrew speakers. In other words, the orientation of timeline was represented in
accordance with their reading/writing direction. Ouellet et al. (2010) showed the
similar role of reading/writing directions in the mental timeline. Spanish (who read
and write from left to right, same as English speakers) and Hebrew speakers were
asked to do an auditorily STARC task, in which they judged the auditorily presented
words referring to the past or future with either left or right hand. Results showed
that Spanish participants were faster responding to past words with the left hand and
to future words with the right hand, whereas Hebrew participants showed the oppo-
site pattern. Fuhrman and Boroditsky (2010) also replicated the findings of left-to-
right mental timeline for English speakers and right-to-left mental timeline for
Hebrew speakers in both temporal arrangement task and non-linguistic STARC
tasks, consistent with their corresponding reading/writing directions.
Further, Chen et al. (2013) dissociated this experience of reading/writing direc-
tion from language usage by recruiting participants who were from Taiwan or
mainland China and shared the same language, but differed in their experience of
reading/writing directions. Specifically, mainland China adopts the national pol-
icy of printing all texts horizontally, but this is not the case in Taiwan (i.e., both
horizontal and vertical reading/writing directions are used in Taiwan). In their
study, participants were asked to do a reading task involving either horizontally or
vertically arranged texts (i.e., a priming task), and then a STARC task. Results
showed that Taiwan participants showed a larger vertical STARC effect than the
horizontal one, however such a vertical bias was not observed for participants
from mainland China. This could be attributed to Taiwan participants’ signifi-
cantly more frequent experience of encountering vertical texts than that of main-
land China participants. Moreover, the vertical bias of Taiwan participants was
modulated by their immediate reading experience such that it disappeared when
they read horizontally arranged texts prior to the STARC task. The modulation
effect of the immediate reading experience was also observed for the mainland
China participants, but in the opposite direction. The horizontally arranged texts
did not lead to any vertical bias but the vertically arranged texts brought about a
significant horizontal bias in mainland China participants. Given that both Taiwan
and mainland China participants share the linguistic expressions of space–time
metaphors, the significant influence of reading/writing direction on their ways to
represent time suggested that the experience of reading/writing direction did have
an independent role in time representations.
Chen and O’Seaghdha (2013) added one more group of participants from the
United States who shared the same experience of reading/writing direction with
participants from mainland China. Such paired comparisons provided the dissoci-
ated role of language vs. experience of reading/writing directions in time represen-
tation. That is, if participants from mainland China showed similar results with
Taiwan participants but a reversed pattern of results with US participants, language

might play a more determining role compared to the experience of reading/writing
direction. Otherwise, the experience of reading/writing direction might play a more
determining role than language (see also Chan & Bergen, 2005). Results showed
that participants from mainland China and the United States showed a STARC
effect when the key responses were horizontally assigned, but not when vertically
assigned; but the opposite pattern was observed for Taiwan participants, suggesting
the reading/writing direction played a more significant role than language in shap-
ing participants’ time representation. Similarly, Bergen and Lau (2012) extended
these studies in a temporal arrangement task (i.e., spatially placing pictures in the
temporal order). In this study, English participants were all English monolinguals
and participants from mainland China and Taiwan were all Chinese–English bilin-
guals. Results showed that English participants only adopted the left-to-right
arrangement to place the pictures. The bilinguals from mainland China showed a
strong tendency to use the left-to-right arrangement and a few of them used the top-
to-bottom arrangement. Inconsistent with English monolinguals and Chinese–
English bilinguals from mainland China who never used the right-to-left
arrangement, Chinese–English bilinguals from Taiwan also showed a small portion
of right-to-left arrangement other than the left-to-right and top-to-bottom arrange-
ments. These findings again confirmed the independent role of reading/writing
direction in time representation.
Casasanto and Bottini (2014) showed the causal role of reading/writing direction
on the representation of mental timeline. Dutch speakers were randomly assigned into
four groups, with instruction and stimuli being written in standard, mirror-reversed, or
rotated orthography (rotated 90° clockwise or counterclockwise). Then they did the
STARC task, in which they judged whether the stimuli referred to an earlier (e.g., een
jaar daarvoor [a year before]) or a later (een dag daarna [a day after]) time by press-
ing a left-right assigned key in the standard and mirror-reversed orthography condi-
tions, or an up-down assigned key in the two rotated orthography conditions. Results
showed that when participants judged temporal words written in standard orthogra-
phy, their response times were consistent with a rightward-direction mental timeline
(i.e., left-past and right-future). However, the pattern was reversed when the instruc-
tion and stimuli were written in mirror-reversed orthography (i.e., left-future and
right-past). Moreover, when standard orthography was rotated 90° clockwise (down-
ward) or counterclockwise (upward), participants’ mental timelines were rotated
accordingly, too. These findings also suggested that the online experience of reading/
writing plays a significant role in the representation of time.
Conclusion
In the previous sections, we introduced linguistic relativity, embodiment, and con-

ceptual metaphor with the aim to learn more about the role of language and embodi-
ment in conceptual metaphors and then provided a comprehensive review of the
space–time metaphors as a litmus test of the linguistic relativity hypothesis. In the

following section, we summarize, again using the space–time metaphor, the role of
language and embodied experience in time representation, and then put forward
some potential future directions.
Role of Language in Time Representation
Time can be expressed in various ways, such as stationary as ego/object- moving

perspective and moving as time-moving perspective, distance (e.g., long time) or
amount (e.g., much time), timelines with difference directions across cultures (e.g.,
horizontal for English speakers; vertical and horizontal for Chinese speakers), and
absolute cardinal directions and environmental-based directions (e.g., uphill/down-
hill, or river direction). Empirical evidence so far has supported the role of language
in shaping time representation in both linguistic and non-linguistic tasks (e.g.,
Boroditsky, 2001; Boroditsky et al., 2011; Chan & Bergen, 2005; Chen &
O’Seaghdha, 2013; Fuhrman et al., 2011; Miles et al., 2011). However, it is note-
worthy that some controversial findings were also reported (e.g., Chen, 2007;
January & Kako, 2007; Tse & Altarriba, 2008). Further, several training programs
were conducted to examine the causal relationship between the language usage of
conceptual metaphor and time representation. For example, training English speak-
ers to use vertical spatial terms to talk about time (Boroditsky, 2001) did shape the
way of thinking about time by showing a vertical bias of a mental timeline. Similar
findings were observed in Casasanto et al. (2004) and Casasanto (2008) where
English speakers were trained to use distance or amount metaphor to speak time and
which indeed shaped their ways to represent time. Overall, even though some con-
troversial findings were reported regarding the role of language in space–time con-
ceptual metaphor, there has been evidence that language has an impact on the time
representation via space–time conceptual metaphor, providing support for the lin-
guistic relativity hypothesis. Nevertheless, we should be cautious to conclude that
the way people describe time plays a determining role in mental representation of
time because other factors (e.g., reading/writing direction) may also shape the men-
tal representation of time.
Role of Embodied Experience in Time Representation
Reading/writing direction is considered another kind of experience of language use

that exhibits embodied characteristics. As Casasanto (2012) pointed out, language
is not just superficial signals, but is also grounded in sensorimotor experience. From
this point, the experience that language is grounded on might also play a role in time
representation. This was supported by the evidence that the experience of reading/
writing direction shaped the way people used space–time metaphors and, in turn,
their mental representations of time (e.g., Chen et al., 2013; Fuhrman & Boroditsky,
2007, 2010; Ouellet et al., 2010). However, language expression is not always con-
sistent with the experience of language use (e.g., reading/writing direction) in
space–time conceptual metaphors. For example, while a horizontal mental timeline
is consistent with the reading/writing directions (from left to right for English
speakers, from right to left for Hebrew speakers), it is not the case in the linguistic
expression in daily language. In such cases, do linguistic expression and embodied
experience play different roles in time representation and which of them would play
a more determining role when they are in conflict? From the empirical evidence that
participants’ left-right mental timeline was consistent with their left-right reading/
writing directions, even when there was no left-right space–time metaphor in lin-
guistic expressions (e.g., Chen et al., 2013; Fuhrman & Boroditsky, 2007, 2010;
Ouellet et al., 2010), it suggests that embodied experience plays a more determining
role than linguistic expression when they were in conflict. This was also consistent
with the view of embodied cognition, which offers the ground of conceptual meta-
phor to represent abstract concepts, like time, the representation of which is depen-
dent on the spatial information. The relationship between linguistic metaphors and
conceptual metaphors is evident in such a way that the former is the surface linguis-
tic manifestation of the latter (see a similar view in Kövecses, 2005). Metaphors
exist in language due to its existence in thought, which is grounded on sensorimotor
and embodied experiences. Thus, the embodied experience of reading/writing direc-
tions may be activated faster than linguistic expression and may in turn go beyond
the influence of linguistic expressions, when they are mismatched. Following this
perspective, we may explain the effect of other embodied factors on the mental
representation of time, such as the situated environment (e.g., Pormpuraaw and
Yupno speakers, who depend on their local topographic patterns to represent time)
(Boroditsky & Gaby, 2010; Gaby, 2012; Núñez et al., 2012).
rigin of Conceptual Metaphor: Preexisting or Acquired

O
Via Learning
Research has suggested that the space–time metaphor may be acquired through
learning. Specifically, participants could learn a new pattern to represent time after
a short training program. For example, Casasanto and Bottini (2014) found that the
direction of mental timeline could be changed by presenting the instruction and
stimuli in different directions, showing that the reading/writing directions could
shape the directions of the mental timeline. Alternatively, there may be no direction
for mental timeline originally, which can be induced by some relevant factors, such
as the experience of reading/writing direction or linguistic expressions. Cai, Connell,
and Holler (2013) adopted a non-linguistic task to test the space–time metaphor and
the direction of mental timeline. In their task, participants saw a singer’s video in
which a singer moved a hand along with short or long distance from left to right or
from right to left in the movement conditions. In non-movement condition, singer’s

hands kept short or long distance in front of the body. Then participants judged the
duration of the singing. Results showed that participants’ estimation on the duration
of singing was increased by a long distance movement relative to a short one, which
was similar in all three conditions. The leftward and rightward moving elicited simi-
lar distant effect on the duration estimation, suggesting the spatial representation of
time is non-directional in the transverse axis. Casasanto and Bottini found that the
direction of mental timeline could be reversed, rotated clockwise or counterclock-
wise by manipulating the direction of instruction and stimuli. Hence, it is possible
that mental timeline does not have any default direction. Nevertheless, future stud-
ies should verify this hypothesis by testing the space–time relationship in prelin-
guistic infants who have not acquired any specific language or experience of reading/
writing direction (see Dolscheid et al., 2014). Further, Merritt, Casasanto, and
Brannon (2010) found that rhesus monkeys showed a symmetrical relationship
between space and time. Both spatial and temporal manipulation showed similar
magnitude of effects on spatial and temporal judgments respectively. In contrast,
spatial information influences more on temporal judgment than the other way
around for humans. Merritt et al. attributed the asymmetric space–time relationship
among humans but symmetric for monkeys to the availability of language for
humans. Specifically, language plays a role in the formation of metaphoric associa-
tions and encourages humans to use concrete domains to speak abstract concepts
(e.g., Boroditsky, 2000; Gentner & Bowdle, 2008). On the other hand, monkeys do
not have the experience of using linguistic metaphor that could lead to the asym-
metric space–time relationship. Nevertheless, it does not mean that there is no con-
ceptual metaphor due to the cognitive nature of conceptual metaphor. Future studies
should investigate whether monkeys and/or other animals could use spatial refer-
ence frame (e.g., left-right) to understand time in order to shed further light on the
origins of conceptual metaphor.
Linguistic Association or Embodiment
The Conceptual Metaphor Theory holds that abstract concepts are represented and
understood based on sensorimotor experience from concrete concepts via meta-
phoric association. This view is in line with the basis of embodied cognition.
However, there is another possibility that conceptual metaphor processing, such as
the association between abstract and concrete concepts, is built on a linguistic net-
work (see Louwerse & Jeuniaux, 2008, 2010). Louwerse and Jeuniaux proposed
that language could be understood via embodied activation of sensorimotor experi-
ence or via symbolic representation regarding the interdependencies of words (i.e.,
the symbol interdependency hypothesis). In other words, language can be either
embodied or symbolic. Thus, the question is extended to the nature of language or
concepts. If the metaphoric association is embodied in nature, the effect of concep-
tual metaphor should be in accordance with embodied experience. For example,
Zwaan and Yaxley (2003) reported the iconicity effect that participants responded
faster to iconic configurations (e.g., the word attic presented above the word base-
ment than reverse-iconic configurations (e.g., the word basement presented above
attic). In other words, the conceptual processing was facilitated when the orienta-
tion of two words was matched with embodied experience. Otherwise, if the meta-
phoric association is symbolic in nature, the effect of conceptual metaphor should
be in accordance with linguistic factors (e.g., the frequency of word order) (Louwerse
& Jeuniaux, 2008). For example, the frequency of occurrence of the word pairs in
an iconic (e.g., attic-basement) or a reverse-iconic (e.g., basement-attic) order could
account for the variance in response times better than the iconicity did. Louwerse
and Jeuniaux (2010, see also Tse, Kurby, & Du, 2010) found that the conceptual
processing could be both symbolic and embodied, which were modulated by task
demand, with embodiment factors being stronger in iconicity judgments for pic-
tures and linguistic factors being stronger in semantic judgments for words.
However, few studies have tested the role of linguistic association in the processing
of conceptual metaphor (see Huang & Tse, 2015, additional analyses in the General
Discussion for an exception). Future studies should explore the relationship between
linguistic and embodied factors in the processing of conceptual metaphors to further
our understanding for the nature of conceptual metaphor processing.
References
Au, T. K.-F. (1983). Chinese and English counterfactuals: The Sapir-Whorf hypothesis revisited.
Cognition, 15, 155–187.
Barsalou, L. W. (1999). Perceptual symbol systems. Behavioral and Brain Sciences, 22, 577–660.
Barsalou, L. W. (2003). Situated simulation in the human conceptual system. Language and
Cognitive Processes, 18, 513–562.
Barsalou, L. W. (2008). Grounded cognition. Annual Review of Psychology, 59, 617–645.
Bender, A., Beller, S., & Bennardo, G. (2010). Temporal frames of reference: Conceptual analy-
sis and empirical evidence from German, English, mandarin Chinese and Tongan. Journal of
Cognition & Culture, 10, 283–307.
Bergen, B. K., & Lau, T. C. (2012). Writing direction affects how people map space onto time.
Frontiers in Psychology, 3, 109. doi:10.3389/fpsyg.2012.00109
Berlin, B., & Kay, P. (1969). Basic color terms: Their universality and evolution. Berkeley, CA:
University of California Press.
Boot, I., & Pecher, D. (2010). Representation of categories metaphorical use of the container
schema. Experimental Psychology, 58, 162–170.
Boroditsky, L. (2000). Metaphoric structuring: Understanding time through spatial metaphors.
Cognition, 75, 1–28.
Boroditsky, L. (2001). Does language shape thought? Mandarin and English speakers’ conceptions
of time. Cognitive Psychology, 43, 1–22.
Boroditsky, L. (2011). How language shapes thought. Scientific American, 304, 62–65.
Boroditsky, L., Fuhrman, O., & McCormick, K. (2011). Do English and Mandarin speakers think
about time differently? Cognition, 118, 123–129.
Boroditsky, L., & Gaby, A. (2010). Remembrances of times east: Absolute spatial representations
of time in an Australian aboriginal community. Psychological Science, 21, 1635–1639.
Boroditsky, L., & Ramscar, M. (2002). The roles of body and mind in abstract thought.
Psychological Science, 13, 185–189.
Brown, P. (2012). Time and space in Tzeltal: Is the future uphill? Frontiers in Psychology, 3, 212.
doi:10.3389/fpsyg.2012.00212
Cai, Z. G., Connell, L., & Holler, J. (2013). Time does not flow without language: Spatial dis-
tance affects temporal duration regardless of movement or direction. Psychonomic Bulletin &
Review, 20, 973–980.
Casasanto, D. (2008). Who’s afraid of the big bad Whorf? Crosslinguistic differences in temporal
language and thought. Language Learning, 58, 63–79.
Casasanto, D. (2009). When is a linguistic metaphor a conceptual metaphor? In V. Evans &
S. Pourcel (Eds.), New directions in cognitive linguistics (Vol. 24, pp. 127–145).
Casasanto, D. (2012). Whorfian hypothesis in Oxford bibliographies online: Anthropology.
New York: Oxford University Press.
Casasanto, D., & Boroditsky, L. (2008). Time in the mind: Using space to think about time.
Cognition, 106, 579–593.
Casasanto, D., Boroditsky, L., Phillips, W., Greene, J., Goswami, S., Bocanegra-Thiel, S., et al.
(2004). How deep are effects of language on thought? Time estimation in speakers of English,
Indonesian, Greek, and Spanish. In Proceedings of the 26th Annual Conference Cognitive
Science Society.
Casasanto, D., & Bottini, R. (2014). Mirror reading can reverse the flow of time. Journal of
Experimental Psychology: General, 143, 473–479.
Chan, T. T., & Bergen, B. (2005). Writing direction influences spatial cognition. In Proceedings of
the 27th Annual Conference of the Cognitive Science Society.
Chen, J.-Y. (2007). Do Chinese and English speakers think about time differently? Failure of rep-
licating Boroditsky (2001). Cognition, 104, 427–436.
Chen, J.-Y., Friederich, M., & Shu, H. (2013). Whether Chinese speakers think about time more
vertically depends on their immediate and lifetime experience of reading horizontal or vertical
texts: Evidence from contextual priming. In Proceedings of the 35th Annual Conference of the
Cognitive Science Society.
Chen, J.-Y., & O’Seaghdha, P. G. (2013). Do Mandarin and English speakers think about time
differently? Review of existing evidence and some new data. Journal of Chinese Linguistics,
41, 338–358.
Crawford, L. E., Margolies, S. M., Drake, J. T., & Murphy, M. E. (2006). Affect biases mem-
ory of location: Evidence for the spatial representation of affect. Cognition & Emotion, 20,
1153–1169.
Dolscheid, S., Hunnius, S., Casasanto, D., & Majid, A. (2014). Prelinguistic infants are sensitive to
space-pitch associations found across cultures. Psychological Science, 25, 1256–1261.
Dolscheid, S., Shayan, S., Majid, A., & Casasanto, D. (2013). The thickness of musical pitch:
Psychophysical evidence for linguistic relativity. Psychological Science, 24, 613–621.
Dong, W. (2004). The Chinese conceptualization of time revisited. Contemporary Linguistics, 6,
189–190.
Fedden, S., & Boroditsky, L. (2012). Spatialization of time in Mian. Frontiers in Psychology, 3,
485. doi:10.3389/fpsyg.2012.00485
Franklin, A., & Davies, I. R. L. (2004). New evidence for infant colour categories. British Journal
of Developmental Psychology, 22, 349–377.
Franklin, A., Drivonikou, G. V., Bevis, L., Davies, I. R. L., Kay, P., & Regier, T. (2008). Categorical
perception of color is lateralized to the right hemisphere in infants, but to the left hemisphere
in adults. Proceedings of the National Academy of Sciences of the United States of America,
105, 3221–3225.
Fuhrman, O., & Boroditsky, L. (2007). Mental time-lines follow writing direction: Comparing
English and Hebrew speakers. In Proceedings of the 29th Annual Conference of the Cognitive
Science Society.
Fuhrman, O., & Boroditsky, L. (2010). Cross-cultural differences in mental representations of

time: Evidence from an implicit nonlinguistic task. Cognitive Science, 34, 1430–1451.
Fuhrman, O., McCormick, K., Chen, E., Jiang, H., Shu, D., Mao, S., et al. (2011). How linguistic
and cultural forces shape conceptions of time: English and Mandarin time in 3D. Cognitive
Science, 35, 1305–1328.
Gaby, A. (2012). The Thaayorre think of time like they talk of space. Frontiers in Psychology, 3,
300. doi:10.3389/fpsyg.2012.00300
Gentner, D., & Bowdle, B. (2008). Metaphor as structure-mapping. In R. Gibbs (Ed.), The
Cambridge handbook of metaphor and thought (pp. 109–128). New York: Cambridge
University Press.
Gentner, D., Imai, M., & Boroditsky, L. (2002). As time goes by: Evidence for two systems in
processing space time metaphors. Language and Cognitive Processes, 17, 537–565.
Gibbs, R. W. (1994). The poetics of mind: Figurative thought, language, and understanding.
Cambridge: Cambridge University Press.
Gibbs, R. W., Costa Lima, P. L., & Francozo, E. (2004). Metaphor is grounded in embodied experi-
ence. Journal of Pragmatics, 36, 1189–1210.
Gilbert, A. L., Regier, T., Kay, P., & Ivry, R. B. (2006). Whorf hypothesis is supported in the right
visual field but not the left. Proceedings of the National Academy of Sciences of the United
States of America, 103, 489–494.
Gozli, D. G., Chow, A., Chasteen, A. L., & Pratt, J. (2013). Valence and vertical space: Saccade
trajectory deviations reveal metaphorical spatial activation. Visual Cognition, 21, 628–646.
Hardin, C., & Banaji, M. R. (1993). The influence of language on thought. Social Cognition, 11,
277–308.
Hauk, O., Johnsrude, I., & Pulvermuller, F. (2004). Somatotopic representation of action words in
human motor and premotor cortex. Neuron, 41, 301–307.
Heider, E. R. (1972). Universals in color naming and memory. Journal of Experimental Psychology,
93, 10–20.
Heider, E. R., & Olivier, D. C. (1972). The structure of the color space in naming and memory for
two languages. Cognitive Psychology, 3, 337–354.
Heider, K. G. (1970). The Dugum Dani. New York: Aldine.
Horstmann, G. (2010). Tone-affect compatibility with affective stimuli and affective responses.
Quarterly Journal of Experimental Psychology, 63, 2239–2250.
Huang, Y., & Tse, C.-S. (2015). Re-examining the automaticity and directionality of the activa-
tion of the spatial-valence “Good is Up” metaphoric association. PLoS One, 10, e0123371.
doi:10.1371/journal.pone.0123371
Hunt, E., & Agnoli, F. (1991). The Whorfian hypothesis: A cognitive psychology perspective.
Psychological Review, 98, 377–389.
January, D., & Kako, E. (2007). Re-evaluating evidence for linguistic relativity: Reply to
Boroditsky (2001). Cognition, 104, 417–426.
Kay, P., & Kempton, W. (1984). What is the Sapir-Whorf hypothesis? American Anthropologist,
86, 65–79.
Kövecses, Z. (2005). Metaphor in culture: Universality and variation. Cambridge: Cambridge
University Press.
Lai, V. T., & Boroditsky, L. (2013). The immediate and chronic influence of spatio-temporal met-
aphors on the mental representations of time in English, Mandarin, and Mandarin-English
speakers. Frontiers in Psychology, 4, 142. doi:10.3389/fpsyg.2013.00142
Lai, V. T., Rodriguez, G. G., & Narasimhan, B. (2014). Thinking-for-speaking in early and late
bilinguals. Bilingualism: Language and Cognition, 17, 139–152.
Lakoff, G. (1993). The contemporary theory of metaphor. In A. Ortony (Ed.), Metaphor and
thought (Vol. 2, 2nd ed., pp. 202–251). New York: Cambridge University Press.
Lakoff, G., & Johnson, M. (1980). Metaphors we live by (Vol. 111). Chicago: University of
Chicago Press.
Lakoff, G., & Johnson, M. (1999). Philosophy in the flesh: The embodied mind and its challenge
to western thought. Chicago: University of Chicago Press.
Levinson, S. C. (2003). Space in language and cognition: Explorations in cognitive diversity (Vol.
5). Cambridge: Cambridge University Press.
Louwerse, M., & Jeuniaux, P. (2008). Language comprehension is both embodied and symbolic.
In M. de Vega, A. Glenberg, & A. C. Graesser (Eds.), Symbols, embodiment, and meaning
(pp. 309–326). Oxford: Oxford University Press.
Louwerse, M., & Jeuniaux, P. (2010). The linguistic and embodied nature of conceptual process-
ing. Cognition, 114, 96–104.
Meier, B. P., & Robinson, M. D. (2004). Why the sunny side is up: Associations between affect and
vertical position. Psychological Science, 15, 243–247.
Merritt, D. J., Casasanto, D., & Brannon, E. M. (2010). Do monkeys think in metaphors?
Representations of space and time in monkeys and humans. Cognition, 117, 191–202.
Miles, L. K., Tan, L., Noble, G. D., Lumsden, J., & Macrae, C. N. (2011). Can a mind have two
time lines? Exploring space-time mapping in mandarin and English speakers. Psychonomic
Bulletin & Review, 18, 598–604.
Niedenthal, P. M., Barsalou, L. W., Winkielman, P., Krauth-Gruber, S., & Ric, F. (2005).
Embodiment in attitudes, social perception, and emotion. Personality and Social Psychology
Review, 9, 184–211.
Núñez, R., Cooperrider, K., Doan, D., & Wassmann, J. (2012). Contours of time: Topographic
construals of past, present, and future in the Yupno valley of Papua New Guinea. Cognition,
124, 25–35.
Ostinelli, M., Luna, D., & Ringberg, T. (2013). When up brings you down: The effects of imagined
vertical movements on motivation, performance, and consumer behavior. Journal of Consumer
Psychology, 24, 271–283.
Ouellet, M., Santiago, J., Israeli, Z., & Gabay, S. (2010). Is the future the right time? Experimental
Ozgen, E., & Davies, I. R. L. (1998). Turkish color terms: Tests of Berlin and Kay’s theory of color
universals and linguistic relativity. Linguistics, 36, 919–956.
Piaget, J., & Inhelder, B. (1972). The psychology of the child (Vol. 5001). New York: Basic Books.
Roberson, D., & Davidoff, J. (2000). The categorical perception of colors and facial expressions:
The effect of verbal interference. Memory & Cognition, 28, 977–986.
Roberson, D., Davies, I., & Davidoff, J. (2000). Color categories are not universal: Replications
and new evidence from a stone-age culture. Journal of Experimental Psychology: General,
129, 369–398.
Roberson, D., Pak, H., & Hanley, J. R. (2008). Categorical perception of colour in the left and right
visual field is verbally mediated: Evidence from Korean. Cognition, 107, 752–762.
Scott, A. (1989). The vertical dimension and time in Mandarin. Australian Journal of Linguistics,
9, 295–314.
Shayan, S., Ozturk, O., Bowerman, M., & Majid, A. (2014). Spatial metaphor in language can
promote the development of cross-modal mappings in children. Developmental Science, 17,
636–643.
Smith, E. R., & Semin, G. N. R. (2004). Socially situated cognition: Cognition in its social context.
Advances in Experimental Social Psychology, 36, 53–117.
Smith, L. B. (2005). Cognition as a dynamic system: Principles from embodiment. Developmental
Review, 25, 278–298.
Tse, C.-S., & Altarriba, J. (2008). Evidence against linguistic relativity in Chinese and English: A
case study of spatial and temporal metaphors. Journal of Cognition and Culture, 8, 335–357.
Tse, C.-S., Kurby, C. A., & Du, F. (2010). Perceptual simulations and linguistic representations
have differential effects on speeded relatedness judgments and recognition memory. Quarterly
Journal of Experimental Psychology, 63, 928–941.
Ulrich, R., & Maienborn, C. (2010). Left-right coding of past and future in language: The mental
timeline during sentence processing. Cognition, 117, 126–138.
Walker, P., Bremner, J. G., Mason, U., Spring, J., Mattock, K., Slater, A., et al. (2010). Preverbal infants’
sensitivity to synaesthetic cross-modality correspondences. Psychological Science, 21, 21–25.
Weger, U. W., Meier, B. P., Robinson, M. D., & Inhoff, A. W. (2007). Things are sounding up:
Affective influences on auditory tone perception. Psychonomic Bulletin & Review, 14, 517–521.
Weger, U. W., & Pratt, J. (2008). Time flies like an arrow: Space-time compatibility effects suggest
the use of a mental timeline. Psychonomic Bulletin & Review, 15, 426–430.
Wilson, M. (2002). Six views of embodied cognition. Psychonomic Bulletin & Review, 9, 625–636.
Winawer, J., Witthoft, N., Frank, M. C., Wu, L., Wade, A. R., & Boroditsky, L. (2007). Russian
blues reveal effects of language on color discrimination. Proceedings of the National Academy
of Sciences of the United States of America, 104, 7780–7785.
Zwaan, R. A., & Yaxley, R. H. (2003). Spatial iconicity affects semantic relatedness judgments.
Psychonomic Bulletin & Review, 10, 954–958.
Chapter 2
Inner Speech in Bilinguals: The Example
of Calculation Abilities
Alfredo Ardila and Mónica Roselli
Contents
hat Is Inner Speech?
W 27
Inner Speech in Bilinguals 29
Mental Calculation in Bilinguals 30
Calculation Abilities and Language 31
The Dynamic Process of the Inner Language of Mental Calculation 34
Conclusions 35
References 36
What Is Inner Speech?
The idea that there is an inner speech—internal language for ourselves—has a long
history. As a matter of fact, the idea has existed at least since Plato (1961;
Theaetetus189e − 190a and Sophist 263e) that thinking means to use an inner
speech; that is, thinking to a significant extent means to talk to ourselves. This idea
has been expressed by different authors throughout the modern and contemporary
history. This mental language—as it frequently has been called—differs from ordi-
nary language by consisting solely of meanings, i.e., as signified without signifiers
(Wiley, 2006).
A. Ardila (*)
Florida International University, Miami, FL, USA
e-mail: ardilaa@fiu.edu
M. Roselli
Department of Psychology, Florida Atlantic University, Davie, FL, USA
e-mail: mrossell@fau.edu

28 A. Ardila and M. Roselli
Vygotsky (1934b/2012) systematized the concept of inner speech when he

referred to three different types of speech: external speech (or social speech, that
is, the speech used in social interactions), egocentric speech (or private speech,
the speech for ourselves), and inner speech (internalized social speech). Important
to note, there is a process of internalization in which the first is transformed into
the second and finally into the third. As a matter of fact, private speech represents
a kind of halfway between external and inner speech but with very distinctive
properties. So for Vygotsky, egocentric (private) speech is the link between
social (external) speech and organized inner speech. Social speech represents the
overt, external speech addressed to others (words, sentences) with the purpose of
social interaction and communication; whereas inner speech is subvocalized
speech directed and adapted to oneself; private speech—as a midway between
external and inner speech—is vocalized speech addressed and adapted to one-
self. Thus, private speech is neither social communication nor silent thought, but
vocalized thought (Ehrich, 2006; Jones, 2009; Vygotsky (1934b/2012).
In Vygotsky’s conception, speech begins as a social medium and became inter-
nalized as inner speech, that is, verbalized thought. Following Vygotsky, it has been
proposed that inner speech has four distinguishing features: (a) silence, that is, not
overtly produced, (b) syntactical ellipses or short-cuts, (c) semantic embeddedness,
i.e., highly condensed word meanings, and (d) egocentricity, or highly personal
word meanings (Johnson, 1994; Jones, 2009). Vygotsky (1929, 1934a/1968,
1934b/2012) argued that thought (and so-called complex psychological processes)
is associated with some inner speech.
Attempts have been made to find the neurological substrate of inner speech.
McGuire et al. (1996) found that auditory verbal imagery is associated with
increases in the left inferior frontal gyrus, the left premotor cortex, the supple-
mentary motor area, and the left temporal cortex. They suggested that the silent
articulation of sentences involves activity in an area concerned with speech gen-
eration, while imagining speech is associated with additional activity in regions
associated with speech perception. Morin and Michaud (2007) clearly demon-
strate that the left inferior frontal gyrus represents the brain area sustaining inner
speech use. Geva et al. (2011) showed that inner speech abilities were affected by
lesions to the left pars opercularis (Brodmann area 44, usually known as Broca’s
area) in the inferior frontal gyrus and to the white matter adjacent to the left
supramarginal gyrus, over and above overt speech production and working
memory areas.
An alternative proposal to Vygotsky’s idea of inner speech as crucial element in
verbal thought is the so-called mentalese hypothesis (Chomsky, 1968; Fodor,
1975). A mentalese is a hypothetical language in which concepts and propositions
are represented in the mind without words. It is assumed that as a previous step to
select the language to designate something, there is a mentalese, and only later do
we provide the language form in English, Spanish, or Chinese, for example.
Mentalese is prelinguistic and occurs before the representations we are thinking
about are turned into a specific language. A mentalese consequently represents the
fundamental language of thought; a particular type of language without words or,
2 Inner Speech in Bilinguals: The Example of Calculation Abilities 29
rather to say, previous to the words. But at any rate, a particular language (English,
Spanish, Chinese) will be selected to express the mentalese. How do bilingual
individuals select that particular language?
Inner Speech in Bilinguals
In bilingual individuals a crucial question is: what is the language used in inner
speech, or more precisely, what is the language used for verbal thinking?
Grosjean (2010) conducted a survey with bilinguals and trilinguals in which the
question in what language you think was presented; that is, what is the internal lan-
guage (inner speech) used to talk to ourselves; 70% of the participants replied both
languages or all languages (for trilinguals). Grosjean considers this answer unsur-
prising because bilinguals use their languages for different purposes, in different
domains of life, and with different people. So, the inner speech in bilinguals can be
either one, depending on the specific context.
Larsen, Schrauf, Fromholt, and Rubin (2002) studied Polish immigrants in
Denmark and their use of inner speech for autobiographical memories retrieval. The
authors found that memory recalls in the native language (i.e., Polish) dropped signifi-
cantly after immigration to Denmark. In addition, they found significant differences in
inner speech usage relative to age of immigration. Those individuals who migrated
later in adulthood tend to have more instances of inner speech in Polish when com-
pared to those who migrated in their early years. Such findings might indicate that
memory representation and retrieval are closely associated to the linguistic system in
which those events (i.e., autobiographical) took place. Therefore, memories may be
linguistically encoded and dependent on the language in which the individual experi-
enced those memories. That is the case of later immigrants who may have linguisti-
cally encoded memories that are closely related to their years pre-migration in one
language (i.e., Polish) and in another (i.e., Danish) for those years after migration.
It has been suggested that L1 is preferred for emotional inner speech, even when
this language is partly attrited (Söter, 2001). Dewaele (2011) studied 386 bi- and
multilingual adults; participants were proficient in both L1 and L2 and had a con-
tinuous active use of both languages. A quantitative analysis revealed that partici-
pants preferred to use L1 for communicating feelings of anger, swearing, addressing
their children, performing mental calculations, and using inner speech. They also
perceived their L1 to be emotionally stronger than their L2 and reported lower levels
of communicative anxiety in their L1 (Dewaele, 2010). Participants reported that
their multilingualism and multiculturalism gave them a sense of empowerment and
a feeling of freedom. Using an extended sample of 1459 multilinguals (1040
females, 419 males) speaking a total of 77 different L1 s, a further analysis of inner
speech in bilinguals and multilinguals was advanced. There were 221 bilinguals,
362 trilinguals, 390 quadrilinguals, and 486 pentalinguals. Data were collected
through an online questionnaire (Dewaele & Pavlenko, 2001–2003). ANOVAs con-
firmed that languages acquired later in life are less likely to be used for inner (emo-
tional) speech compared to early acquired languages. Significant differences in

language choice for general use, inner speech and inner emotional speech in the L2,
L3, L4, and L5 were disclosed. These languages were used less frequently for inner
speech than for general use, and even less frequently for emotional inner speech.
Language used in dreaming represents a very idiosyncratic type of communica-
tion. Grosjean (2010) found in a survey that almost as many bilinguals as trilinguals
(64% in all) said that they dreamed in one or the other language, depending on the
dream (when a language was involved, of course). The conclusion seems evident:
depending on the situation and the person we are dreaming about, we will use the
one language, the other, or both. Grosjean refers to the interesting observation that
some people have reported speaking a language fluently in a dream when they are
not actually fluent in that language.
Ardila, Benettieri, Church, Orozco, and Saucedo (n.d.) analyzed private speech in
a sample of 31 18–40 years Spanish/English early sequential bilinguals (learning L2
before the age of 10 years; their native language was Spanish, and English was
learned as a consequence of moving to the US from a Latin American country); in
64% of the cases English was the language that they reported to use for thinking
(e.g., planning, reflecting) and in 68% of the cases for problem-solving/reasoning
(e.g., solving math problems); only about 15% of the participants reported to prefer
Spanish for these complex forms of cognition, and the rest considered that they used
both languages for thinking and problem-solving/reasoning. In these individuals,
although their native language was Spanish, English had become the dominant lan-
guage as a consequence of attending school in English. Simultaneous bilinguals with
English as a dominant language, on the other hand, used English in about 90% of the
cases for thinking and problem-solving/reasoning; in the remaining cases they used
both languages; Spanish was never used in thinking and problem-solving/reasoning
situations. It was concluded that in bilinguals private speech to a significant extent
corresponds to the dominant language, rather than to the native language.
Mental Calculation in Bilinguals
Mental calculations could be considered examples of inner speech. People use men-
tal calculation when they compute numbers in their heads, without using any other
computing tools, such as calculators; the most frequent mental calculations include
additions, multiplications, and divisions; however, mental calculation could involve
the use of specific mental strategies devised for specific types of problems. Many of
these strategies take advantage of or rely on the decimal numeral system. Also, the
choice of base 10 defines what methods to use and which calculations are easier to
perform mentally. For example, multiplying or dividing by ten is an easy task when
working in decimals (just move the decimal point); or in adding or subtracting, it is
the ability to recognize a number in the teens as comprising a ten and a single-digit
number (for a review see Thompson, 1999, 2000).
Some authors distinguish between mental calculations and mental arithmetic; the
first one refers to the strategy used in solving an arithmetic problem, whereas the
latter requires the mental recall of arithmetic facts alone. Mental strategies are
applied to known or quickly calculated number facts in combination with specific
properties of the number system to find the solution of a calculation whose answer
is not known (Thompson, 1999). Mental arithmetic, on the other hand, becomes
automatic with practice and may become a recall process with no need to think. For
example, doing one-digit operations (3 + 4; 8–5; 5 × 8; 10/2) or performing the mul-
tiplication tables will require just a recall process in older children or in adults.
Frequently, mental calculations contain also mental arithmetic.
Dewaele (2007) defines metal calculations as cognitive operations that involve
both language dependent and language independent processes. The following sec-
tions describe how much mental calculations depend on language and are therefore
affected by the learning of a second language in bilingual individuals.
Calculation Abilities and Language
For some developmental cognitive psychologists, language is an essential part of

mathematics. Some of the first words that are learned by toddlers are the words to
count, although mastery of counting requires the understanding of concept words
and of counting procedures (Jara-Ettinger, Piantadosi, Spelke, Levy, & Gibson,
2016). In parallel, children learned to identify magnitude, although for many the
concept of numerosity is present at birth (Gallistel & Gelman, 1992). Development
can be seen in terms of an increasingly sophisticated understanding of numerosity
and its implications, and in increasing skill in manipulating numerosities
(Butterworth, 2005). Despite the fact that children are born with an ability to recog-
nize quantity by responding to differences in the number of elements in a set, lan-
guage development plays an important role in mathematical learning. Language
development marks children’s progress to acquire the basics of elementary school
arithmetic—addition, subtraction, multiplication and division.
One important source of behavioral evidence on the association between lan-
guage and calculation abilities has come from the study of arithmetic memory in
bilingual individuals (Rusconi, Galfano, & Job, 2007). Dehaene, Spelke, Pinel,
Stanescu, and Tsivkin (1999) presented evidence that language-based processes
mediate exact arithmetic, whereas a language-independent magnitude representa-
tion mediates approximate arithmetic. To test the dissociation between these two
processes, the authors trained Russian–English bilinguals on sets of exact and
approximate multi-digit arithmetic problems in their two languages. The testing of
exact calculations after training revealed large response time (RT) costs of language
switching and poor generalization to novel problems for both languages. In contrast,
approximate arithmetic showed language independence at test, and training
generalized to new problems without loss. Dehaene et al. concluded that training
exact arithmetic was language specific, whereas approximate arithmetic depended

upon a language-independent quantity representation.
Spelke and Tsivkin (2001) supported these findings through three experiments
also performed with Russian–English bilingual college students; participants were
taught new numerical operations (Experiment 1), new arithmetic equations
(Experiments 1 and 2), or new geographical or historical facts involving numerical
or non-numerical information (Experiment 3) in either Russian or English. In all the
experiments, subjects retrieved information about exact numbers more effectively
in the language of training, and they solved trained problems more effectively than
untrained problems. In contrast, subjects retrieved information about approximate
numbers and non-numerical facts with equal efficiency in their two languages, and
their training on approximate number facts generalized to new facts of the same
type. These findings further support the idea that language appears to have an
important role in learning about exact numbers in a variety of contexts, a finding
that according to the authors has implications for bilingual education.
The association of language and math found for simple arithmetic operations has
also been seen in the solving of more complex mathematical operations, such as
word problems. Bernardo (2005) investigated the effects of using a bilingual stu-
dent’s first or second language to solve word problems in arithmetic. The author
tested two groups of Filipino–English bilingual students in the understanding and
solving of easy and difficult arithmetic word problems in students’ first and second
languages. The first group of students had Filipino and the second group had English
as their first language. The results indicated a first language advantage, meaning the
students were better able to understand and solve problems in their first language,
whether it was English or Filipino. Moreover, this advantage depended on the level
of difficulty of the problems, being more evident with the easy problems.
The findings described so far demonstrate that new learned arithmetic facts are
stored at least temporarily in the language in which they are learned; to further
explore whether there is language specificity for learning simple arithmetic facts in
bilinguals, Campbell and Dowd (2012) tested the inter operation learning transfer,
from simple multiplications (6 × 8) to additions (6 + 8) in the two languages of
bilinguals. The authors hypothesized that bilinguals may possess memory networks
of simple addition and multiplication facts in each language. To pursue this,
Campbell and Dowd (2012) examined retrieval-induced interference in Chinese–
English bilinguals’ addition-fact memory that is induced by retrieval practice of
multiplication counterparts. Participants practiced simple multiplication problems
(e.g., 6 × 7 =?), one set in Chinese and a different set in English. After training, sepa-
rate groups answered corresponding addition problems (6 + 7 =?) and control addi-
tion problems in either Chinese or English. The results demonstrated response time
(RT) costs for addition problems after their multiplication counterparts were prac-
ticed in the same language, relative to practice in the other language. This slowing
in retrieval represents direct manipulation of the long-term memory representations
of addition facts. Thus, the Campbell and Dowd experiment provided evidence of
language-specific retrieval-induced interference in Chinese–English bilinguals’
memory for addition facts.
To further explore the specificity of language in bilinguals’ math learning,

Chen, Yanke, and Campbell (2016) analyzed the degree of generalization of arith-
metic learning from one language to the other in bilingual participants. The
authors trained Chinese–English bilingual adults on different subsets of simple
addition and multiplication problems. Each operation was trained in one language
or the other. At testing after training, all addition and multiplication problems
were presented in both languages. Results showed that the RT gains for trained
problems relative to untrained problems were greater in the trained language than
in the untrained language. Subsequent analysis showed that English training pro-
duced larger RT gains for trained problems relative to untrained problems in
English at test, relative to the untrained Chinese language. In contrast, there was
no evidence with Chinese training that problem-specific RT gains differed between
Chinese and the untrained English language. The authors suggested that training
in Chinese promoted a translation strategy for English arithmetic (particularly
multiplication) that produced solid cross-language generalization of practice,
whereas training in English strengthened relatively weak, English-language arith-
metic memories and produced little generalization to Chinese (i.e., English train-
ing did not induce an English translation strategy for Chinese language trials).
The results support even more the existence of language-specific strengthening of
memory for arithmetic facts.
Further evidence for the effects of language-specific experience on arithmetic
processing in bilinguals at the cortical brain level is reported by Lin, Imada, and
Kuhl (2012) in a functional magnetic resonance imaging (fMRI) study. They
analyzed the neural correlates associated with the performance differences
observed between bilinguals’ two languages during a mental addition task. Brain
activation was recorded during mental addition when auditory problems were
presented in participants’ L1 and L2. Chinese–English bilinguals heard 2-digit
addition problems that required exact or approximate calculations. fMRI results
showed that orally presented multi-digit addition in bilinguals activates bilateral
inferior parietal and inferior frontal regions in both L1 and L2. Language differ-
ences were observed in the form of greater activation for L2 exact addition in the
left inferior frontal area. A negative correlation between brain activation and
behavioral performance during mental addition in L2 was observed in the left
inferior parietal area.
The view that there exists a close association between language and simple men-
tal calculations is not shared by all cognitive scientists. For example, Gelman and
Butterworth (2005) analyzed data from Amazonian cultures who had very restricted
number vocabularies but who were capable of manipulating magnitudes correctly,
demonstrating that numerical concepts could be language independent. The debate
is to determine the extent to which math is dissociable from other cognitive func-
tions, including language. Some authors have found that language does not always
affect word problem-solving performance among bilingual students. For example,
older students are less likely to experience difficulties associated with language-
related aspects of word problem solving (Bernardo, 2005). These results were
attributed to the students’ acquisition of problem schemas in the domain. Problem
schemas are abstract representations of the underlying structure of the word prob-
lems that allow problem solvers to model and solve the word problems in ways that
are largely unaffected by the superficial (e.g., linguistic) features of the problems
(Bernardo, 2002).
he Dynamic Process of the Inner Language of Mental

T
Calculation
The conventional view has been that people access mathematical concepts, such as
multiplication tables, more efficiently in the language in which these were learned.
So, for example, immigrants who spoke Spanish first and learned basic math in their
native language as children will turn to that language to calculate math later in life,
even if they have become proficient in another language. This hypothesis is in line
with the self-reported language choice described by Dewaele (2007) in 1454 adult
multilinguals from a variety of linguistic, social and ethnic backgrounds, in which
L1 was usually reported as the speakers’ preferred language for mental calculation.
Multilinguals’ preference for the L1 may be linked to the fact that this specific cog-
nitive operation has most probably been learnt in the L1, which was typically also
the dominant language that in most cases is the language of instruction.
The dominance of a language in bilinguals is a dynamic process that can change
over the lifetime and, therefore, the language in which math concepts are learned in
childhood does not necessarily remain the dominant language in adulthood. The
language that is more dominant changes based on the bilingual environment, par-
ticularly the educational setting. It is clear that mental calculation does not happen
exclusively in the L1. Dewaele (2007) found that L2 is reported by his multilingual
sample, on average, to be used sometimes in mental calculations. Moreover, 3% of
Dewaele’s participants reported never using the L1 anymore for mental calculation,
which shows the flexibility of language used for calculations in individuals who
have learned more than one language.
In support of this plasticity of the switch of language in calculations from L1 to
L2 is the report by Velez-Uribe and Rosselli (in press), who asked Spanish/English
bilinguals (age = 21.82 years; SD = 5.60) whose native language was Spanish and
who have learned English at an average age of 6.92 years, about the frequency of
language use in mathematical calculations; they found that 58% of them reported
using English (L2), whereas only 13.9% reported using Spanish (L1). Most of these
bilinguals had been educated in English. Likewise, Bernardo (2002) showed that
bilingual high school students performed better on arithmetic in their language of
schooling, even though it was not their native language.
More recently, Salillas and Wicha (2012) came to a similar conclusion when they
examined the relationship between bilingualism and math using brain event-related
potentials (ERPs). In their first study, they measured electrical activity in the brains
of 22 Spanish–English bilinguals performing basic multiplication calculations. All
of the volunteers were college students who grew up in Spanish-speaking families

but became proficient in English by the age of 15. They were asked to solve simple
math problems, some of which were presented in digits and others in words, in both
Spanish and English. The results confirmed the advantage for the language of
instruction, but they also showed that some individuals responded faster in the lan-
guage they used regularly and not automatically in the language in which they ini-
tially learned basic math.
In a follow-up study, Martinez-Lincoln, Cortinas, and Wicha (2015) found that
arithmetic memory networks established in childhood can be changed by experi-
ence in adulthood. They studied 14 elementary school bilingual teachers (11 female)
whose first language (L1) was also their language of learning arithmetic (LA+).
Half of the participants taught primarily in LA+ (learn = teach) and half in LA−
(learn ≠ teach). All participants showed stronger access to multiplication tables
when using the language in which they learned arithmetic during childhood (LA+)
than the other language (LA−), implying language-specific encoding of math facts.
However, most bilinguals used LA+ throughout their life, confounding the impact
of encoding and use. The authors then tested if using arithmetic facts in LA− could
reduce this LA− disadvantage. They measured event related brain potentials while
bilingual teachers judged the correctness of multiplication problems in each of their
languages. Earlier N400 peak latency was observed in both groups for the teaching
than non-teaching language, showing more efficient access to these facts with use.
LA+ teachers maintained an LA+ advantage, while LA− teachers showed equiva-
lent N400 congruency effects (for incorrect versus correct solutions) in both lan-
guages. LA− teachers also showed a late positive component that may reflect
conflict monitoring between their LA+ and a strong LA−. The authors concluded
that the LA− disadvantage for exact arithmetic established in early bilingual educa-
tion can be lessened by later use of LA−.
Conclusions
Regardless that the idea of inner speech has a long history, it was systematized by
Vygotsky who referred to three different types of speech: external or social speech,
egocentric or private speech, and inner speech. Vygotsky interprets inner speech with
an internalized verbalized thought and suggests that complex psychological pro-
cesses are associated with inner speech. From the neurological perspective, inner
speech has been related with the activity of the left inferior frontal gyrus (Broca’s
area). By the same token, inner speech is affected by lesion in this cortical area.
Bilinguals can use both languages in inner speech, depending upon the specific
circumstances. However, languages acquired later in life are less likely to be used
for inner (emotional) speech compared to early acquired languages and L1 is pre-
ferred for emotional inner speech even when this language is partly attrited.
In the case of number processing, bilinguals seem to use L1 more frequently as
the inner speech for mental calculations; however the use of L2 is also reported for
mental mathematics, particularly in cases in which education has been received in

L2; therefore, bilinguals present flexibility of the language used for mental calcula-
tions based on environmental experiences. Moreover, it seems that bilinguals
develop memory networks for calculations that are specific for each language with
distinguishable brain networks.
Acknowledgement Our gratitude to Deven Christopher for her editorial support.
References
Ardila, A., Benettieri, K., Church, Y., Orozco, A. & Saucedo, C. (in press). Private speech in simul-
taneous and early Spanish/English bilinguals. Applied Neuropsychology.
Bernardo, A. B. I. (2002). Language and mathematical problem solving among bilinguals. The
Journal of Psychology, 136(3), 283–297. doi:10.1080/00223980209604156
Bernardo, A. B. I. (2005). Language and modeling word problems in mathematics among bilin-
guals. The Journal of Psychology, 139(5), 413–425. doi:10.3200/JRLP.139.5.413-425
Butterworth, B. (2005). The development of arithmetical abilities. Journal of Child Psychology
and Psychiatry, 46(1), 3–18.
Campbell, J. I. D., & Dowd, R. R. (2012). Interoperation transfer in Chinese–English bilinguals
arithmetic. Psychological Bulletin Review, 19, 948–954. doi:10.3758/s13423-012-0277-z
Chen, Y., Yanke, J., & Campbell, J. I. (2016). Language-specific memory for everyday arithmetic
facts in Chinese-English bilinguals. Psychonomic Bulletin & Review, 23(2), 526–532.
Chomsky, N. (1968). Language and mind. New York: Harcourt, Brace & World.
Dehaene, S., Spelke, E., Pinel, P., Stanescu, R., & Tsivkin, S. (1999). Sources of mathematical
thinking: Behavioral and brain-imaging evidence. Science, 284(5416), 970–974.
Dewaele, J. M. (2007). Multilinguals’ language choice for mental calculation. Intercultural
Pragmatics, 4–3, 343–376.
Dewaele, J. M. (2010). Emotions in multiple languages. Basingstoke: Palgrave Macmillan.
Dewaele, J. M. (2011). Self-reported use and perception of the L1 and L2 among maximally pro-
ficient bi-and multilinguals: A quantitative and qualitative investigation. International Journal
of the Sociology of Language, 2011(208), 25–51.
Dewaele, J. M., & Pavlenko, A. (2001-2003). Web questionnaire ‘Bilingualism and Emotions’.
London: University of London.
Ehrich, J. F. (2006). Vygotskian inner speech and the reading process. Australian Journal of
Educational & Developmental Psychology, 6, 12–25.
Fodor, J. A. (1975). The language of thought. Cambridge: Harvard University Press.
Gallistel, C. R., & Gelman, R. (1992). Preverbal and verbal counting and computation. Cognition,
44(1), 43–74.
Geva, S., Jones, P. S., Crinion, J. T., Price, C. J., Baron, J. C., & Warburton, E. A. (2011). The
neural correlates of inner speech defined by voxel-based lesion–symptom mapping. Brain,
134(10), 3071–3082.
Gelman, R., & Butterworth, B. (2005). Number and language: how are they related?. Trends in
Cognitive Sciences, 9(1), 6–10.
Grosjean, F. (2010). Bilingual: Life and reality. Cambridge, MA: Harvard University Press.
Jara-Ettinger, J., Piantadosi, S., Spelke, E. S., Levy, R., & Gibson, E. (2016). Mastery of the
logic of natural numbers is not the result of mastery of counting: Evidence from late counters.
Developmental Science.
Johnson, J. R. (1994). Intrapersonal spoken language: An attribute of extrapersonal competency. In

D. R. Vocate (Ed.), Intrapersonal communication: Different voices. Different minds (pp. 169–
192). Hillsdale, NJ: Lawrence Erlbaum Associates.
Jones, P. E. (2009). From ‘external speech’ to ‘inner speech ‘in Vygotsky: A critical appraisal and
fresh perspectives. Language & Communication, 29(2), 166–181.
Larsen, S. F., Schrauf, R. W., Fromholt, P., & Rubin, D. C. (2002). Inner speech and bilingual
autobiographical memory: A Polish-Danish cross-cultural study. Memory, 10(1), 45–54.
Lin, J. F. L., Imada, T., & Kuhl, P. K. (2012). Mental addition in bilinguals: An fMRI study of task-
related and performance-related activation. Cerebral Cortex, 22(8), 1851–1861.
Martinez-Lincoln, A., Cortinas, C., & Wicha, N. Y. (2015). Arithmetic memory networks estab-
lished in childhood are changed by experience in adulthood. Neuroscience Letters, 584,
325–330.
McGuire, P. K., Silbersweig, D. A., Murray, R. M., David, A. S., Frackowiak, R. S. J., & Frith,
C. D. (1996). Functional anatomy of inner speech and auditory verbal imagery. Psychological
Medicine, 26(1), 29–38.
Morin, A., & Michaud, J. (2007). Self-awareness and the left inferior frontal gyrus: Inner speech
use during self-related processing. Brain Research Bulletin, 74(6), 387–396.
Plato. (1961). Collected dialogues. In E. Hamilton & H. Cairns (Eds.), Bollingen series LXXI.
New York: Pantheon.
Rusconi, E., Galfano, G., & Job, R. (2007). Bilingualism and cognitive arithmetic. In I. Kecskés
& L. Albertazzi (Eds.), Cognitive aspects of bilingualism (pp. 153–174). Dordrecht: Springer.
Salillas, E., & Wicha, N. Y. (2012). Early learning shapes the memory networks for arithmetic
evidence from brain potentials in bilinguals. Psychological Science, 23(7), 745–755.
Söter, A. (2001). Straddling three worlds. In D. Belcher & U. Connor (Eds.), Reflections on multi-
literate lives (pp. 67–73). Clevedon: Multilingual Matters.
Spelke, E. S., & Tsivkin, S. (2001). Language and number: A bilingual training study. Cognition,
78(1), 45–88.
Thompson, I. (1999). Mental calculation strategies for addition and subtraction. Part 1. Mathematics
in School, 28(5), 2–4.
Thompson, I. (2000). Mental calculation strategies for addition and subtraction: Part 2. Mathematics
in School, 29(1), 24–26.
Vélez-Uribe, I & Rosselli, M. (in press) The Auditory and Visual Appraisal of Emotion-Related
Words in Spanish/ English Bilinguals. Bilingualism: Language and Cognition.
Vygotsky, L. S. (1929). The problem of the cultural development of the child II. The Journal of
Genetic Psychology, 36, 415–432.
Vygotsky, L. S. (1934a/1968). Mind in society. Cambridge, MA: Harvard University Press.
Vygotsky, L. S. (1934b/2012). Thought and language. Cambridge, MA: MIT Press.
Wiley, N. (2006). Inner speech as a language: A Saussurean inquiry. Journal for the Theory of
Social Behaviour, 36(3), 319–341.
Chapter 3
Emotional Associations of Words
in L1 and L2 in Bilinguals
Contents
Introduction 39
Emotional Associations in L1 and L2 Vocabulary 40
Research Approaches for Studying Emotions in Bilinguals 42
Conclusions and Future Research 68
References 70
Introduction
Bilingualism can be defined as the habitual use of two languages in one person or in
one region (Real Academia de la Lengua Española, 2001). However, bilingualism is
significantly heterogeneous, generating a wide arrange of variables to consider in
research. In the first place, being bilingual is not a dichotomous decision that results in
a simple Yes/No answer to the question: Are you bilingual? Bilingualism exists in a
continuum, with great variability among the bilingual population in the second lan-
guage (L2) domain. Such variability depends on when the L2 was acquired (childhood
vs. adulthood), how it was acquired (academic vs. informal learning), and how well it
was learned (level of proficiency). Moreover, the level of proficiency within each lan-
guage may be unequal across different linguistic domains within an individual. For
example, the second generation Spanish–English bilingual living in Miami may have
the phonology (i.e., sound system) of a native speaker in both English and Spanish, but
M. Roselli
I. Vélez-Uribe (*)
Florida Atlantic University, Davie, FL, USA
e-mail: ivelezur@fau.edu
A. Ardila

40 M. Roselli et al.
the grammatical use may be distant form that of a native speaker in the language in
which no formal training has been received (which in most cases is Spanish).
Some additional variables should be taken into account, such as the social context
and language patterns of the bilingual individual under study. For example, the social
context and bilingual cultural and linguistic environments are different for cases of
either elective bilingualism or nonelective (forced/circumstantial) bilingualism. In
elective bilingualism, the L2 is acquired through formal language classes, in a lan-
guage other than what is used in the bilingual’s daily social environment. For instance,
an English-speaking person living in an English-speaking country may learn Spanish
as their L2, though this adoption may not be deemed a necessity. In contrast, non-
elective or circumstantial bilingualism, results when a person is required to learn the
L2 in order to live in a new linguistic environment, as it is the case for an English
speaker whose migration to Spain makes learning Spanish a necessity.
The age of acquisition (AoA) of an L2, independent of the context in which the
languages are learned, also plays an important role in the definition of bilingualism.
By some definitions, learning L2 before age 12 defines early bilingualism, whereas
late bilingualism implies the learning of L2 after age 12. Early bilingualism is fre-
quently associated with simultaneous learning of the two languages, while late
bilingualism is related to successive learning of them (L2 is learned after the first
language [L1]). The linguistic experience, particularly that involving emotion
words, could differ significantly depending on the AoA of the L2.
An additional distinction in bilingualism that may influence how emotion words are
perceived in L1 and L2 is the functional distance between the two languages; this dis-
tance can be large (strong bilingualism) or small (weak bilingualism; Ardila, 2007). In
strong bilingualism, the two languages are significantly different linguistically, or dis-
tant (for example Chinese and English) whereas in weak bilingualism the two spoken
languages are more similar (for example Italian and Spanish). Although to our knowl-
edge no research on emotions has been done comparing these specific dimensions of
bilingualism, one may propose that in strong bilingualism the emotional response to
each language is dissimilar and independent of each other, whereas in weak bilingual-
ism the emotional experience with one language may influence the other.
Finally, the culture identification, or the attitude towards the culture correspond-
ing to the target language, may be a pertinent variable in adopting the emotional
value of words in a specific language. A factor known as emotional acculturation
greatly impacts the level of emotional concordance of the individual to the host
culture (De Leersnyder, Mesquita, & Kim, 2011). Attitudes toward a culture may
predispose the individual to a more or less evident automatic emotional response to
stimuli that represent the language of that culture.
Emotional Associations in L1 and L2 Vocabulary
One of the debates in bilingualism research focuses on whether bilinguals process

information differently in each language and whether these individuals experience
emotions in a different way, depending on which language they are using. Even
3 Emotional Associations of Words in L1 and L2 in Bilinguals 41
though the acquisition of semantic and syntactic elements of language seems suc-
cessful, sequential bilinguals, who have learned their languages at different ages,
frequently report the subjective impression that L2 is less emotional than L1
(Dewaele, 2004a, 2008).
Words can have an emotional value. This is particularly true with special words
used to express emotional states (i.e., sadness) or those called emotion laden words
that induce an emotion (i.e., war). To measure the emotional value of these words,
researchers have used emotional ratings whereby participants rate their perception
of emotional words (Bradley & Lang, 1999; Redondo, Fraga, Padrón, & Comesaña,
2007). The first and most important rating concerns the valence (or pleasantness) of
the emotions invoked by the word. Valence is commonly measured within a con-
tinuum, from extremely negative to extremely positive (e.g., unhappy to happy; or
unpleasant to pleasant), indicating the extent to which a stimulus, in this case a
word, is negative or positive. The second type of emotional rating addresses the
degree of arousal evoked by the word, or the extent to which the word is calming
(decreases arousal) or exciting (increases arousal). Valence and arousal are theoreti-
cally orthogonal (Kuperman, Estes, Brysbaert, & Warriner, 2014): negative words
can be either relaxing (e.g., dirt) or exciting (e.g., snake), and positive words can
also be relaxing (e.g., sleep) or exciting (e.g., sex). Differences in the arousal and
valence of words correlate with variations in automatic nervous system response
measured through changes in the electrical conductivity of the skin (Harris, Aycicegi
& Gleason, 2003; Harris, 2004). In addition, neurological dissociations in arousal
and valence have been reported resulting in the activation of different cortical net-
works (Kensinger & Corkin, 2004; LaBar & Cabeza, 2006).
It has been observed that the emotional associations of these words can differ for
the L1 and L2 of the bilingual speaker. Bilingual and multilingual participants in
studies of self-report ratings (Dewaele, 2004a, 2008) have consistently described
emotional differences across languages related to the chronological order of the
acquisition of each language. It has been suggested that the first acquired language
is associated with deeper emotional connotations, possibly related to the context of
acquisition (Altarriba, 2008), whereas the L2 is capable of providing greater emo-
tional distance.
The emotional dissociation of the two languages in bilinguals has also been doc-
umented in the clinical psychological settings. For instance, some bilinguals seem
to resort to selective code-switching (switching between languages), transferring to
the language that provides greater emotional distance when discussing embarrass-
ing topics (Bond & Lai, 1986), and use it as a way to manage the emotional tone of
the therapy sessions (Dewaele & Costa, 2013).
In a recent study, Costa et al. (2014) provided additional support for the hypoth-
esis of greater emotional detachment in L2. The authors proposed that if L2 indeed
provides more detachment, it could be, as a result, more utilitarian. Since reduced
emotionality would increase rationality, it would allow for decisions that are more
useful when the individual is confronted with a moral dilemma in L2 than in L1.
The authors found that when bilinguals are faced with the decision of pushing a man
in front of a train, killing him, but saving five people, participants chose signifi-
cantly different when the option was presented in L1 or L2; more participants
selected the utilitarian option when L2 was used. In a less emotional condition
involving switching the train tracks to choose whether to kill one man or five, bilin-
gual participants selected equally, regardless of the language of presentation. The
results of this study showed that L2 seemingly reduced emotionality and increased
utilitarianism, with an influence of level of proficiency; higher levels of proficiency
decreasing the difference in responses between L1 and L2.
Research Approaches for Studying Emotions in Bilinguals
A variety of research paradigms have been used to study affective processing in

bilingual speakers. We describe six of these approaches following Pavlenko’s (2012)
classification: clinical, introspective, cognitive, autonomic, electrophysiological,
and neuroimaging approaches.
Clinical Approaches
Clinical approaches focus on the affective reactivity or the pattern of arousal and
language disturbances displayed in conversations of negatively valenced topics in
patients with psychological disorders (Pavlenko, 2012). Clinical studies provide
evidence of the bilingual patients’ discrepant affective reactivity when using taboo
words or discussing the same subject in different languages (Amati-Mehler,
Argentieri, & Canestri, 1993); however, case studies of patients in therapy are the
frequent research methodology used by clinical approaches limiting the generaliza-
tion of results. Also, as stated by Pavlenko (2012), limitations of this approach
include insufficient information provided about the patients’ bilingualism; most fre-
quently, there is no description of the language learning paths, language dominance,
or language proficiency, which limits even more the significance of results from this
approach in the understanding of bilingualism and emotions.
Introspective Approaches
Introspective approaches analyze bilingual speakers’ personal perceptions of emo-

tionality of their respective languages, language choice for emotional expression,
and language attitudes among others. As indicated before, the acquisition of emo-
tional connotations in an L2 might be delayed when compared to the acquisition of
semantic and syntactic elements. Bilinguals’ self-reports suggest that this delay in
the acquisition of emotionality in L2 might facilitate approaching difficult or embar-
rassing topics (Bond & Lai, 1986) and anxiety-provoking subjects (Dewaele &
Pavlenko, 2002) by providing more emotional distance. This detaching effect might
also affect swear and taboo words which could feel less strong in an L2 (Dewaele,
2004b); similarly, words connoting positive emotions, such as terms of endearment,
might feel stronger when expressed in the dominant language (Dewaele, 2008).
However, differences in language choice for swearing might reflect not only order
of acquisition, but also frequency of use and the context in which each language was
acquired (Dewaele, 2004b).
Pursuing this further, Dewaele (2004a) asked 1039 bilinguals if swear and taboo
words had the same emotional weight in their two languages. The analyses indicated
that swearwords were indeed felt more strongly in L1 than in L2. Further language
learning increased the order of acquisition, thereby lessening the strength of feelings
elicited by swearwords. Dewaele found an influence of context of acquisition, where
naturalistic (immersion in second culture) and mixed context (immersion and
instructed) of acquisition provided stronger associations to taboo words, along with
frequency of use, earlier AoA and higher level of proficiency (Dewaele, 2004a). These
findings provide support for the hypothesis that the acquisition of emotional elements
in an L2 is delayed with respect to other linguistic elements and raises questions about
the variables that can possibly intervene in this process. Dewaele (2008) proposes that
multiple sociocultural, linguistic, situational, and relational factors influence the way
individuals appraise the emotional weight of a sentence. When analyzing participants’
appraisal of the phrase I love you, the author found that language dominance had a
stronger influence than gender, trait emotional intelligence, and education level,
although overall this sample was highly educated. Frequency of use seemed to
decrease the perceived emotional strength of the sentence; this is probably an effect of
desensitization, since higher frequency of use led to reports of a decrease in intensity.
The emotional weight of the phrase I love you was also reported with almost 50% of
the participants considering it to have more emotional weight in their native language.
These findings suggest the meaning of the expression I love you in L2 has been
grasped, although the full conceptual representation has not yet been acquired, mak-
ing its emotional significance different from that of the same expression in L1.
Even though single words are commonly used in studies of emotions in bilin-
guals, studies analyzing differences in the appraisal of emotional intensity in verbal
stimuli are scarce. Winskel (2013) analyzed ratings of emotional intensity of single
words in bilingual participants (Thai and English) who had acquired their L2
(English) in an instructional setting with no immersive experience, and an English
monolingual group. Comparison between ratings of neutral and negative words did
not show any significant differences between languages. Furthermore, the ratings of
English words did not differ between bilingual and monolingual participants. More
recently, Vélez-Uribe and Rosselli (in press) tested the valence of emotionally
loaded words in L1 (Spanish) and L2 (English) in 101 Spanish–English bilinguals
living in the USA. Participants appraised three categories of words (positive, nega-
tive, and taboo) in each language and in two sensory modalities (visual and audi-
tory). The effect of language was present in two directions in both sensory modalities:
positive word ratings were rated as more positive in English than in Spanish; simi-
larly, negative words were judged as more negative in English than in Spanish.
However, taboo words were appraised as more negative in Spanish than in English.
Using regression analyses, it was found that in the visual category, percent of life in
the USA and Spanish proficiency were significant predictors of the valence of posi-
tive words, whereas in the auditory sensory modality, both English and Spanish
proficiency had significant predictor value for positive words. In the visual negative
category, English proficiency and Latino cultural identity were significant predic-
tors. As expected, correlation analyses reveled that participants with a higher per-
cent of life in the USA tended to have higher proficiency in English and lower
proficiency in Spanish. In addition, participants who acquired English later in life
presented lower levels of English proficiency and higher levels of Spanish profi-
ciency. There was a rather unusual significant correlation in these results: partici-
pants with higher self-reported levels of Spanish proficiency tended to have higher
scores in the USA cultural identity scores. It could be possible that those individuals
in this study who were highly proficient in Spanish were more assimilated to the
mainstream American culture than to their Latin culture of origin.
Attitudes toward a language have also been studied in the context of the emotions
and language in bilinguals’ emotion. Language attitudes are measured by asking
participants to give affective and non-affective attributes to taped texts in different
languages and dialects and with and without foreign accents. Results show that non-
linguists are as good as linguists in recognizing dialects in their L1 and in distin-
guishing between accented and non-accented speech (e.g., Bresnahan, Ohashi,
Nebashi, Liu, & Shearman, 2002). These differences are frequently conferred with
affective meanings, with participants commonly rating speakers of their own lan-
guage significantly higher on affective attributes than accented speakers and speakers
of socially marginalized/minority groups, such as African-American (e.g., Bresnahan
et al., 2002). Attitudes concerning a language can be influenced by AoA. For exam-
ple, negative attitudes may be associated with languages learned later in life (Dewaele,
2010) and can also be influenced by negative experiences even with people who
speak the native language. For example, Jewish Holocaust survivors developed nega-
tive attitudes toward the German language after the war, despite the fact that German
was the native language for many of them (Pavlenko, 2012).
Cognitive Approaches
Cognitive approaches use behavioral methods to determine whether emotional

words are processed differently from neutral words in bilingual participants’ L1
and L2; these words are presented visually or auditorily with the assumption that
their processing is automatic (or involuntary) and immediate. The classical cogni-
tive paradigms use words with different valence (positive, negative, neutral) and
levels of arousal (high, moderate, low). The emotional effects measured with
these methods include changes in memory recall and facilitation or interference
through congruency or affective priming. Table 3.1 presents the description of
different studies using these two paradigms; each of the studies presented in the
table is described below.
Table 3.1 Cognitive paradigms in the study of emotional language in bilingual participants
Familiarity
Paradigm/ N Mean age N Type of Active of words
reference Bilinguals (SD) F:M Languages Monolinguals bilinguals Findings Type of stimuli bilinguals tested
1. Emotion–memory effect: recall tasks
Anooshian 36 30 years Not Spanish/ None Proficient late More emotional than neutral words Surprise recall Yes No
and Hertel reported English bilinguals: were recalled when the words were after rating
(1994) L2 acquisition presented in the native language, emotional and
after age 8 but not when presented in the neutral words
(mean age 16.3 second language based on level
for the Spanish of difficulty and
native speakers emotionality
and 18.4 for the
English native
speakers)
Ayçiçeği 42 28 23:17 Turkish/ None Late learners of Stronger emotion–memory effect Surprise free Yes No
and Harris English English (after in L2 for the free recall of words recall or
(2004) age 12 and 18) carrying negative associations recognition
more proficient while recall of positive words was after rating
in reading and similar in L1 and L2; in negative and
understanding recognition in L2 positive, taboo words,
negative, and taboo words had an and reprimands
emotion–memory advantage; for
L1 the advantage was observed
only for taboo words
(continued)
Table 3.1 (continued)
Familiarity
Ayçiçegi- 59 21 (3.9) 52:7 Turkish/ None Nonproficient Across the four study tasks, Surprise recall No No
Dinn and English bilinguals who emotion–memory effects were of emotion
Caldwell- acquired equally strong in L1 and L2, except words after
Harris instructional for higher recall in L2 for emotion rating
(2009) English after reprimands. Taboo words and task, letter
age 12 reprimands elicited higher recalled counting, and
rates than positive words, which in translation tasks
turn, presented higher recall rates
than negative and neutral words. This
was true on all tasks and through
both languages. Negative and neutral
words presented similar recall rates
in all languages. Recall differences
were related to the type of task
Ferre et al. 75 19.9 (4.2) 61:14 Spanish/ None Proficient early The two groups showed a Surprise free Yes No
(2010) Catalan bilinguals. Mean superiority in recall for emotional recall for
Experiment (n = 33) age of words that was of the same positive,
2 and Acquisition of magnitude in their first and second negative, and
Catalan/ L2 Spanish 2.8 languages neutral words
Spanish (2.8) and
(n = 42) Catalan 2.2 (1.5)
Ferre et al. 35 26.8 (9.3) 22:13 Spanish/ None Proficient/ No difference between positive and Surprise free No No
(2010) English unbalanced negative words either in Spanish or recall for
Experiment acquisition of in English positive,
3 L2 at age 8.6 No effect of the type of word and negative, and
(3.80) language, nor their interaction, on neutral words
the magnitudes of the emotionality
effects
2. Lexical decision task
Ponari Total: Early L2 Early Spanish/ 95 Native Early L2: mean No difference in the type and Lexical Yes No
et al. 156 22.63, (6.77) 65:7 English, speakers of age of English direction of the emotion effect decision task on
(2015) Early:77 Late L2: Late: early and English age acquisition between native speakers and both late tightly matched
Late: 79 31 (9.03) 60:19 late high 22.58 (7.30) 4.07, (1.89); and early high-proficient L2 speakers. negative,
proficient 60 females Late L2 Valenced words were recognized positive, and
bilinguals mean AoA faster than neutral words by both neutral words
11.95 (2.88) native speakers and L2 participants,
despite Late L2 being overall slower
than native speakers in performing
the task. This valence effect was not
modulated by language similarity or
language immersion: They found
comparable emotion effects across all
language families, regardless of AEA,
with no evidence for an effect of
cognates and, crucially, no evidence
for an effect of country of residence,
duration of stay, frequency, and
domain of language use
3. Emotional Stroop and Simon interference tasks
Eilola et al. 34 28.4 (5.35) 29:5 Finnish/ None Late/ Significantly more errors were Emotional No Yes
(2007) English unbalanced; age made when taboo words were Stroop paradigm
of acquisition compared to neutral words but not including
of L2 9.7 years when compared to negative or positive,
positive words negative, taboo,
Similar pattern of negativity effect and neutral
for both L1 and L2 in emotional words
and Taboo Stroop
(continued)
Familiarity
Eilola and 31 23.4 (3.1) 22:9 Greek/ 32 (7 males Proficient Native English speakers made Emotional Yes Yes
Havelka English and 25 bilinguals more errors than nonnative; larger Stroop paradigm
(2011) females) Learned latencies for negative and taboo including
English after words; no effect of language group positive,
the age of 6. negative, taboo,
Mean 8.4 (1.9) and neutral
words
Sutton 64 Not Not Spanish/ None Early highly Equal size of interference effect in Emotional Yes No
et al. specified specified English proficient L1 and L2 (response latencies). No Stroop task
(2007) bilinguals; both difference in errors between
languages languages was found
learned before
the age of 7.
English
4.9 years,
Spanish 1.9
Colbeck 24 21.7 16:8 Chinese/ 20 native Bilinguals were More correct responses were found Rapid Visual Yes No
and Bowers English English studying for a in the neutral condition than the Presentation
(2012) speakers (12 degree in taboo condition. Native English task using a
females, age English no speakers did more poorly on the modified
M = 20.4) specifications taboo/sexual word condition than English Stroop
are given the Chinese/English bilinguals task
regarding
history of
bilingualism
Altarriba 32 College Not Spanish/ None English was Planned comparisons revealed Affective Yes No
and students age specified English acquired before significant congruency effects in Simon Task
Basnight- is not age 7; slightly English for both the positive and including
Brown specified more proficient negative items. Congruency effects positive and
(2010) in English—the in Spanish were significant for negative words
Experiment second participants and marginally half in Spanish
3 language significant for items for the and the other
negative items, and significant for half in English
both participants and items for the
positive items. Overall, the effects
appear to be more robust in English
than in Spanish, for emotion-laden
words. However, none of the
individual congruency effects was
significantly different from each
other. The effects of valance and
congruency were equal across
languages as they proceeded
emotion-laden words
Altarriba 34 College Not Spanish/ None English was Significant congruency effects in Affective Yes No
and students age specified English acquired before English for both the negative and Simon Task
Basnight- is not age 7; slightly positive conditions. In Spanish, the including words
Brown specified more proficient effect was significant only for the that label
(2010) in English—the negative condition. None of the positive and
Experiment second significant effects were reliably negative
4 language different from each other, while emotions half
negative items produced an in Spanish and
Affective Simon Effect, positive the other half in
items did not, in English for the English
monolingual participants and in
Spanish for the bilingual
participants
(continued)
Familiarity
Winskel 54 20.90 (2.08) 34:20 Thai / 54 Mean age of Stroop effect was present only in Emotion Stroop No Yes
(2013) English acquisition = 7. the native language of bilinguals with negative
All learned in (Thai) and monolinguals (English). emotional and
instructional No significant differences in negative
setting with no ratings of emotional intensity emotion words
immersion/ between languages and, neutral
naturalistic words;
experience followed by an
emotionality-
rating task
The Emotion–Memory Effect: Recall Tasks
Stimuli with high emotional valence seem to take precedence in memory encoding
processes and result in higher recall rates in memory tasks. This effect, known as
emotion–memory effect, is reflected in the superior recall of emotion words com-
pared to neutral words (Reber, Perrig, Flammer, & Walter, 1994) and it is observed
when participants perform a task (such as rating the emotional intensity of words or
making a lexical decision) and are later asked to recall as many words as possible
from the prior task. The superior recall of emotion words over neutral words is
explained by the automatic elicitation of a deeper level of processing than neutral
words; the deeper level of processing is likely due to the emotion words’ inherent
interest or because emotional stimuli recruit higher levels of attention (MacKay &
Ahmetzanov, 2005). In bilinguals, this memory phenomenon combined with the
finding that L1 is frequently experienced as more emotional than L2, results in dif-
ferences in the recall rates between L1 and L2 in a memory task.
If the emotional intensity of L2 is in fact lower, then recall rates might be lower
for emotion words in L2. Anooshian and Hertel (1994) hypothesized that this effect
would be evident when analyzing AoA as a factor. They studied the free recall of
Spanish and English emotional and neutral words in 36 late Spanish–English bilin-
guals who had acquired fluency in their L2 after 8 years of age. Half of the sample
had Spanish as the native language/L1 (mean AoA of English/L2 was 16.3 years)
and the other half had English as their native language/L1 (mean AoA of Spanish/
L2 was 18.4 years). Findings indicated that emotional words presented in the native
language result in higher recall rates than neutral words, when they were rated for
ease of pronunciation (unemotional tasks) and emotional intensity (emotional task);
with an overall recall advantage for L1 in all groups (Anooshian & Hertel, 1994).
These results were unrelated to the specific native language (Spanish or English),
suggesting that while the native language results in the simultaneous development
of linguistic and emotional connotations, the learning of the L2 after age 12 (most
did) results in separate development of the linguistic/emotional connotations of that
language. According to Anooshian and Hertel (1994), this emotional dissociation
between L1 and L2 seems to be independent of the characteristics of the language
or to the bilingual’s experience in a particular linguistic subculture.
In a similar study, Ayçiçeği, and Harris (2004; see also Heredia & García, this
volume) evaluated differences between auditory and visual processing and the
resulting free recall/recognition rates in L1 and L2 with the prediction that the audi-
tory presentation might lead to deeper processing possibly augmenting the emo-
tion–memory effect. Turkish–English bilinguals who had learned English after
12 years of age were presented with stimuli in the two sensory modalities and five
categories, positive, negative, neutral and taboo words, and childhood reprimands.
Surprise recall (in writing) and recognition (from a list of written words) were tested
after a rating task was performed. The authors hypothesized that negative words
might trigger avoidance mechanisms and possibly present a recall disadvantage that
would be higher in L1; on the other hand, positive words and reprimands would
present a recall advantage in L1. Results showed no sensory modality effect. The
emotion word advantage score was obtained by subtracting the number of neutral
words from mean recall of each emotion word category. In L2, all emotion word
categories except negative words showed an emotion word advantage. In L1, taboo
words and negative words were significantly higher than neutral words. To further
test the strength of the emotion advantage of L2, the authors obtained the L2 supe-
riority score by subtracting L1 emotional advantage mean from the L2 means, con-
cluding that the emotion–memory effect using surprise recall was stronger for
words with negative connotation presented in L2, while the recall of positive words
was similar in L1 and L2. Contrary to their expectations, reprimands showed good
recall in L2 (Ayçiçeği & Harris, 2004). In the recognition condition, the L2 advan-
tage was observed in the recall of positive, negative and taboo words, for L1 the
advantage was only seen for taboo words. The authors propose that the novelty and
unusualness of the taboo words and reprimands in the L2 could have played a role
in making them more salient, hence improving their recall. One shortcoming of the
study is the lack of control for the participants’ familiarity with the emotion words
used in the experiment since the authors point out that several participants infor-
mally expressed more familiarity with the Turkish words than the English words,
perceiving them as more natural since they have had experience with these words
within the childhood context. Although these participants were immersed in the
context of the L2 (studying in an English-speaking country) at the time of testing,
they had been living there only for an average of 2.5 years; a very short period to
experience emotion words in the immersion context. Therefore, in this study, low
familiarity of participants with the emotion words in L2 was seemingly a confound-
ing factor in the high emotion advantage scores found for this language.
In a follow-up study, Ayçiçegi-Dinn and Caldwell-Harris (2009) proposed that
these results could be related to the immersion context favoring activation of the
language dominant in the testing conditions and proposed that the memory effect
could be obtained by controlling for levels of processing in different tasks. They
suggested that emotional activation elicited by items presented in L2 can be favored
in participants immersed in the L2 environment, and the tasks that require deeper
levels of processing would increase recall rates. In four different tasks varying in
level of processing, 59 Turkish native speakers with high proficiency in English
rated the items for emotional intensity in (a) a rating task (rating the items for emo-
sional intensity); (b) a shallow processing task (counting the number of letters con-
tained in a closed circle); (c) a translation task; and (d) a word association task
(providing as many word associates as possible in 10 s). The emotion–memory
effect was equally strong for both languages averaged over the four tasks, but it was
restricted to native language in the emotion intensity-rating task. On the letter-
counting task, the emotion–memory effect was very similar in both languages and
was the strongest of all four tasks, suggesting that automatic processing of emotion
stimuli was similar in both languages (Ayçiçegi-Dinn & Caldwell-Harris, 2009; see
also Heredia & García, this volume).
In support of the equivalent emotion–memory effects in L1 and L2 are recent
findings by Ponari et al. (2015) and Ferré, García, Fraga, Sánchez-Casas, and
Molero (2010). Ponari et al. (2015) used a lexical decision task on negative, posi-
tive, and neutral words, to compare the emotion effect on 95 English monolinguals
and 156 highly proficient bilinguals whose L2 was English, but who spoke different
languages as their L1. Bilinguals showed the same facilitation in processing emo-
tionally valence? words as native English speakers, regardless of their L1, AoA of
English, or frequency and context of English use. In a series of experiments, Ferré
et al. (2010) tested memory for positive, negative, and neutral words in two groups
of proficient bilinguals of Spanish and Catalan who had acquired the L2 early in life
in an immersion context and who differed in their language dominance; 42 were
dominant in Catalan and 33 were dominant in Spanish. They also tested a group of
35 proficient Spanish–English bilinguals who had learned the L2 later in life but
before age 12 in an instructional setting. They found no differences in recall between
the two languages in early balanced bilinguals or late unbalanced bilinguals. The
null differential effects of emotions in L1 and L2 reported by Ferré et al. (2010) in
the Spanish–Catalan bilinguals could be interpreted as the result of the linguistic
similarities between L1 and L2 (Catalan and Spanish). In addition, both languages
were acquired in parallel (simultaneously) and before the age of three. Furthermore,
these bilinguals were active bilinguals living in a Spanish–Catalan bilingual social
context. However, words seemed to have the same emotional intensity in the first
and in the L2 in the third group of bilinguals (as well Spanish–English bilinguals)
and the effects of age and context of acquisition were not significant. Altogether,
these results led the authors to conclude that neither language dominance nor the
type of context where the language is learned, the AoA of the L2. or the similarity
between languages, seem to have any effect on memory for emotional words in the
L2, at least in proficient bilinguals.
As stated by Ferré et al. (2010), it could be possible that the effect of AoA
can be observed only between groups of extreme bilinguals, for instance, when
early and late bilinguals are compared. Ayçiçegi-Dinn and Caldwell-Harris
(2009) argue for the importance of considering variations in results depending
on the task. The so called emotion–memory effect is obtained when participants
perform certain tasks, therefore it seems that the type of task is a relevant factor
affecting the emotion–memory effect. Future research should look at this effect
in tasks of different levels of difficulty and comparing performance in shallow
versus deep processing tasks (Ayçiçegi-Dinn & Caldwell-Harris, 2009). Finally,
many studies use translation of the stimulus words from one language to another
without taking into consideration the important culture differences in the use of
these words.
As shown in the three studies mentioned in the preceding paragraphs, there
seems to be a recall advantage for words with emotional associations; however,
some contradictions have emerged with regard to which language presents this
recall advantage. For example, in late bilinguals, Anooshian and Hertel (1994)
found that emotion words were better recalled in L1; Ayçiçeği and Harris (2004)
found the effect in the opposite direction. Altogether, this seems to suggest that the
emotion–memory effect is not confined to the native language, indicating that addi-
tional factors might have been involved. Additionally, Ayçiçegi-Dinn and Caldwell-
Harris (2009) found that even when controlling for depth of processing, the effect
was still not exclusive to L1.
There seem to be three major implications for these results. First, the emotion–
memory effect in bilinguals seems to be more related to the characteristics of the
stimuli (emotionality or novelty) than to the order and AoA of the languages in
which they are presented. Second, these inconsistencies may suggest that recall
tasks are not the best suited to test the differences in the emotional experience of
bilinguals in L1 and L2. Third, the theme of the differences in the characteristics of
the samples recurs: some samples consisted of only late bilinguals (Anooshian &
Hertel, 1994; Ayçiçeği & Harris, 2004; Ayçiçegi-Dinn & Caldwell-Harris, 2009),
while others made comparisons only between early bilinguals (Ferré et al., 2010).
The Emotional Stroop and Other Emotional Interference Tasks
A commonly used task to evaluate latencies elicited by emotional versus nonemo-

tional stimuli is the emotional Stroop task. In this variation of the traditional color
Stroop task, participants are presented with emotion words (e.g., anger, fear, jeal-
ousy) and neutral words (e.g., boat, car, train) instead of color congruent (RED in
red ink) and color incongruent (RED in blue ink) words of the traditional Stroop
task. Participants are asked to simply respond, manually or vocally, to the color of
the words presented. The interference effect in the emotional Stroop task is a result
of the emotional content of the words; emotional stimuli tend to result in longer
reaction times (RTs) than neutral stimuli, which are interpreted as an indication of
higher interference for those items. In bilingual studies, hypotheses are based on the
rationale that if there were more automaticity in L1 than in L2, then higher interfer-
ence levels in L1 would be present, resulting in higher RTs. Therefore, this effect
would reflect the influence of valence and the level of automaticity present in each
language, offering some insight into the differential processing of emotion words in
bilinguals (Altarriba, 2008).
Eilola, Havelka, and Sharma (2007) investigated this hypothesis—higher Stroop
interference in L1 compared to L2—at the level of single words that imply threat
(negative and taboo). Negative and taboo words produced significantly longer RTs
compared to neutral words when presented in the native language. These differences
were not observed between positive and neutral words. However, the effect for nega-
tive and taboo words was not exclusive to L1, thus extending to L2. The participants
were 20–44-year-old Finnish–English bilinguals (they started to learn English
between the ages of 7 and 13) living in their native language environment, but very
proficient in both languages (Eilola et al., 2007). The lack of differences between the
effect in L1 and L2 could be attributed to high levels of proficiency in both languages
presented in the sample, as well as early exposure of the participants to both lan-
guages. Although the authors refer to their bilingual sample as late bilinguals, the fact
that learning of L2 took place on average at the age of 9.7 years makes most partici-
pants in this sample early bilinguals, according to Pavlenko’s (2012) criteria. The
authors concluded that emotion words in L1 and L2 could produce the same emo-
tional activation, and suggest that perhaps single words activate emotional aspects of
word meaning to a lesser extent than more complex linguistic stimuli, and as a con-
sequence, differences between L1 and L2 might not necessarily be detected. In sup-

port of these findings, Eilola and Havelka (2011) found similar levels of emotional
Stroop interference in both languages in Greek–English early proficient bilinguals
who differently from the bilingual sample in Eilola et al. (2007) were immersed in
the English-speaking environment (L2) at the moment of testing.
In a similar experiment, Sutton, Altarriba, Gianico, and Basnight-Brown (2007)
found faster RTs for neutral words than for emotion words in bilinguals, faster RTs
in L2-English than L1-Spanish. However, the emotional Stroop effect was present
in both languages, but was greater in L2 than L1, possibly resulting from a shift in
dominance to L2 in their sample. The bilinguals in their study reported using English
(L2) more often than Spanish (L1); however, they reported extremely high profi-
ciency ratings in both languages. Overall, these results indicate that interference
effects, as reflected in the emotional Stroop, can be present in bilinguals in both
languages, and that the effect can be modulated not only by the AoA of L2, but also
by the level of proficiency and dominance. The latter becomes a particularly impor-
tant factor to consider when samples are composed of individuals whose native/L1
is not the dominant language. This is commonly found in children of immigrant
parents born and/or raised from an early age in their L2 environment, and who
received their education in their L2. Nevertheless, in similar experiment with nega-
tive and neutral words, including Thai–English bilingual participants, Winskel
(2013) found significant differences in latencies between the two types of words in
Thai (L1), but not in English (L2). After analyzing a bivariate correlation, the
authors concluded that there was no significant influence of proficiency in the
results. However, their participants did not have immersive experience with their
L2, which, as it relates to the results detailed before, could have been an important
contributing factor, making the sample highly dominant in their native language
while lacking the contribution of naturalistic learning.
Furthermore, Colbeck and Bowers (2012) administered a rapid serial visual pre-
sentation (RSVP) task, which has been previously applied to the study of the auto-
maticity in emotion activation in monolinguals, hypothesizing that level of
proficiency would be the most important factor. The task was to press a key on the
numeric pad upon presentation of the target word (color word) and to ignore all
other words in the set, which included taboo and neutral words. Participants were 20
English monolinguals and 24 Chinese–English bilinguals living in England with
high levels of proficiency in L2. Taboo words generated higher interference and
lower performance on target word identification in English monolinguals, and this
was weaker for the bilingual group (Colbeck & Bowers, 2012). Similarly, Altarriba
and Basnight-Brown (2010) applied the Affective Simon Task (a modification of the
classic Simon task, with presentation of nouns and adjectives with emotional con-
notation) in two experiments, one with emotion-laden words and another with emo-
tion words. In the bilingual group, the level of interference was larger in L2 than in
L1, which the authors attributed to a dominance shift from L1 to L2 (Altarriba &
Basnight-Brown, 2010). These somewhat conflicting results are consistent with the
idea that that connections of emotion words in L2 are influenced by a set of vari-
ables including proficiency, age, context of acquisition, and word frequency.
Overall, the results from the affective Simon task (Altarriba & Basnight-Brown,
2010), and the RSVP task (Colbeck & Bowers, 2012) might indicate that increased
interference in L2 could be related to a higher level of automaticity in this language.
This could be reflective of the differences in proficiency levels and language domi-
nance. Thus, if there is a shift in dominance from L1 to L2, as L2 becomes the domi-
nant language, it can increase automaticity of emotion word processing in L2,
making it very similar to the level of automaticity in L1, and reducing the differ-
ences in processing of emotion words in both languages.
In sum, as reflected in the emotional Stroop and RSVP tasks, reaching a high
level of proficiency in L2 can increase the level of interference of incongruent trials
in that language. With equal levels of proficiency, the level of automaticity in both
languages can become almost equal, decreasing the difference in the Stroop effect
between L1 and L2, hence the similar levels of interference in both languages.
Consistent with findings in different tasks, this suggests a shift in dominance from
L1 to L2. As seen above, tasks requiring high levels of automaticity have not yielded
consistent results that aid in finding an explanation as to why bilinguals feel words
differently in their different languages.
Autonomic Approaches: Skin Conductance Responses (SCRs)
Measures such as skin conductance responses (SCRs), which provide information

about activation of the sympathetic nervous system in response to emotional stim-
uli, have provided some evidence about the differences in which bilinguals experi-
ence emotional content in both languages. However, these measures seem to be
more sensitive to negative and threatening than to neutral or positive stimuli.
Table 3.2 shows a summary of the results of research studies using SCR.
Harris, Ayçiçeği, and Gleason (2003) measured the psychophysiological reac-
tions to taboo words and found that auditory stimuli provoked stronger SCRs in
native language. Taboo words elicited stronger SCRs in both languages among
word categories, stronger in L1 and the auditory channel. Furthermore, Harris
(2004) found that stimuli presented aurally elicited higher SCRs than stimuli pre-
sented visually for early learners but not for late learners in both languages. Taboo
words showed the highest reactivity across word categories. Late learners showed
higher SCRs in L1, but early learners did not show a significant difference in read-
ings. The authors concluded that L1 is more emotional only when it is the most
proficient language (Harris, 2004). In a further study, Caldwell-Harris and
Ayçiçeği-Dinn (2009) found overall autonomic reactivity stronger for items pre-
sented in the L1, with the largest difference observed in the reprimands category
(Caldwell-Harris & Ayçiçeği-Dinn, 2009). Eilola and Havelka (2011) found greater
reactivity to negative and taboo words in L1. In sum, autonomic reactivity seems
to be greater for stimuli presented in L1 than in L2, although it appears to depend
on the type of word, with taboo words provoking the stronger responses. The audi-
tory modality seems to elicit higher reactivity, consistent with the early learning
Table 3.2 Studies using skin conductance response (SCR) paradigm to study emotions in bilinguals
Familiarity
Bilinguals/ Type of Active of words
Reference N Age (SD) F:M language Monolinguals Bilinguals type Findings stimuli bilinguals testeda
Skin conductance responses (SCRs)
Harris 32 28 17:15 Turkish/ None Proficient SCR amplitudes were higher in Turkish Heard/read 5 Yes Yes
et al. English bilinguals; late than in English and amplitudes differed category of
(2003) acquisition of for word emotional type. The largest words
English after difference between languages (and the (neutral,
age 12 only statistically significant pairwise positive,
comparison) occurred for reprimands, aversive,
with reprimands in Turkish eliciting taboo words,
stronger SCRs than reprimands in and
English. This difference held for both reprimands) in
auditory and visual presentation. Taboo Turkish and
words elicited larger SCR in Turkish English while
than in English, but statistical the skin
significance was obtained only when conductance
the analysis was restricted to the response was
auditory modality. Responses to Taboo monitored
words displayed in the visual modality
did not differ between the two
languages
(continued)
Familiarity
Harris 15 early 20.5 Not Spanish/ None Highly 1. Stimuli in the auditory modality Heard/read 5 Yes Yes
(2004) learners specified English proficient; learn elicited higher SCRs than in the visual category of
21 late English in modality, but only for the early learners, words
learners average at age The late learners had comparable (neutral,
of 7.9 (1–16) reactivity to visual and auditory stimuli. positive,
English No other modality effects or aversive,
interactions were significant taboo words,
2. On the early learners a main effect of and
stimulus type was found, this effect was reprimands) in
specific to taboo words. English taboo Spanish and
words differed from the single word English while
conditions as did Spanish taboo words the skin
3. For the late learners, the main effect conductance
of stimulus type was significant. Taboo response was
words elicited elevated SCRs monitored
(compared to the single word
condition) in both English and Spanish.
The reprimands also elicited SCRs that
were greater than the single word
condition, but only in Spanish
4. Averaging over learning group,
Spanish endearments were rated as
more pleasant than English
endearments
5. The only significant difference
occurred for the late learners’ ratings of
reprimands
Reprimands were rated as more
unpleasant in Spanish than in English
Familiarity
Caldwell- 70 20.5 (1.5) 61:9 Turkish/ None Learners of The main finding in the Emotional Listening to 4 No Yes
Harris English English after: Phrases Task was the overall stronger emotional
and age 12 (n = 42) autonomic responding in the first categories:
Ayçiçeği- age 15 (n = 11) language, with the L1–L2 difference endearments,
Dinn and 18 (17). strongest for reprimands. Authors insults,
(2009) Nonproficient concluded that the first language more reprimands,
late bilinguals emotional finding is not simply an and neutral
artifact of nostalgia for the first words I
language for immigrants, and can be Turkish and
extended to bilinguals who are residing English; rating
in an environment where their first sentences for
language is dominant emotional
intensity while
electrodermal
activity was
recorded
Eilola and 31 23.4 (3.1) 22:9 Greek/ 32 (7 males Proficient SCR similar in both Native speakers Skin Yes Yes
Havelka English and 25 bilinguals and nonnative speakers of English; conductance
(2011) females) Learned English however in native English speakers during an
after the age of negative and taboo words produce emotional
6 higher SCR than positive and neutral Stroop
words; this difference was not observed interference
in nonnative English speakers task
a
A yes is included here if in the participants’ rating the unknown meaning category was included
hypothesis proposed by Caldwell-Harris and Ayçiçeği-Dinn (2009), in which the

context and AoA of emotion words would be the primary determinant of future
reactions to such words.
Even though the results of previous studies are inconsistent, there is enough evi-
dence to suggest that it might be possible to find physiological support of differen-
tial processing of emotion words in bilinguals. It has been found for example, that
in late sequential bilinguals, cortical representation of the two languages overlaps
but seems more differentiated than in early simultaneous bilinguals (Abutalebi,
2008), indicating that linguistic processing in L2 might recruit larger cortical areas
than in L1. However, measures such as SCR, which provide information about sym-
pathetic activation, can be more reactive to negative and threatening than to neutral
or positive stimuli, and seem to be more sensitive to differences in the processing of
differently valenced stimuli.
lectrophysiological Approaches: Event-Related Potentials

E
(ERPs)
From all the methodologies applied to investigate how bilingual individuals process
emotional content in both languages, it seems like the one option to obtain evidence
that is more conclusive would be the utilization of brain-based approaches. There
are not many studies in this topic utilizing event-related potentials or neuroimaging
techniques though. Table 3.3 describes the studies using these techniques.
Four ERP studies are available from the literature. The first study to attempt to
find evidence through this method was an unpublished dissertation (Kim, 1993).
The sample consisted of 20 English monolinguals and 40 Korean–English bilin-
guals in two levels: 20 becoming bilinguals and 20 stable bilinguals who performed
a valence decision task, classifying words in English into positive, neutral, or nega-
tive categories. The becoming bilingual group showed greater latencies in P300 and
N200 waves, attributed to interference between both languages causing a process-
ing delay. N200 amplitudes differed significantly between the becoming bilingual
and monolingual groups only, which the authors attributed to increased amount of
resources needed to recognize and process words in the least proficient language.
However, the stimulus set did not include words in Korean to compare with the
words in English. Furthermore, the analysis of emotion word processing in bilin-
guals might require, in addition to comparing reactions in both languages, analyzing
earlier components including electrodes located in occipitoparietal areas, which are
associated with visual processing and attentional mechanisms elicited by emotional
stimuli (Bradley, Hamby, Löw, & Lang, 2007).
In a more recent study, Conrad, Recio, and Jacobs (2011) presented words with
negative, positive, and neutral valences in a lexical decision task to German or
Spanish late bilingual participants who differed in order of acquisition (40 German–
Spanish and 26 Spanish–German late bilinguals). The components of interest were
the early posterior negativity (EPN), which seems to be enhanced by emotional
Table 3.3 Summary of studies that used event-related potentials (ERPs) and functional magnetic resonance (fMRI) to study emotions in bilinguals
Mean age
(Range
Paradigm/ or Gender
reference N SD) F:M Languages Monolinguals Type of bilinguals Findings Stimuli
1. Event-related potentials (ERPs)
Kim (1993) 40 23.1 Not Korean/ 20 (age Stable Korean/ No differences were found in P300 ERPs elicited by
specified English 21.3 years) English bilinguals amplitude as a function of words’ positive,
(n = 20), becoming emotional valence or participants’ negative, and
Korean/English English proficiency neutral emotional
bilinguals (n = 20) English words
Conrad 40 26.20 (20–33) 12−28 German/ None The two L1 groups A highly similar pattern of results was ERP elicited
et al. (2011) (native 28.54 (20–38) 14−12 Spanish are not strictly obtained across L1 and L2 processing: during a visual
Germans comparable in terms event-related potential waves generally lexical decision
speakers) of L2 exposure, reflected a nearly posterior negativity task presenting
26 because all Spanish and a late positive complex for words words with either
(native participants were with positive or negative valence positive, neutral,
Spanish living and being compared to neutral words regardless of or negative
speakers tested in the country the respective test language and its L1 valence
of their L2, whereas or L2 status
German participants Only Spanish native speakers currently
were presently living living in the L2 country showed no
in their native effects for negative as compared to
country—though neutral words presented in L2
most of them had
reported past stays in
Spanish speaking
countries for several
months
(continued)
Mean age
(Range
Paradigm/ or Gender
Opitz and 32 18–28 years old 24:8 German/ None Sequential, EPN was consistently enhanced to Frequently used
Degner French unbalanced proficient emotionally valent (positive and French and
(2012) (16); bilinguals negative) as compared to neutral words, German nouns
French/ replicating previous results of native that varied in
German language research. Valence effects their emotional
(16) showed a bilateral occipitotemporal valance (neutral,
negativity, which is the typical scalp negative, or
distribution of the EPN. However, the positive) and
EPN emotion effect in the present study some pseudo-
was observed slightly later than emotion words were
effects recently reported by others included in the
When reading emotional words in L2 German and the
the EPN was similar in magnitude but French word
delayed compared to L1 word reading lists. The
participant
performed a
lexical
monitoring task
(press a bar when
a pseudo-word
was shown) EEG
was continuously
recorded from 59
electrodes
Chen et al. 17 22.5 (1.8) 8:9 Chinese/ None Proficient, late The effect of emotional valence was While recording
(2015) English unbalanced only significant in L1 with faster an EEG,
Experiment responses to positive compared to participants
1 neutral and negative words. In contrast, performed a
the main effect of emotional valence lexical decision-
was not significant in L2 making task
In the accuracy analyses, the main (judge if the
effect of language was significant, stimulus was a
indicating that accuracy for words in L1 word) of emotion
was higher than that in L2 words—positive,
The main effect of emotional valence negative, neutral,
was also significant Further comparison and pseudo
showed that the accuracy for positive words)
words was higher than for neutral words
and negative words. Furthermore, the
interaction of language and emotional
valence was significant
The ERP results showed that only
positive words in L1 elicited a larger
negative-going waveform on the
posterior sites (EPN) during both 250
and 300 ms and 300 and 350 ms time
windows compared to neutral words
(continued)
Mean age
(Range
Paradigm/ or Gender
2. fMRI
Hsu et al. 20 23.85 (3.6) Not German/ None Proficient, late Results for distinguishing L1 and L2 in Text passages
(2015) specified English unbalanced (age of the emotion-related regions already from Harry
Acquisition of suggested different patterns of Potter books
English (10.6 years) processing emotional material across including
the two languages. A robust modulating passages in the
effect of language was found in the Fear condition,
behavioral data for the evaluation of the Happy
happy and though to a lesser degree, condition, and
fear passages the Neutral
The modulatory effect for processing condition were
happy passages was consistent across read by
behavioral and fMRI data. Happy participants
passages were rated as happier when
read in L1 and fMRI factorial analysis
revealed that the likely neural substrates
involved are left precentral gyrus (the
head/face area on the somatotopy) and
bilateral amygdala
Chen et al. 22 22.4 (1.5) 16:8 Chinese/ None Proficient, late The main effect of emotional valence was fMRI data was
(2015) English unbalanced located in the right superior parietal lobe recorder while
Experiment (BA 7). Further analysis revealed that participants
2 activation in this brain area was greater for performed a
neutral word than for both positive and lexical decision
negative words. The interaction between making task
language and emotional valence was (judge if the
significant in the left cerebellum stimulus was a
Further comparisons showed that the word) of emotion
emotional valence main effect was words—positive,
significant both in L1 and in L2. Pairwise negative, neutral,
comparisons in each language showed and pseudo
that in L1, positive words had weaker words)
activation than negative and marginally
weaker activation than neutral words; in
L2, positive words had stronger activation
than neutral words
In the ROI analyses, the emotional
valence effect was significant in the left
superior frontal gyrus. Pairwise
comparison showed that activation for
negative words was greater than neutral
word. The interaction between language
and emotional valence was significant
in the middle occipital gyrus. Post hoc
comparisons revealed that the emotional
valence main effect was only significant
in L1, showing that positive words had
smaller activation than both neutral and
negative words On the contrary, the
emotional valence effect was not
significant in L2
valence particularly for positive words when compared to neutral words, and seems
to indicate an attention shift toward words with emotional relevance at early pro-
cessing stages; and the late positive complex (LPC), which seems to reflect a higher
level semantic evaluation of words. The authors hypothesized that no modulation of
the EPN would be present in L2, possibly because of a reduction in emotion–cogni-
tion coupling in L2, therefore, requiring more elaborate processing and time. The
emotion sensitivity to L2 words would be reflected in the LPC. If no differences
were found in either component, this would be an indication of similar processing
of emotional content in both languages. Analyses of ERPs for German speakers
reading words in German showed significant effect of valence. The results suggest
that EPN and LPC effects can be elicited in both L1 and L2 in positive words. The
differential effect was present for L1 versus L2 for negative words only in German
native speakers, whereas Spanish native speakers showed no such effect. This could
possibly reflect a positivity bias induced by being immersed in the L2 culture
(Conrad et al., 2011), or, as suggested by Pavlenko (2012), could be indicative of
differences in positivity-negativity biases between the German and Spanish lan-
guages, or could relate to the actual stimuli chosen. L2 effects, as apparent from
EPN and LPC effects, were similar to the effects of L1 processing. Onsets of EPN
emotion effects suggest only general processing delay (Conrad et al., 2011). The
authors interpreted the presence of valence effects in both languages and longer
latencies for both components in L2 as indication of only quantitative differences in
processing emotion words between languages, but not differential processing of
emotion content. However, the two participant groups were both late learners of L2,
differing in whether they were native German or native Spanish speakers. Therefore,
since the two bilingual groups were of similar levels of proficiency in L2, the ques-
tion remains of whether controlling for different levels of proficiency in two other-
wise similar groups of bilinguals would yield different results between languages.
Additionally, the lexical decision task requires analysis of the orthographic proper-
ties of the word rather than a deeper semantic analysis; this analysis taps on r elatively
automatic processing of emotional connotations, which creates the necessity of
exploring emotion effects in a task involving deeper processing levels.
Pursuing this further, Opitz and Degner (2012) tested 16 French–German and 17
German–French bilingual participants living in Germany (AoA of L2; M = 12, rang-
ing from 7 to 12) in a lexical monitoring task, in which participants simply press a
key upon appearance of a nonword. As it relates to valence, no differences in the
processing of negative words were found, as participants in both groups showed
greater EPNs for positive and negative words in both languages, with only a delay in
L2. They attributed these differences to interference and to processing being costlier
in cognitive resources in a second rather than in an L1, which would result in the
flattening of affect for an L2 commonly reported by bilinguals. The EPN was consis-
tently enhanced for both categories of emotion words as compared to neutral words,
showing a bilateral occipitotemporal negativity, typical of the EPN. However, the
authors expected the EPN elicited by emotionally valent words in L2 to be reduced
in amplitude as compared to L1, as well as a delayed peak latency of the EPN rather
than reduction of its amplitude (Opitz & Degner, 2012). Additionally, positive and
negative words elicited a more negative-going deflection at temporooccipital recording

sites as compared to neutral words. The mean of the EPN amplitude differed between
neutral words and both positive and negative words, but not between positive and
negative words. Significant differences in latency for all word categories were
observed for all participants in their L2 for emotion words as compared to neutral
words. The authors suggest that the delayed latency in L2 corresponds to similar
attentive processes, but the conceptual identification of the words in this language
takes longer. While participants were reading emotion words in L2, the EPN was
similar in magnitude but delayed compared to L1 emotion word reading. The authors
concluded that lexical access to emotional words in highly proficient L2 users is
delayed due to higher interference between L1 and L2 lexical representations; there-
fore, because processing is costlier in L2, the perception of emotional valence in
words is attenuated (Opitz & Degner, 2012). The longer latency in the EPN in this
study is consistent with the findings of Conrad et al. (2011); both studies used bilin-
gual groups with similar characteristics. For example, in both studies, the bilingual
groups consisted of late bilinguals differing only in the order of acquisition of their
respective languages. Additionally, proficiency was controlled only to ensure that
both groups were comparable in their proficiency levels in L2. Finally, both studies
utilized a lexical decision or monitoring task, which only require an orthographic
decision and leave out the possibility of analyzing the effects of valence processing.
Chen, Lin, Chen, Lu, and Guo (2015) had 24 Chinese–English bilinguals who had
never visited an English-speaking country perform a lexical decision task. Overall,
all conditions evoked similar early ERP components (P1 and N2). In L1, emotion
words elicited larger negative deflections than neutral words in the 250-400 ms win-
dow at posterior electrode sites, resembling the classical EPN. Additionally, in L1,
emotion words generated smaller positive waves than neutral words starting at 500-
800 ms window at centroparietal electrodes, consistent with the LPC. In L2, neutral
words presented lower positivity than emotion words at the parietal sites during the
400-500 ms window (Chen et al., 2015). However, valence effects in L2 were signifi-
cantly delayed with respect to previous studies and resembled the scalp distribution
of the N400 component rather than the LPC. The participants in this sample were all
late unbalanced bilinguals, with significantly lower levels of proficiency in L2 and
who had never been immersed in the L2 environment, suggesting that differences in
component amplitude for this sample might originate in actual differences between
the processing in both languages, as could be concluded from the detection of a com-
ponent associated with semantic processing such as the N400. Moreover, since the
effects of valence and language were analyzed only within one group of participants,
there was no comparison between participants with different linguistic profiles, such
as different levels of proficiency.
Overall, ERP studies have suggested differences in the processing of emotion
words in each language in bilinguals. Conrad et al. (2011) and Opitz and Degner
(2012) suggest quantitative differences, as reflected in delay latencies in L2 com-
pared to L1. On the other hand, Chen et al. (2015) suggest that this difference could
be also qualitative, as indicated by the differences in amplitude in the ERP components
in L2 compared to L1.
euroimaging Approaches: Functional Magnetic Resonance

N
Imaging (fMRI)
Thus far, and to the authors’ knowledge, only two studies have looked into the neu-
ral bases of differential emotion content processing in bilinguals utilizing fMRI. Hsu,
Jacobs and Conrad (2015) observed the neural substrates associated with reading
literary passages with emotional content in the L1 or the L2 in 20 late German–
English bilinguals. While in the magnetic resonance (MR) scanner, participants had
to read short text passages from Harry Potter books characterized by a negative
(fear-inducing passages) or positive (happiness-inducing passages) versus neutral
emotional valence manipulation. Following the experiment in the MR scanner, par-
ticipants rated all 120 passages in the language version they had read inside the
scanner. Results showed that bilateral visual cortices were more active when read-
ing in L1 (German) than in L2 (English), irrespective of emotion conditions. This
result was attributed to differences between orthographic systems in the two lan-
guages. The authors found that the amygdala and the hippocampal and parahippo-
campal cortex were associated with the processing of emotional stimuli; other areas
activated were the temporal lobe and the temporoparietal junction. In contrasting
these activations in L1 and L2 the authors found the activation in the amygdala to be
significant in the contrast [L1 > L2], while the activation of the anterior insula was
significant in the contrast [L2 > L1]. Happy passages were rated as happier when
read in L1 and fMRI factorial analysis revealed that the likely neural substrates
involved are left post-central gyrus (the head/face area on the somatotopy) and bilat-
eral amygdala. The overall neural responses to emotion content were more differen-
tiated in L1 than in L2. The authors concluded that reading literature in one’s L1
could induce a more intense emotional experience than reading it in one’s L2.
Furthermore, Chen et al. (2015) recorded fMRI during a lexical decision task in
22 late Chinese–English bilinguals. During this experiment, the mid-occipital gyrus
showed decreased activity in response to positive words when compared to negative
and neutral in L1, but it was inactive in response to emotional content in L2. An
unexpected finding was a differential activation of the left cerebellum, where posi-
tive words showed decreased activation when compared to neutral words in L1, but
increased activation in L2.
These two studies suggest some level of dissociation in brain activation in bilin-
guals’ perception of L1 and L2 emotion words; however, more research is needed to
obtain findings that are more conclusive.
Conclusions and Future Research
The relationship between emotion processing and bilingualism is complex. Despite

the fact that bilinguals report differences in affect between their two languages and
consistently describe their L1 as more emotional, experimental research results are
inconsistent. Bilingual individuals are often assumed most proficient in their first
language, even though this is not necessarily true. Dominance and proficiency can
change over time depending on the individual history of bilingualism. The high
variability among the linguistic histories of bilinguals and factors influencing L2
learning is unquestionable. Additionally, learning processes are continuous, occur-
ring at all times in all contexts in which the learner is immersed, thus increasing
variability within the same individual across time. Therefore, the conditions of use
and acquisition of an L2 are difficult to quantify. Hence, the answer to the question
of whether bilinguals experience emotions differently in both languages and if these
differences can be found reflected in either behavioral or physiological measures,
might depend on factors that are highly variable. Therefore, conclusive evidence
might be found only by controlling for inter- and intraindividual variability.
Evaluative methods in the studies reviewed here consist of tasks including emo-
tion words in both languages. Such tasks include, but are not limited to recall tasks
(Anooshian & Hertel, 1994; Ayçiçeği & Harris, 2004; Ayçiçegi-Dinn & Caldwell-
Harris, 2009; Ferré et al., 2010), emotional Stroop tasks (Eilola et al., 2007; Sutton
et al., 2007), and interference tasks (Colbeck & Bowers, 2012).
The most consistent finding in this review is that in early—simultaneous—pro-
ficient bilinguals, the processing of emotion words is identical in the two lan-
guages, whereas emotional dissociations are more likely in late sequential
nonproficient bilinguals. It has been suggested that the emotional advantage for
the native language found in sequential late bilinguals is explained by a delay in
cognition–emotion coupling in L2 (Anooshian & Hertel, 1994; Conrad, et al.,
2011). Moreover, it has been found that the processing of emotional information
provided in an L2 differs not only qualitatively but also quantitatively when com-
pared with that in the native language. Pavlenko (2012) proposes that these pro-
cessing differences would be most differentiated between early and late bilinguals;
whereas late bilinguals would process emotional valence in L2 at a semantic level,
early bilinguals would present more embodied representations of L2, as reflected
in instances of skin conductance reactivity such as in Harris (2004). However, the
question of which conditions favor emotion–cognition coupling in bilinguals’ L2
and which variables contribute the most still remains unanswered. While some
investigators propose AoA as the predominant factor (Anooshian & Hertel, 1994),
others postulate frequency of use (Dewaele, 2004b) and even the context in which
each language has been acquired (Dewaele, 2010). Other proposed mediating fac-
tors in the emotional dissociation of L1 and L2 are the types of experimental tasks
(Ayçiçegi-Dinn & Caldwell-Harris, 2009), language proficiency (Eilola et al.,
2007; Harris, 2004), and language dominance (Altarriba & Basnight-Brown,
2010; Sutton et al., 2007). Additionally, the matter of the separation of the two
languages in the communication of emotional content does not seem limited to
the productive aspect of language. Consistently, comprehension of emotion con-
cepts in L2 would require not just a high level of linguistic competence but also
the understanding of the corresponding cultural background. For instance, bilin-
guals may encounter problems in understanding humor in their L2 since its inter-
pretation requires knowledge of semantic and culture-specific word associations
in order to be able to decode the meaning (Vaid, 2000).
Future research should also look at the implication of the distance between
languages in the emotional effect; for example, this effect should not be consid-
ered equivalent between Spanish–Catalan bilinguals compared to Turkish–English
bilinguals since Catalan and Spanish are closer linguistically compared to
Turkish–English. Moreover, differences in orthographic systems should be con-
sidered since quite often the participants’ reading and writing skills are included
in the experimental paradigms of emotions. Also, it will be interesting to look at
the strategies used by bilinguals in the recalling process. For example when recall-
ing the words from L2, do bilinguals use translation from L1? Finally, the emo-
tional stimuli used in the detection of the emotional advantage need better
definition and further exploration. Recent research has shown that there are differ-
ences in the processing of emotion words compared to emotion-laden words
among bilingual participants. For example, Kazanas and Altarriba (2015) found
that Spanish–English bilinguals reacted faster to emotion-laden words, but only in
English. Not all potentially emotion-inducing stimuli can be generalized across
languages; what is and what is not an emotion word depends on a complex inter-
play of informational, contextual, cultural, and individual factors (Brosch,
Pourtois, & Sander, 2010).
Nonetheless, if emotional processing differences indeed exist between L1 and
L2, the implications for bilingual populations, especially those living in a host
country, could be of great importance, not only for maintaining successful trans-
cultural relationships but also to successfully adapt to a host culture. This would
raise the need to reevaluate didactic and clinical settings or anywhere bilinguals are
evaluated by monolingual speakers through testing materials in their L2, which
could potentially yield biased results and possibly lead to unsuitable therapeutic
approaches.
Acknowledgement Our most sincere gratitude to Deven Christopher for her editorial support and
valuable suggestions.
References
Abutalebi, J. (2008). Neural aspects of second language representation and language control. Acta
Psychologica, 128(3), 466–478.
Altarriba, J. (2008). Expressions of emotions as mediated by context. Bilingualism: Language and
Cognition, 11(2), 165–167. doi:10.1017/S1366728908003295
Altarriba, J., & Basnight-Brown, D. M. (2010). The representation of emotion vs. emotion-laden
words in English and Spanish in the affective Simon task. International Journal of Bilingualism.
doi:10.1177/1367006910379261
Amati-Mehler, J., Argentieri, S., & Canestri, J. (1993). The Babel of the unconscious: Mother
tongue and foreign tongues in the analytic dimension. Madison, CT: International Universities
Press.
Anooshian, L. J., & Hertel, P. T. (1994). Emotionality in free recall: Language specificity in bilin-
gual memory. Cognition and Emotion, 8(6), 503–514. doi:10.1080/02699939408408956
Ardila, A. (2007). Bilingualism in the contemporary world. In A. Ardila & E. Ramos (Eds.),
Speech and language disorders in bilinguals (pp. 1–20). New York: Nova Science.
Ayçiçeği, A., & Harris, C. L. (2004). Brief report: Bilinguals’ recall and recognition of emotion
words. Cognition and Emotion, 18(7), 977–987. doi:10.1080/02699930341000301
Ayçiçegi-Dinn, A., & Caldwell-Harris, C. (2009). Emotion-memory effects in bilingual speak-
ers: A levels-of-processing approach. Bilingualism: Language and Cognition, 12(3), 291–303.
doi:10.1017/S1366728909990125
Bond, M. H., & Lai, T. (1986). Embarrassment and code-switching into a second language. The
Journal of Social Psychology, 126(2), 179–186.
Bradley, M. M., Hamby, S., Löw, A., & Lang, P. J. (2007). Brain potentials in perception: Picture
complexity and emotional arousal. Psychophysiology, 44(3), 364–373.
Bradley, M. M., & Lang, P. J. (1999). The International affective digitized sounds (IADS): Stimuli,
instruction manual and affective ratings. Gainesville, FL: NIMH Center for the Study of
Emotion and Attention.
Bresnahan, M. J., Ohashi, R., Nebashi, R., Liu, W. Y., & Shearman, S. M. (2002). Attitudinal and
affective response toward accented English. Language & Communication, 22(2), 171–185.
Brosch, T., Pourtois, G., & Sander, D. (2010). The perception and categorization of emotional
stimuli: A review. Cognition and Emotion, 24, 377–400.
Caldwell-Harris, C., & Ayçiçeği-Dinn, A. (2009). Emotion and lying in a non-native language.
International Journal of Psychophysiology, 71(3), 193–204. doi:10.1016/j.ijpsycho.2008.09.006
Chen, P., Lin, J., Chen, B., Lu, C., & Guo, T. (2015). Processing emotional words in two languages
with one brain: ERP and fMRI evidence from Chinese–English bilinguals. Cortex, 71, 34–48.
Colbeck, K. L., & Bowers, J. S. (2012). Blinded by taboo words in L1 but not L2. Emotion, 12(2),
217–222. doi:10.1037/a0026387
Conrad, M., Recio, G., & Jacobs, A. M. (2011). The time course of emotion effects in first and
second language processing: A cross-cultural ERP study with German-Spanish bilinguals.
Frontiers in Psychology, 2, 351. doi:10.3389/fpsyg.2011.00351
Costa, A., Foucart, A., Hayakawa, S., Aparici, M., Apesteguia, J., Heafner, J., et al. (2014). Your
morals depend on language. PloS One, 9(4), e94842.
De Leersnyder, J., Mesquita, B., & Kim, H. S. (2011). Where do my emotions belong? A
study of immigrants’ emotional acculturation. Personality and Social Psychology Bulletin,
0146167211399103.
Dewaele, J. M. (2004a). The emotional force of swearwords and taboo words in the speech of mul-
tilinguals. Journal of Multilingual and Multicultural Development, 25(2-3), 204–222.
Dewaele, J. M. (2004b). Blistering barnacles! What language do multilinguals swear in? Estudios
de Sociolinguıstica, 5, 83–105.
Dewaele, J. M. (2008). The emotional weight of “I love you” in multilinguals’ languages. Journal
of Pragmatics, 40(10), 1753–1780.
Dewaele, J. M. (2010). Emotions in multiple languages. London: Palgrave Macmillan.
Dewaele, J. M., & Costa, B. (2013). Multilingual clients’ experience of psychotherapy. Language
and Psychoanalysis, 2(2), 31–50.
Dewaele, J. M., & Pavlenko, A. (2002). Emotion vocabulary in interlanguage. Language Learning,
52(2), 263–322.
Eilola, T. M., & Havelka, J. (2011). Behavioural and physiological responses to the emotional and
taboo Stroop tasks in native and non-native speakers of English. International Journal of
Bilingualism, 15(3), 353–369. doi:10.1177/1367006910379263.
Eilola, T. M., Havelka, J., & Sharma, D. (2007). Emotional activation in the first and second lan-
guage. Cognition and Emotion, 21(5), 1064–1076. doi:10.1080/02699930601054109
Ferré, P., García, T., Fraga, I., Sánchez-Casas, R., & Molero, M. (2010). Memory for emotional
words in bilinguals: Do words have the same emotional intensity in the first and in the second
language? Cognition and Emotion, 24(5), 760–785.
Harris, C. L. (2004). Bilingual speakers in the lab: Psychophysiological measures of emotional
reactivity. Journal of Multilingual and Multicultural Development, 25(2-3), 223–247.
Harris, C. L., Aycicegi, A., & Gleason, J. B. (2003). Taboo words and reprimands elicit greater
autonomic reactivity in a first language than in a second language. Applied PsychoLinguistics,
24(04), 561. doi:10.1017/S0142716403000286
Hsu, C. T., Jacobs, A. M., & Conrad, M. (2015). Can Harry Potter still put a spell on us in a second
language? An fMRI study on reading emotion-laden literature in late bilinguals. Cortex, 63,
282–295.
Kazanas, S. A., & Altarriba, J. (2015). Emotion word processing: Effects of word type and valence
in Spanish-English bilinguals. Journal of Psycholinguistic Research, 45(2), 395–406.
Kensinger, E. A., & Corkin, S. (2004). Two routes to emotional memory: Distinct neural processes
for valence and arousal. Proceedings of the National Academy of Sciences of the United States
of America, 101(9), 3310–3315.
Kim, M. (1993). The effects of emotional connotations of English words on event-related potentials
(ERPs) in Korean-English bilinguals and English-speaking monolinguals. (Ph.D., University
of Georgia). ProQuest Dissertations and Theses, Retrieved from http://search.proquest.com/
docview/304058599?accountid=10902. (304058599).
Kuperman, V., Estes, Z., Brysbaert, M., & Warriner, A. B. (2014). Emotion and language: valence
and arousal affect word recognition. Journal of Experimental Psychology General, 143(3),
1065–1081. doi:10.1037/a0035669
LaBar, K. S., & Cabeza, R. (2006). Cognitive neuroscience of emotional memory. Nature Reviews:
Neuroscience, 7(1), 54–64.
MacKay, D. G., & Ahmetzanov, M. V. (2005). Emotion, memory, and attention in the taboo stroop
paradigm an experimental analogue of flashbulb memories. Psychological Science, 16(1),
25–32.
Opitz, B., & Degner, J. (2012). Emotionality in a second language: It’s a matter of time.
Neuropsychologia, 50(8), 1961–1967. doi:10.1016/j.neuropsychologia.2012.04.021
Pavlenko, A. (2012). Affective processing in bilingual speakers: Disembodied cognition?
International Journal of Psychology, 47(6), 405–428. doi:10.1080/00207594.2012.743665
Ponari, M., Rodríguez-Cuadrado, S., Vinson, D., Fox, N., Costa, A., & Vigliocco, G. (2015).
Processing advantage for emotional words in bilingual speakers. Emotion, 15(5), 644.
Real Academia de la Lengua. (2001). Diccionario de la lengua española (22nd ed.). Madrid: Real
Academia Española.
Reber, R., Perrig, W. J., Flammer, A., & Walther, D. (1994). Levels of processing and memory for
emotional words. Swiss Journal of Psychology, 53(2), 78–85.
Redondo, J., Fraga, I., Padrón, I., & Comesaña, M. (2007). The Spanish adaptation of ANEW
(affective norms for English words). Behavior Research Methods, 39(3), 600–605. doi:10.3758/
BF03193031
Sutton, T. M., Altarriba, J., Gianico, J. L., & Basnight-Brown, D. (2007). The automatic access
of emotion: Emotional Stroop effects in Spanish-English bilingual speakers. Cognition and
Emotion, 21(5), 1077–1090. doi:10.1080/02699930601054133
Vaid, J. (2000). New approaches to conceptual representations in bilingual memory: The case for
studying humor interpretation. Bilingualism: Language and Cognition, 3(01), 28–30.
Velez-Uribe, I., & Rosselli, M. (in press). The auditory and visual interpretation of emotionally
loaded words in Spanish/English bilinguals. Bilingualism: Language and Cognition.
Winskel, H. (2013). The emotional Stroop task and emotionality rating of negative and neutral
words in late Thai–English bilinguals. International Journal of Psychology, 48(6), 1090–1098.
Part II
Bilingual Language Representation
Chapter 4
Bilingual Figurative Language Processing
Anna B. Cieślicka
Contents
I ntroduction: The Importance of Figurative Language 75
Theoretical Frameworks 79
Figurative Processing in L2 and Bilingual Language Users 85
Summary and Conclusions 106
References 109
Introduction: The Importance of Figurative Language
The prevalence of figurative language is a well-established fact, quoted repeatedly in

each and every research report concerned with the topic. Estimates have been pro-
vided for the number of figurative expressions produced per minute of speech (Pollio,
Barlow, Fine, & Pollio, 1977), per week (Hoffman, 1980), or for the number of fro-
zen and novel metaphorical expressions uttered within a 60-year life span (Glucksberg,
1989). Others report the percentage of formulaic expressions, such as idioms, collo-
cations, or speech formulas in databases of spoken (Erman & Warren, 2000) or writ-
ten (Howarth, 1998) discourse, to name just a few. More recently, based on the data
from naturally occurring discourse, Van Lancker Sidtis, Cameron, Bridges, and
Sidtis (2015) estimated that formulaic expressions make up around 25% of spontane-
ous speech in healthy populations, regardless of the style or topic of conversation.
Given the widespread occurrence of figurative language in spoken and written
discourse, many scholars view the study of figurative comprehension as a necessary
prerequisite to achieving a full understanding of how the human mind processes
A.B. Cieślicka (*)

Department of Psychology and Communication, Texas A&M International University,
Laredo, TX, USA
e-mail: anna.cieslicka@tamiu.edu

76 A.B. Cieślicka
language in general (cf. Colombo, 1993; Johnson-Laird, 1993). Going even further,
researchers working in the cognitive linguistics framework and the Conceptual
Metaphor Theory (Gibbs, 1994, 1996, 1998; Lakoff, 1987, 1990, 1993; Lakoff &
Johnson, 1980; Lakoff & Turner, 1989) have placed figurative language at the center
of human cognition and argued that essentially all language is based on metaphors
that structure our conceptual system. The Conceptual Metaphor Theory, as well as
its most recent extension, the Neural Theory of Metaphor (Lakoff, 2009) strongly
emphasize the centrality of metaphor in everyday language use suggesting that
many expressions perceived as literal are, in fact, deeply rooted in metaphoric
thought. Rather than viewing metaphorical expressions merely as a verbal phenom-
enon, Lakoff and Johnson (1980) argued that verbal metaphors elaborate conven-
tional metaphorical patterns of human thought in which a certain conceptual domain,
or a set of concepts, is understood in terms of another domain, or another set of
concepts. For example, in the conceptual metaphor life is a journey, mappings
between the physical (source) domain of motion and the abstract (target) domain of
time are developed, conceptualizing the passage of time as a motion along the path
and helping us understand an abstract concept of the time of our biological existence
in terms of a more concrete concept of a journey. These ontological correspon-
dences, or conceptual mappings, reflect embodied experience in the sense they
reflect experiences of our somatosensory, perceptual, etc., systems and structure the
way we think about the world. Each metaphor establishes a mapping of systematic
correspondences between the source and target domains. In life is a journey
example above, people’s experiential knowledge about journeys allows drawing fur-
ther metaphorical inferences, such as the fact that, like journeys, life has a purpose
and a destination. Those metaphorical conceptualizations are broadly universal, as
documented by extensive cross-linguistic and cross-cultural research into concep-
tual metaphors (Kövecses et al., 2015) and might be stored as fixed neural circuits in
the brain that get activated automatically when processing conventional metaphors
(Lakoff, 2009). Overall, the idea of embodied cognition explains abstract concepts,
such as emotions, thought, and causality as semantically grounded in basic bodily
and sensory experiences, such as heat, space, and motion, thus placing figurative
language and thought at the center of human cognitive operations (see also Katz &
Bowes, 2016, for the embodied cognition perspective to bilingualism).
Because of the importance of figurative language for understanding how the
mind works, various tropes of figurative language such as metaphor, idioms, prov-
erbs, irony, sarcasm, and metonymy have long been the major focus of psycholin-
guistic research. The main research questions asked in the figurative language
literature have been motivated by the fact that figurative expressions pose a chal-
lenge for the language processing mechanism in that their intended meaning is dif-
ferent than their literal meaning. As an example, consider the somewhat novel
metaphor My new rescue cat is a cactus. While it is obvious that the intended figura-
tive meaning cannot be derived from the compositional word-by-word analysis of
the phrase, as a feline cannot possibly be a succulent perennial, literal meaning does
contribute in some way to the process of comprehending this metaphor, as the con-
cept of cactus, called the vehicle for the metaphor, is applied to the characterization
4 Bilingual Figurative Language Processing 77
of the cat (the topic or tenor of the metaphor). Accordingly, only some of the attributes
of the concept of cactus such as having prickles, rather than growing primarily in
arid regions or bearing edible fruits, will be considered in relation to the topic cat.
This incompatibility between the literal meanings of the topic and vehicle creates a
rhetorical tension between them and constitutes the ground or inferred interpreta-
tion of the metaphor. Even assuming that we restrict the range of semantic features
of the vehicle that can reasonably be attributed to the topic, the metaphor still allows
a wide range of possible interpretations. Is the cat prickly because it gets easily
irritated and acts aggressive towards other cats in the household? Or, maybe, does
being prickly refer to the deplorable state of its fur which is clumped and tangled in
knots and make the cat look as if he indeed had cactus-like spines?
Idioms, or fixed phrases belonging to the vast category of formulaic sequences
appear even more problematic for the language processing mechanism. Whereas
some idioms cannot be analyzed on a word-by-word basis, as they are either seman-
tically anomalous (e.g., cut the mustard, be on cloud nine, rack your brains) or
syntactically ill-formed (e.g., through thick and thin, jump the gun, cut to the chase),
others do have a plausible literal meaning (e.g., break a leg, hit the nail on the head,
bite off more than you can chew), which makes them ambiguous and adds additional
challenge to any linguistic theory attempting to account for their comprehension. To
make matters more complicated, not all literally plausible idioms are identical, as
some are more likely to be interpreted literally (i.e., high-literalness idioms such as
have your hands full) than others (i.e., low-literalness idioms such as a thousand
dollars is chicken feed), and this difference crucially affects how they are processed
(Cronk & Schweigert, 1992; Cronk, Lima, & Schweigert, 1993).
In addition to the degree of their literalness, ambiguous idioms differ along the
dimension of semantic decomposition (Nunberg, 1978) or analyzability (Gibbs, 1994;
Gibbs & Nayak, 1989; Gibbs, Nayak, & Cutting, 1989; Glucksberg, 1991), that is, the
extent to which literal meanings of idiom constituent words contribute to the phrase’s
figurative interpretation. For instance, while in the decomposable idiom break the ice
there is a one-to-one correspondence between literal meanings of idiom component
words and their figurative meanings, in that the word break corresponds to the idiom-
atic sense of changing a tense atmosphere and the word ice relates figuratively to
social tension, a literal analysis of the nondecomposable idiom chew the fat will not
lead to inferring its figurative interpretation talk without purpose. It seems then that
the most economical strategy for the language processing mechanism when dealing
with idiomatic sequences would be to suppress or otherwise ignore altogether the
incompatible literal meanings while processing nondecomposable idioms (e.g., chew
the fat, a piece of cake, or a chip on your shoulder) and to hold literal meanings active
while processing semantically decomposable ones (e.g., keep your fingers on the
pulse, save your skin, or play with fire). How does the language processing mecha-
nism know which idiom is semantically decomposable and which is nondecompos-
able and when to either suppress or hold literal meanings of idiom constituents active,
depending on their contribution to the idiom’s figurative meaning? Which of the
meanings, literal or figurative, becomes available first? Is the activation of literal
meaning an obligatory step in arriving at a figurative interpretation or can literal anal-
78 A.B. Cieślicka
ysis be skipped altogether, especially in idioms where word-by-word linguistic

processing does not contribute to the phrase’s overall meaning? Those questions have
motivated a lot of research into the processing of idiomatic expressions.
If metaphors and idioms seem challenging enough for the language comprehen-
sion system, in that they entail resolving lexical- and phrase level ambiguity in order
to arrive at the appropriate interpretation, other figurative tropes such as metonymy,
irony, or sarcasm call for even more intricate processing strategies as they rely on
contextual and pragmatic information to be successfully understood. For example,
in understanding irony a speaker’s belief and attitude have to be inferred by consid-
ering the implied figurative meaning, which is different, or even completely oppo-
site, to the utterance’s literal meaning. Thus, the phrase I love it! cannot not be
understood as intended (expressing frustration rather than positive enthusiasm)
unless presented in a larger discourse context such as, Ania is not a morning person
as she stays up late working on her research, so when Maria asks her how she copes
with her cat scratching at the bedroom window at 7 a.m. every day to be let back in
after his nocturnal adventures, she replies, “I love it!” To be properly understood,
ironic statements like the one above require not only linguistic and semantic analy-
sis on a word-by-word level but, more importantly, an in-depth pragmatic analysis
based on extra-linguistic information such as the situational context in which the
ironic statement is uttered and knowledge of the speaker’s communicative style
(such as whether the speaker typically engages in ironic comments or not).
Similarly, comprehending metonymy poses a challenge for the language pro-
cessing system, as it involves both conceptual and linguistic processes whereby a
word denoting one concept shifts in meaning to represent a related concept, as in
The buses are on strike, where the concept buses stands for the concept bus drivers
(Kövecses & Radden, 1998, p. 42) or as in The ham sandwich is sitting at table 20,
where ham sandwich refers to the customer who ordered this particular dish
(Nunberg, 1978). Broadly, metonymies express a stands for relation; for example, a
producer stands for the product (Shakespeare referring to the book he wrote, as in
Shakespeare is on the top shelf), object used stands for its user, as in the bus strike
example earlier, place stands for the event that happened there (Vietnam referring to
the war in Vietnam, as in A lot of Americans protested during Vietnam), place stands
for the institution located there (White House referring to the government, as in The
White House blocks CNN from press briefing), or part of a thing stands for the whole
thing (the hungry mouth referring to the hungry person, as in Now they have another
hungry mouth to feed) (Frisson & Pickering, 1999; Kóvesces & Radden, Kövecses
& Radden, 1998; see also Lowder & Gordon, 2013). While the precise relationship
between literal and metonymic meanings of a word varies in different types of
metonymy, what they all have in common is that the word’s metonymic and literal
meanings are always somehow related. For more familiar metonymies whose com-
prehension requires activating the metonymic meaning that is already stored in the
mental lexicon, the challenge for the language processing mechanism is hence to
select between the available literal and metonymic senses. In this regard, the com-
prehension of familiar metonymies resembles that of idioms where both literal and
figurative meanings are available to choose from. On the other hand, unfamiliar
metonymies require the language user to actively create the metonymic meaning
based on the literal meaning available in the mental lexicon, and here the level of
challenge resembles that involved in understanding novel metaphors, whose figura-
tive interpretation needs to be inferred, rather than retrieved.
In short, a brief overview of the different figurative tropes and nuances involved
in their understanding clearly shows the many challenges they pose for the language
processor. Some of the questions that arise are: How and when do figurative meaning
gets activated and literal meanings suppressed in the course of language comprehen-
sion? Is the activation of literal meanings obligatory, even in cases when literal anal-
ysis does not contribute to the intended figurative meaning, as in nondecomposable
idioms? What is the role of context in the activation of one meaning and promoting
suppression of another? These questions have guided a lot of research into the mech-
anisms of figurative language comprehension and resulted in the development of a
number of theoretical approaches. Broadly speaking, four major accounts have been
developed to explain how the language processing mechanism copes with figurative
language. They have motivated recent decades of psycholinguistic research into
figurative processing and have substantially influenced more recent theories and
models developed for second language and bilingual language users. These theoreti-
cal frameworks are reviewed in the next section.
Theoretical Frameworks
The Literal-First Model
This classical approach to explaining how figurative language is understood, also

known as the standard pragmatic model (Grice, 1975; Searle, 1975, 1979) echoes
Aristotelian views expressed in Poetics, under which metaphor should be viewed as
deviant from literal language. The standard pragmatic view has been labeled the
three-stage model (Clark & Lucy, 1975; Janus & Bever, 1985) to reflect the three
stages of nonliteral language comprehension it assumes. Accordingly, in stage (1),
the listener computes the literal meaning of the utterance; in stage (2), s/he com-
pares the literal meaning against the context of the utterance to decide if the literal
meaning could also be the intended meaning; finally, in stage (3), upon determining
that the literal meaning makes no sense, s/he computes the figurative meaning.
Because of the multiple stages and the serial nature of this process, the literal-first
model has also been referred to as the multistage model or the serial model (cf.
Temple & Honeck, 1999). A number of processing implications emerge from the
literal-first model (see Glucksberg, 1989; Glucksberg, Gildea, & Bookin, 1982).
First of all, literal and figurative comprehension processes are essentially different.
Second, literal analysis is obligatory whereas figurative optional, as it is carried out
only when the literal interpretation fails. The failure of literal interpretation is thus
a necessary trigger for figurative processing to be initiated. Finally, given the serial
nature of the process, figurative processing is assumed to take place later during the
comprehension process, thus appearing to take longer than literal processing.
80 A.B. Cieślicka
The literal-first view has influenced much of the idiom processing research and
served as a framework for the idiom list hypothesis (Bobrow & Bell, 1973), which
assumes that literal meanings of idiomatic expressions are processed faster than
figurative ones. Idioms are stored in a special list of idiomatic expressions, or an
idiom lexicon, which is separate from the main word lexicon. This idiom lexicon is
accessed through a special idiom mode of processing which is instantiated when
literal analysis of an idiom fails to provide a meaningful interpretation.
While some support has been found for the literal-first model (e.g., Bobrow &
Bell, 1973; Brannon, 1975; Clark & Lucy, 1975; Janus & Bever, 1985), many of its
assumptions have been challenged in subsequent research investigating the process-
ing of figurative language (e.g., Bock & Brewer, 1980; Gibbs, 1981, 1983, 1994;
Gildea & Glucksberg, 1983; Keysar, 1989; Ortony, Schallert, Reynolds, & Antos,
1978; Pollio, Fabrizi, Sills, & Smith, 1984; Shinjo & Myers, 1987). The model has
been criticized as too simplistic and failing to distinguish between the various figu-
rative tropes, each of which might call for slightly different processing strategies
(Morgan, 1979). For example, understanding a fixed multi-word phrase very likely
evokes a different processing mechanism than interpreting a novel poetic metaphor
(Levin, 1979). In addition, the model has been criticized for failing to take into
account the role of context which might aid in the comprehension of figurative
meaning and bypass the need for literal activation (Gibbs, 1994; Gerrig & Healy,
1983). However, the real theoretical issue is not whether context influences the com-
prehension of nonliteral language (we know it does!), but where and when during
the comprehension process. Interestingly, patterns of literal and figurative activation
obtained in recent electrophysiological studies in metaphor (e.g., Lai, Curran, &
Menn, 2009) and metaphor and metonymy (Weiland, Bambini, & Schumacher,
2014) have been interpreted as supportive of the standard pragmatic model (but see
the parallel-processing account).
The Figurative-First Model
The figurative-first approach is the most radical departure from the standard prag-
matic view, as it emphasizes the primacy of figurative meaning, such that activating
the utterance’s literal meaning is not at all needed to retrieve the intended figurative
interpretation (Gibbs, 1983, 1985, 1994). Support for the model was mainly found
in studies which showed better recall (e.g., Bock & Brewer, 1980) or faster process-
ing times for figurative than for literal meanings of nonliteral expressions. For
example, Gibbs (1986) found that people processed sarcastic expressions You are a
fine friend (meaning You are a bad friend) faster than they did either literal uses of
the same expressions or nonsarcastic remarks. Similarly, Kemper (1981) found
shorter reading latencies for proverbs used figuratively than for proverbs used liter-
ally (cf. Temple & Honeck, 1999).
An example of the figurative-first model developed to account for the processing
of idioms is the direct access model (Gibbs, 1980, 1985). Also known as the
figurative first hypothesis (cf. Cronk & Schweigert, 1992) or idiomatic processing
model (cf. Schweigert & Moates, 1988), it suggests that the figurative meaning of
an idiom is accessed directly and that literal meanings of idiom component words
do not need to be analyzed at all. This is especially true of highly conventional idi-
oms whose figurative meanings are quickly accessible from the mental lexicon. The
view that figurative language is understood effortlessly and directly is also reflected
in the Conceptual Metaphor approach mentioned earlier, which assumes that most
figurative expressions are instantiations of preexisting metaphorical mappings
between conceptual domains. Since our conceptual system is structured via such
metaphorical mappings, we understand figurative language as effortlessly and auto-
matically as we do literal language (Gibbs, 1992, 1994, 1996, 1998, 2001). The
metaphorical expression Our marriage is a rollercoaster is understood by automati-
cally activating the underlying conceptual metaphor love is journey where the
abstract domain of experience (love) is mapped onto the more concrete domain of
experience (that of journeys). The figurative-first approach has been subsequently
modified so as to allow for some degree of literal activation to occur prior to, or
simultaneous with the computation of figurative meaning (Gibbs, 2002), which ren-
ders it more compatible with the third major approach, the parallel-processing
model, to be discussed next.
The Parallel-Processing Model
The parallel-processing model states that both literal and nonliteral meanings can
be processed simultaneously and that there are no fundamental differences between
the comprehension of literal and figurative language. Giora and Fein (1999a, 1999b,
1999c) refer to this view as the equivalence hypothesis to emphasize the fact that
understanding figurative language does not require any special mechanism separate
from the one employed for literal language. The model emerged in the wake of the
studies which failed to support the idea that figurative processing takes longer than
literal processing (e.g., Blank, 1988; Harris, 1979; Pollio et al., 1984) and those
which specifically emphasized how the role of context had been downplayed in the
literal-first model (Gildea & Glucksberg, 1983; Keysar, 1989; Ortony et al., 1978;
Shinjo & Myers, 1987). For example, Gildea and Glucksberg (1983) found that
comprehension of metaphorical expressions was considerably faster when a disam-
biguating context was provided for less conventional novel metaphors, such as
Some marriages are iceboxes (see also Frisson & Pickering, 1999).
The parallel-processing view has been very influential as a framework for a num-
ber of models developed to account for L1 idiom comprehension. One of them, the
lexical representation hypothesis claims that idiom understanding involves i nitiating
simultaneous computations of both the literal and figurative meanings (Swinney &
Cutler, 1979). This model, referred to in the idiom literature as the lexicalization
hypothesis (cf. Glucksberg, 1993), the simultaneous processing hypothesis (cf.
Cronk & Schweigert, 1992), the unitization hypothesis (cf. Cutting & Bock, 1997),
82 A.B. Cieślicka
or the dual-process model (cf. Gibbs & Nayak, 1989) further assumes that idioms
are stored in the mental lexicon as long words. It thus dispenses with the need to
posit the existence of a special idiom lexicon or a special processing mode for com-
prehending idioms, as was the case with the idiom list hypothesis. Whether the idi-
om’s literal or figurative meaning is processed faster depends on the relative speed
with which full linguistic processing on the one hand and idiom access on the other
can be completed. Since idioms are stored as long words with single entries in the
lexicon, their figurative meaning can be retrieved before the semantic and syntactic
processing necessary for literal analysis is completed. Support for the lexical repre-
sentation model has been found in a number of studies (e.g., Glass, 1983; Hillert &
Swinney, 2001; Swinney & Cutler, 1979).
Another idiom model compatible with the parallel-processing view is the idiom
decomposition model (Gibbs & Nayak, 1989; Gibbs et al., 1989), under which idiom
processing is affected by the degree of idiom semantic decomposability. As explained
earlier, idiom decomposability or analyzability refers to the degree to which indi-
vidual meanings of idiom constituent words contribute to the idiom’s figurative inter-
pretation. On this view, language users normally perform literal analysis of idiom
constituents as part of the process of determining their figurative meanings. Since for
nondecomposable idioms (e.g., kick the bucket) such analysis fails to facilitate recog-
nition of an idiom’s figurative meaning (i.e., die), language users must abandon com-
positional processing and recover the figurative meaning directly from the lexicon,
which slows down the processing time. On the other hand, in decomposable idioms
(e.g., pop the question) individual constituents’ literal meanings contribute to the
idiom’s figurative meaning (pop= utter, question=marriage proposal), which facili-
tates their processing and makes those idioms easier to understand than nondecom-
posable ones. The idiom decomposition model has motivated a lot of research into
monolingual idiom processing, yielding contradictory results, with some studies sup-
porting the processing differences between decomposable and nondecomposable idi-
oms (e.g., Caillies & Butcher, 2007; Hamblin & Gibbs, 1999; Titone & Connine,
1994, 1999; Zhang, Yang, Gu, & Ji, 2013) and others failing to do so (Cutting &
Bock, 1997; Libben & Titone, 2008; Sprenger, Levelt, & Kempen, 2006; Tabossi,
Fanari, & Wolf, 2008; Tabossi, Wolf, & Koterle, 2009; Titone & Libben, 2014).
Yet another idiom model is the configuration hypothesis, specifically emphasiz-
ing the role of literal meanings in constructing idioms’ figurative interpretations
(Cacciari & Glucksberg, 1991; Cacciari & Tabossi, 1988). The configuration
hypothesis postulates a distributed representation of idioms in the mental lexicon,
such that the individual words participating in the idiomatic configuration are the
same lexical items which are accessed during literal comprehension. For example,
the word take has a lexical entry which gets activated when processing the literal
sentence The boy took the book. At the same time, the very same word is part of such
idiomatic configurations as take the bull by the horns or take to heart (Cacciari &
Tabossi, 1988). Upon encountering an idiomatic expression, the language compre-
hension system analyzes the idiom literally, while at the same time the phrase may
emerge as an idiomatic configuration. This idiomatic meaning retrieval occurs after
a sufficient portion of the phrase has been processed at a critical point referred to as
the idiomatic key, which is the information in the string that has to be processed
literally before the figurative meaning of an idiom can be activated (Tabossi &
Zardon, 1995, p. 275). Once the idiomatic key is reached and the figurative meaning
accessed, the literal analysis is terminated. One consequence of the idiomatic key is
the fact that high-predictable idioms, in which they key occurs early, permit faster
activation of their idiomatic interpretation than low-predictable idioms, in which the
key occurs late. This effect of idiom predictability on the activation of their literal
and figurative meanings has been subsequently supported in a number of studies
(e.g., Cacciari & Corradini, 2015; Cacciari & Glucksberg, 1991; Tabossi & Zardon,
1993, 1995); Vespignani, Canal, Molinaro, Fonda, & Cacciari, 2009. However,
some research has challenged the assumption that semantic and syntactic analysis
of the idiom is abandoned upon retrieval of its figurative meaning (e.g., Colombo,
1993; Flores d’Arcais, 1993; Konopka & Bock, 2009; Peterson & Burgess, 1993;
Snider & Arnon, 2012; Titone & Connine, 1994; Van de Voort & Vonk, 1995).
More recent, hybrid accounts of idiom processing such as the hybrid model (Caillies
& Butcher, 2007; Cutting & Bock, 1997; Sprenger et al., 2006) or constraint-based
model (Libben & Titone, 2008; Titone & Connine, 1999; Titone, Columbus, Whitford,
Mercier, & Libben, 2015) are also instantiations of the parallel- processing approach.
Those models acknowledge that idioms behave both as compositional phrases, whose
meaning might be assembled through word-by-word analysis and as noncompositional
units, whose figurative meaning can be retrieved directly from the mental lexicon. The
noncompositional nature of idioms is reflected in their conventionality and the fact that
they are highly automatized, multi-word units associated with arbitrarily stipulated
figurative meanings, which can be accessed directly from the mental lexicon. On the
other hand, idioms also behave compositionally, in that some of them are transparent.
Briefly, transparency indicates the degree to which the original motivation of the figu-
rative meaning can be inferred from the literal analysis of the expression (e.g., to take
the bull by the horns; see Nunberg, Sag, & Wasow, 1994).
In the course of language comprehension language users employ all the informa-
tion they have available to ensure successful understanding. Thus both compositional
analysis and direct figurative activation may be undertaken, and their contribution will
vary dynamically at different points during processing as well as for different idioms.
Because idiomatic expressions vary along a number of dimensions such as their pre-
dictability, semantic analyzability, frequency, syntactic flexibility (i.e., the degree to
which they can undergo syntactic modifications, such as not spill a single bean), or
literal plausibility, some idioms might lend themselves more easily to being processed
compositionally, whereas others will benefit from being retrieved holistically.
More importantly, the most recent neuroimaging studies looking at brain corre-
lates of idiom comprehension are also consistent with the parallel-processing view,
pointing to the simultaneous activation of the semantics of idiom component words
and a noncompositional semantic access of holistically stored idiomatic construc-
tions (e.g., Boulenger, Shtyrov, & Pulvermüller, 2012). Likewise, in the domain of
metaphor processing, recent electrophysiological studies support the idea that literal
meanings are activated fast and remain active even after the metaphorical meaning
has been accessed (De Grauwe, Swain, Holcomb, Ditman, & Kuperberg, 2010).
84 A.B. Cieślicka
The Graded Salience Model
Dispensing with the literal-figurative distinction, the graded salience hypothesis

(GSH; Giora, 1997, 2001, 2003; Giora, Fein, & Schwartz, 1998) argues that lan-
guage processing is not so much determined by literalness or figurativeness per se,
but rather by the degree of salience i.e., prototypicality, frequency, familiarity and
conventionality of a given linguistic unit. In other words, salient meaning is lexical-
ized meaning that is retrievable from the mental lexicon rather than from the context
(Giora, 1999, p. 919). In this approach, language comprehension entails the simul-
taneous operation of the bottom-up processes sensitive to salient information and
top-down integrative processes influenced by contextual information. Lexical
access, driven by the bottom-up mechanism, is exhaustive and ordered in such a
way that salient meanings are accessed faster, regardless of contextual bias.
Accordingly, in cases where the most salient meaning is intended, as in the process-
ing of conventional idioms (e.g., add fuel to the fire) their figurative, most salient
meaning (i.e., make things worse) is accessed directly, without the need to process
the less salient literal meaning (e.g., pour gasoline into flames). However, when the
intended meaning is less salient, for example the conventional idiom kick the bucket
is meant literally (i.e., strike the pail with one’s foot), language comprehension will
first start by activating the more salient meaning (die) before deriving the contextu-
ally intended less salient meaning (Giora, 1997). It follows then, that for highly
conventionalized idioms, their figurative interpretation would be actually faster than
their literal (as intended) interpretation.
Support for the GSH has been found in research into the processing of meta-
phors (Peleg, Giora, & Fein, 2001), jokes (Giora, 1991), irony (Giora & Fein,
1999a, 1999b; Giora et al., 1998) and idioms (Giora & Fein, 1999c). In the domain
of metaphor comprehension, the GSH is reflected in the metaphor processing the-
ory called the career of metaphor theory (Gentner & Bowdle, 2001). This view
posits that metaphor processing is contingent upon its conventionality. Thus, while
novel metaphors are processed as comparison statements by aligning the target con-
cept (topic) with the literal base concept (vehicle), over time, stable metaphoric
categories are created so that conventional metaphors are processed by aligning the
target concept with the abstract schema labeled by the base term. This suggests that,
over time, as metaphors become familiar with increased use, they get lexicalized in
the mental storage, thus implying a shift in cognitive processing as a function of
familiarity (saliency).
As shown in this brief review, a number of theoretical frameworks have been
proposed to explain processing of figurative language. Those frameworks have been
adopted to address the question of how bilingual language users and second lan-
guage learners go about the task of acquiring and comprehending figurative expres-
sions. Research into bilingual processing of figurative language is discussed in the
subsequent section.
Figurative Processing in L2 and Bilingual Language Users
While figurative language has been extensively researched in the monolingual lit-
erature, studies exploring how bilinguals process figurative expressions have been
surprisingly scarce, despite the importance of formulaic expressions in successful
communication. Indeed, pedagogically-oriented research has repeatedly shown
how difficult it is for nonnative language users to acquire L2 figurative competence
(e.g., Boers, Demecheleer, & Eyckmans, 2004; Boers & Lindstromberg, 2008;
Bortfeld, 2003; Cooper, 1999; Danesi, 1986, 1992; Kecskes & Papp, 2000; Laufer,
1989; Lazar, 1996; Littlemore, 2001, 2010; Low, 1988; MacArthur, 2010; Wray,
2000, 2002, 2012). L2 learners face considerable challenge when reading (Boers,
Eyckmans, & Stengers, 2007; Martinez & Murphy, 2011), writing (Howarth, 1998;
Kathpalia & Heah, 2011), speaking (Kecskes, 2016) and listening to formulaic lan-
guage (Van Lancker Sidtis, 2003). Contrary to the high volume of pedagogically-
based studies, psycholinguistic research into the nature of the processes underling
bilingual figurative comprehension is still limited. Of the few studies available to
date, most have focused on the processing of idioms and other formulaic expres-
sions such as phrasal verbs (e.g., make up) or collocations (make room). Bilingual
comprehension of other figurative tropes such as metaphor, metonymy, proverbs, or
irony has received even less attention.
Broadly speaking, research questions regarding the processing of figurative lan-
guage in L2 can be divided into four major themes. The first concerns the represen-
tational format of figurative expressions in the mental lexicon and the idea that
formulaic phrases might be stored and processed differently depending on the status
of the language (L1 vs. L2). While there is a wide agreement that formulaic lan-
guage enjoys a processing advantage over nonformulaic language in native speak-
ers, in that multi-word units are retrieved and processed much faster than literal
language, possibly as lexicalized chunks (e.g., Conklin & Schmitt, 2008; Cutter,
Drieghe, & Liversedge, 2014; Schmitt & Carter, 2004; Siyanova-Chanturia,
Conklin, & Schmitt, 2011; Siyanova-Chanturia, Conklin, & van Heuven, 2011;
Underwood, Schmitt, & Galpin, 2004), the question of whether the same holds true
for bilinguals has been debated (see Conklin & Schmitt, 2012). Related to the issue
of how figurative language is stored and retrieved is the time course of literal vs.
figurative meaning activation and factors which might contribute to how strongly
each meaning is activated when bilinguals process nonliteral language. Here a num-
ber of theoretical proposals have been developed, including the suggestion that lit-
eral meanings might be particularly relevant for understanding formulaic language
by nonnative language users (e.g., Abel, 2003; Cieślicka, 2006a, 2006b; Kecskes,
2006; Liontas, 2002).
The next major question focuses on whether figurative language processing
might be subserved by different neural networks in native and nonnative language
users. Accordingly, some researchers looked into possible differences in neural
bases of semantic and figurative processing in native and nonnative languages
(e.g., Cieślicka & Heredia, 2011; Faust, Ben-Artzi, & Vardi, 2012). Finally, since
86 A.B. Cieślicka
a bilingual language user has more than one language at his/her disposal and non-
literal expressions tend to differ across languages (e.g., let the cat out of the bag
in English vs. let the steam out of your mouth in Polish), bilingual figurative
research has also looked at the role of native language in L2 figurative processing
and whether cross-language overlap would help or hinder L2 figurative compre-
hension. The sections below elaborate on each of the major bilingual figurative
research themes.
Holistic Versus Compositional Storage and Retrieval
The holistic vs. compositional storage question mirrors the debate in the monolingual
literature between followers of the figurative-first models, on the one hand, and the
parallel-processing ones, on the other. While figurative-first approaches would claim
that figurative expressions are stored as single units in the mental lexicon and retrieved
directly, the parallel-processing view would allow for compositional processing to be
initiated simultaneously with figurative retrieval (e.g., the lexical representation
model), where literal computations are either terminated upon retrieval of the idiom-
atic meaning (e.g., the configuration hypothesis) or continued for the whole idiom
phrase (e.g., the idiom decomposition hypothesis and the hybrid models). The sug-
gestion that figurative phrases would be stored as single units and retrieved holisti-
cally has also been motivated by the very definition thereof in the lexical processing
literature, as argued by Wray (2002, p. 9), where formulaic sequences are defined as
being stored and retrieved whole from memory at the time of use. Accordingly, since
idiomatic expressions are highly automatized and retrieved holistically from the
mental lexicon, they should be processed faster than novel nonformulaic phrases, the
view referred to as the holistic hypothesis (Jiang & Nekrasova, 2007). However, this
might not be the case for nonnative language users since figurative meanings of L2
idioms are not well-established in their lexical storage and are therefore more likely
to be processed through a word-for-word analysis of their component parts.
A number of studies conducted to verify the holistic hypothesis have employed
the eye-tracking paradigm, where eye-movements are recorded as participants read
linguistic stimuli displayed on a computer screen. Briefly, in the eye-tracking tech-
nique reader’s eye movements are recorded and analyzed for the following
characteristics relevant for either a single word or a whole phrase as a target region:
skipping (a word not being fixated on initially), first fixation duration (how long the
eyes fixate on the word for the first time) and first pass reading time (sum of all fixa-
tions on the phrase before eyes move either to the left or right). These are known as
early measures in that they reflect early automatic processes of lexical access. The
late measures reflect integration processes i.e., how the word/phrase is being incor-
porated into the higher-order interpretation of the sentence and include total reading
time (sum of all fixations on the target word or phrase region that also includes
regressions i.e., return fixations to those parts of the text which were initially diffi-
cult or confusing) and total fixation count (total number of times a target region was
fixated on during the trial). The appeal of the eye-tracking paradigm is that it is
ecologically valid, in that the participant is engaging in a natural reading behavior
and reading measures very straightforwardly capture the ease or difficulty of lin-
guistic processing. For example, the more demanding the text, the larger the number
of fixations and regressions, as well as the longer the fixation durations are likely to
be (see for example, Carrol & Conklin, 2014).
In an eye tracking study investigating native and nonnative formulaic language
processing, Underwood et al. (2004) presented native and nonnative speakers of
English with idiomatic expressions (e.g., honesty is the best policy) and novel
nonformulaic sequences (it seems that his policy of…) where the critical region
was the last word of both the idiomatic phrase and of its control nonformulaic
sentence (policy). The rationale behind employing this technique was that, if for-
mulas are indeed stored and retrieved holistically, then they do not need a detailed
word-by- word analysis which is necessary for understanding nonformulaic
phrases. Accordingly, the last word in formulaic sequences should obtain very
short reading times, compared to when the same word is used in literal nonformu-
laic sequences that require full linguistic analysis in order to be successfully
understood. Indeed, fixation count and fixation duration results showed a clear
processing advantage for formulaic over nonformulaic phrases in the native
speaker data, where the last word of the idiom obtained fewer and shorter fixations
than the same control word used in a novel nonformulaic sequence. In turn, non-
native speaker results were not as clear-cut: while there were fewer fixations on
the last word of idiomatic over non-idiomatic phrases, no differences were found
in fixation duration on the target words, irrespective of whether the words were
part of the idiom or a matched nonformulaic phrase, suggesting that idiomatic
phrases required as much time to process as their literal counterparts.
In a similar study, Siyanova-Chanturia, Conklin, & Schmitt (2011) and Siyanova-
Chanturia, Conklin, & van Heuven (2011) presented idioms to native and nonnative
speakers of English and recorded the number and length of fixations. The idioms
varied in whether they were used figuratively (at the end of the day: finally) or liter-
ally (at the end of the day: in the evening). In addition, as in Underwood et al.’s study,
novel nonformulaic phrases were created (at the end of the war). The eye measures
were recorded both for the whole idiom and its control phrase, as well as for the last
word of the idiom (e.g., day) and its control word in a novel phrase (e.g., war).
Results showed a processing advantage for idioms over novel phrases in the native
speaker data. This advantage held true for idioms both when they were used figura-
tively and literally. In turn, there were no differences in processing times between
idioms and novel phrases in the nonnative speaker data. In addition, idioms used
literally were processed faster than those used figuratively, suggesting that literal
meanings of idioms might be more salient than figurative ones in L2 processing.
Some other studies have demonstrated no advantage for processing formulaic as
compared to nonformulaic expressions, regardless of the language (native vs. non-
native) status. For example, Schmitt and Underwood (2004) employed the self-
paced reading technique to compare reading times of formulaic and nonformulaic
sequences in native and nonnative speakers of English. In a self-paced reading task
88 A.B. Cieślicka
participants are visually presented with a linguistic material in such a way that they
directly control the speed with which they advance from one word (or phrase) to
another. Stimuli are presented one at a time on a computer screen in predetermined
segments (either a single word or a phrase). As participants press the button to
advance to the next stimulus, the computer software measures how long they take to
process the linguistic material presented (see García, Cieślicka, & Heredia, 2015).
To explore the representational status of multi-word sequences, Schmitt and
Underwood compared the reading time for the last word of a formulaic sequence
(e.g., the straw that broke the camel’s back) to the reading time of the same word
when it was used in a nonformulaic control expression (e.g., with one hand tied
behind your back). No significant processing advantage was found for the last word
of formulaic, as compared to nonformulaic sequences and the effect was identical in
both native and nonnative speakers of English. However, a study suffered from a
number of methodological shortcomings, one of them being the fact that a substan-
tial proportion (over 40%) of formulaic sequences was unknown to nonnative
speakers. In addition, word-by-word presentation of the phrases enforced by the
nature of the task might have prevented holistic retrieval by making participants
focus on the compositional analysis of the formulaic phrases.
In a follow-up study, Conklin and Schmitt (2008) presented formulaic (every-
thing but the kitchen sink) and matched nonformulaic control strings (everything in
the kitchen sink) embedded in larger discourse context to native and nonnative
speakers of English. Whereas in Schmitt and Underwood (2004) words were dis-
played one word at a time, here stimuli were displayed one line at a time. Both
native and nonnative speakers processed formulaic sequences faster than nonformu-
laic controls. In addition, this processing advantage was found both when the for-
mulaic sequences were embedded in context biasing their figurative reading (e.g.,
take the bull by the horns: attack a problem) and literal reading (e.g., take the bull
by the horns: wrestle an animal). These results suggest that formulaic sequences are
easier to process than control nonformulaic language, regardless of whether they are
used in their figurative or literal reading and this advantage holds true both for
native and nonnative language users. The fact that no processing differences were
found for idiomatic sequences used figuratively and literally offers support for the
parallel-processing view of figurative language understanding.
Using a different methodology, Jiang and Nekrasova (2007) asked native English
speakers and speakers of English as a second language to make grammaticality
judgments on formulaic and nonformulaic sequences. Participants were shown
word strings such as on the contrary or on the chair and asked to determine if those
word strings were grammatical. The rationale behind this task was that, if formulaic
sequences are stored as holistic phrases and retrieved as such, the time taken to
assess their grammaticality will be much shorter than for nonformulaic phrases
which require a full linguistic analysis. Both groups responded significantly faster
to formulaic sequences and made fewer errors on formulaic, as compared to nonfor-
mulaic sequences, supporting the holistic hypothesis. Further, given equal advan-
tage for formulaic over nonformulaic sequences for native and nonnative speakers,
the authors claim that the holistic hypothesis applies equally to native and nonnative
language users. To preclude the possibility that the demonstrated processing advan-
tage for formulaic expressions was caused by familiarity with the visual shape of
the formulas, in a subsequent experiment Jiang and Nekrasova manipulated the let-
ter case in which the formulas were presented. Accordingly, all the stimuli were
shown in uppercase letters. The same results, i.e., faster processing times and fewer
errors for formulaic sequences in both native and nonnative speaker groups were
again demonstrated, suggesting that the ease of accessing formulaic sequences is
not so much a product of visual shape recognition but of genuinely holistic storage
and retrieval of those phrases from the mental lexicon.
All in all, results are mixed with regard to the holistic processing of formulaic
expressions by bilingual language users. While a number of studies have shown that
bilinguals might be more likely to process idiomatic expressions compositionally
rather than holistically (e.g., Siyanova-Chanturia, Conklin, & Schmitt, 2011;
Siyanova-Chanturia, Conklin, & van Heuven, 2011; Underwood et al., 2004), other
research (e.g., Conklin & Schmitt, 2008; Jiang & Nekrasova, 2007; see also
Paulmann, Ghareeb-Ali, & Felser, 2015) showed no differences between native and
nonnative figurative processing. It seems that the processing strategy employed
might vary for each individual language user and for each individual figurative
expression, in line with hybrid models (e.g., Titone et al.’s, 2015 constraint-base
model) which focus on determining the conditions where either holistic or composi-
tional type of processing would be instantiated. One of such conditions might be L2
proficiency, as shown in the study by Siyanova-Chanturia, Conklin, and Schmitt
(2011) and Siyanova-Chanturia, Conklin, and van Heuven (2011), where native
speakers and highly proficient nonnative speakers performed similarly in processing
binomial expressions (bride and groom) and their reversed forms (groom and bride),
showing equal sensitivity to their frequency as a lexicalized form. However, as
rightly pointed out by Siyanova-Chanturia (2015), a caution needs to be exercised
against equating the speed of processing with the idea that multi-word expressions
are processed as unanalyzed chunks and retrieved as such without any compositional
analysis from the mental lexicon. While the speed of processing on its own might
only be indicative of higher saliency and faster availability of the idiomatic meaning,
it cannot be taken to support the position that constituents of the multi-word expres-
sions are not automatically activated in the course of figurative processing.
Accordingly, studies relying solely on the speed of processing, without considering
activation of individual constituents, would not be best equipped to weigh on the
issue. Clearly, more research is needed focusing on the processing of individual
components of multi-word expressions so as to be able to draw more definitive con-
clusions concerning the holistic hypothesis in monolinguals vs. bilinguals.
Literal or Figurative Salience in L2 Figurative Processing?
The issue of whether literal or figurative meaning has a priority in the course of L2
figurative processing is based on the theoretical assumptions of the graded salience
hypothesis (Giora, 2001) discussed earlier. Under this view the meaning activated
first is the one that is more salient or more easily retrievable from the mental lexicon.
90 A.B. Cieślicka
It seems reasonable to assume that a nonnative language user whose L2 metaphori-

cal fluency is far from native-like might resort to relying on a literal analysis of a less
well-known multi-word expression. This is the assumption that motivated a number
of L2 figurative processing models. One of them, the model of dual idiom represen-
tation (Abel, 2003) suggests separate lexical representations for nondecomposable
idioms, which are stored as lexicalized idiom entries, and for decomposable idioms,
which are represented via constituent entries i.e., lexical entries for their component
words. The model was tested in a decomposability judgment task where German
learners of English were presented with a list of English idioms and asked to decide
if those idioms’ constituent words contributed to their figurative interpretation. The
participants tended to rate many nondecomposable English idioms as decompos-
able, which reflects the L2 learners’ strategy of reliance on literal meanings of L2
idioms and supports the suggestion that literal meanings are particularly prominent
in the course of nonnative figurative language processing.
Similarly, Liontas (2002, 2015) proposed a model of L2 idiom processing called
the idiom diffusion model of second languages that emerged from his studies into the
processing of idioms in English by adult L2 learners from such various L1 back-
grounds as Modern Greek, Spanish, Finnish, French, and German. The model
assumes two stages of figurative processing: the prediction phase and the confirma-
tion or replacement, reconstructive phase. In the prediction phase, the L2 learner
tries to understand an L2 expression by using a literal analysis of its individual lexi-
cal items, especially if the idiom is presented out of context. In this stage the learner
might also resort to activating what seems to be the closest L1 translation equivalent,
so as to be able to infer the phrase’s figurative interpretation. In the subsequent stage,
if the context is provided, it helps in narrowing down the range of possible interpre-
tations, which leads either to the confirmation of earlier predictions or to the replace-
ment and reconstruction of new ones. In line with the literal-first approach, Liontas
(2002) suggests that L2 figurative understanding calls for a special processing mode,
the view compatible with the idiom list hypothesis (Bobrow & Bell, 1973). Even
with highly familiar idioms, word-by-word literal analysis is an o bligatory step in
L2 idiom comprehension, regardless of the contextual information available. Indeed,
Liontas’ (2002) idiom comprehension study revealed that translation was the most
common strategy employed by learners to interpret the meanings of the idiomatic
phrases, which suggests a central role of the learners’ native language in the compre-
hension and interpretation of L2 idioms. Analysis of other strategies (e.g., guessing)
likewise indicated that L2 learners made frequent recourse to their L1 knowledge for
L2 idiom interpretation, a finding supporting Kellerman’s (1983) claim of L1 trans-
fer. Given these results, Liontas concludes that L2 idioms are computed in their lit-
eral sense until this is no longer possible (2002, p. 179).
A similar assumption concerning the importance of literal analysis in L2 figura-
tive processing is reflected in Cieślicka (2006a, 2006b) literal salience model,
which was developed to explain the acquisition and processing of formulaic lan-
guage by L2 learners in a formal L2 setting. The model assumes that salience is
directly correlated with the degree of metaphorical proficiency and learner’s famil-
iarity with a given L2 idiom (see also Kecskes, 2006). While for less proficient L2
learners, figurative expressions are likely to be mainly processed literally using the
parasitic strategy (see Cieślicka, 2015), with time the most familiar idiomatic
expressions develop their own lexicalized entries and might be retrieved directly
from the mental lexicon. Under the parasitic strategy view, L2 lexical development
is initially exclusively driven by reliance on L1. Thus, a newly acquired L2
idiom will not immediately develop a separate lexical representation in the learner’s
lexicon, but will instead first plant itself in the most closely corresponding
L1-translation equivalent, effectively hijacking its lexical and conceptual represen-
tation. The processing of such a less well-known L2 idiom is hence entirely para-
sitic on L1, as it relies on retrieving the L1-translation to help understand its
meaning. Over time, this parasitic idiom entry gets restructured, as the learner’s
fluency and familiarity with the L2 idiom are growing, until finally the L2 idiom
gets to develop its own lexical entry separate from the L1 equivalent. Support for the
literal salience model has been demonstrated in behavioral studies employing the
cross-modal lexical priming (CMLP) paradigm (Cieślicka, 2006a, 2006b), moving
window paradigm (Cieślicka, 2007), split visual presentation technique (Cieślicka
& Heredia, 2011), as well as in production (Cieślicka, 2010) and eye-tracking stud-
ies (Cieślicka, Heredia, & Olivares, 2014).
However, other studies failed to show literal over figurative priority in nonnative
figurative language processing. For example, McPartland-Fairman (1989) used the
CMLP paradigm to explore literal and figurative meaning activation in the course of
processing phrasal verbs i.e., expressions which have an ambiguous verb-preposi-
tion interpretation such as turn on, as in turn on TV versus turning left. The CMLP
paradigm is a behavioral technique combining auditory and visual modes of presen-
tation. In a typical experiment, participants are presented with auditory input via
headphones and make lexical decisions (i.e., decide if a presented target string is a
word, such as CAT or a nonword, such as CRAT) about target stimuli displayed on
a computer screen. The time or reaction time (RT) taken to make a lexical decision,
is indicative of the activation of various meanings of the auditorily processed input
that either speeds up (primes) or slows down (interferes with) the visual target rec-
ognition. In McPartland-Fairman’s (1989) study, participants listened to sentences
biasing either the figurative (e.g., The doctor told the patient he was working too
hard and needed to do more exercise or he would get a heart attack. He didn’t have
any choice, so he signed up the next day for an exercise class) or literal interpretation
of the phrasal verb (e.g., The soldier was writing to his girlfriend and he had a lot to
tell her that day. When he finished, there wasn’t enough space for his name at the
bottom of the letter. He didn’t have any choice, so he signed up the side of the paper).
Immediately at the offset of the phrasal verbs, participants saw visually displayed
targets words for a lexical decision that were related either to the figurative (e.g.,
ENROLL) or literal (e.g., SIGNATURE) meaning of the phrasal verb. While the
naming times recorded were indeed faster for related compared to unrelated targets,
there was no difference in RTs between literally- and figuratively-related targets,
suggesting that both literal and figurative meanings got equally activated in
the course of L2 phrasal verb processing. Those results were taken to imply that
nonnative figurative processing is consistent with the parallel-processing view
92 A.B. Cieślicka
and lexical representation hypothesis, under which both literal and figurative mean-
ings are automatically activated upon encounter of a nonliteral phrase.
The most recent studies exploring the varying salience status of literal and figu-
rative meanings have applied ERPs (event-related potentials) to address the ques-
tion of literal vs. figurative priority in L2 processing (e.g., Paulmann et al., 2015).
Briefly, ERPs provide a noninvasive measure of brain function by examining
stimulus-locked averages i.e., changes in scalp-recorded electrical activity which
occur in response to a specific event (such as a visual or auditory stimulus).
These changes of voltage in the brain electroencephalogram (EEG) are referred to
as ERPs and for the last few decades they have been extensively used to investigate
moment-by-moment perceptual and cognitive processes that underlie language
comprehension, primarily because of their superior temporal resolution and sensi-
tivity (see Kutas and Federmeier, 2000 for a review). One of the most widely
researched ERP components is the N400, named for its negative polarity and identi-
fied as index of semantic processing (see Kutas & Delong, 2008). It peaks around
400 ms after the onset of a critical word and is sensitive to many factors, some of
which are the word’s frequency, concreteness, repetition, position in the sentence
and predictability in a given context (Kutas & Federmeier, 2000, 2011). The N400
component is particularly prominent in response to words semantically mismatch-
ing the preceding context (e.g., I take my coffee with cream and DOG; Kutas &
Hillyard, 1980a, 1980b). The N400 is thus taken as indicative of the ease or diffi-
culty of semantic retrieval from memory and contextual integration, which makes it
an ideal tool for investigating the processes involved in understanding inherently
ambiguous figurative expressions.
In a recent ERP study by Paulmann et al. (2015), the N400 was used to explore
how and when figurative and literal meanings are accessed in the course of native
and nonnative processing of phrasal verbs (e.g., run over), which can be interpreted
both figuratively (kill by driving) and literally as a verb+preposition combination
(walk over something). Monolingual (native English) and bilingual (Arabic speakers
of English) participants were presented with ambiguous verb-preposition strings and
the ERP data were recorded in response to the disambiguating word, which either
biased the figurative reading of the phrasal verb (e.g., I heard that Mr. Smith ran over
the old farmer early this morning) or the literal meaning (e.g., I heard that Mr. Smith
ran over the old bridge early this morning). Both native and nonnative language
groups showed an increased amplitude of the N400 component in response to words
biasing a literal reading of the phrasal verb (farmer), as opposed to words biasing the
figurative reading (bridge). Since a larger amplitude of N400 indicates an enhanced
cognitive effort on the part of the language processing mechanism, these results sug-
gest that both L1 and L2 users processed the phrasal verbs with equal ease and
retrieved their figurative meaning more easily than their literal meaning. Those find-
ings are in line with the figurative-first approach and imply that formulaic phrases
might be stored and retrieved holistically, both for native and nonnative language
users, lending support to the direct access model developed in the monolingual lit-
erature and to the holistic hypothesis discussed in the previous section.
How can we reconcile these seemingly contradictory results? One possibility is

that the type of processing and literal vs. figurative activation will depend on the
figurative trope involved. Whereas research showing literal-only activation in the
course of L2 figurative processing primarily employed idiomatic expressions as
target stimuli, both McPartland-Fairman’s (1989) and Paulmann et al.’s (2015)
studies, which showed no difference between figurative and literal activation, used
phrasal verbs. Idioms and phrasal verbs are both figurative tropes belonging to the
same large category of multi-word expressions, but they differ importantly on a
number of dimensions, the most prominent of which is their length. It is more
likely for short and mostly opaque phrasal verbs to be learned as unanalyzable
chunks and stored as such in the mental lexicon by nonnative users. On the other
hand, most idioms are longer than two words and many are either transparent or
compositional or both, thus lending themselves much more easily to a literal analy-
sis which seems to be undertaken even if it yields incorrect figurative meaning, as
is the case with nondecomposable or opaque idioms. On account of those charac-
teristics, a default L2 idiom processing mode might hence be literal, at least until
they become deeply entrenched in the learner’s mental lexicon. Likewise, a longer
formulaic expression might need much more time to become fully lexicalized and
retrieved holistically than a shorter one, as the longer the phrase the more lexical
links need to be created between its constituents before they can form a configura-
tion (see the configuration hypothesis).
In addition, the strength of literal vs. figurative activation might be modulated by
L2 user’s proficiency or degree of language dominance, as shown by Matlock and
Heredia (2002) in a study exploring the processing differences between early vs.
late bilinguals. In short, the term early bilinguals refers to individuals whose L2 is
learned early in life (i.e., early childhood), whereas the term late bilinguals to those
whose L2 is learned late in life (i.e., after childhood). Cutoff criteria for early vs.
late bilingual distinction vary widely in the bilingual literature, with the cutoff L2
acquisition criterion after the age of four years and before 12 years of age suggested
most recently for a broad definition of early bilingualism, and after 12 years of age
for late bilingualism (see Heredia & Cieślicka, 2014 for extensive discussion). In
Matlock and Heredia’s (2002) study, monolinguals and early Spanish–English
bilinguals understood phrasal verbs used figuratively more quickly than phrasal
verbs used literally. On the other hand, late bilinguals processed literal meanings
faster than figurative ones. According to Matlock and Heredia, those findings sug-
gest that monolinguals and early bilinguals process figurative language directly,
consistent with the figurative-first approach and direct access model, whereas late
bilinguals activate literal meanings first, as predicted by the literal-first view.
This suggestion was further examined in an eye-tracking study by Cieślicka et al.
(2014), who looked at the effect of language dominance in bilingual formulaic lan-
guage processing. In the study, Spanish–English bilinguals, dominant either in
Spanish or English, were presented with literally plausible idioms. Each idiom was
used either figuratively (in deep water: in trouble) or literally (in deep water:
immersed in water) and presented in a sentence where the figurative- or literal-
biasing context either preceded the idiom (e.g., Since both of us were equally guilty
94 A.B. Cieślicka
of causing the overspend, we both knew we were in deep water, and very likely to
lose our jobs) or followed the idiom (e.g., Within seconds she realized she was in
deep water, and that she would very soon come to regret her words). The reading
measures (length and number of fixations, regressions to the idiom region) were
monitored both for the idiom itself and the post-idiom region, that is, the disambigu-
ating part of the sentence biasing either the literal or figurative reading of the idiom.
Results showed that English-dominant bilinguals had shorter total reading time for
the post-idiom region when the idioms were used figuratively than when they were
used literally, whereas the reverse was true for Spanish-dominant bilinguals, with
shorter reading time for the post-idiom region in sentences biasing the literal than
the figurative meaning of the idiom. In addition, English-dominant bilinguals had
significantly fewer fixations for the post-idiom region than Spanish-dominant ones
when the idioms were used figuratively, suggesting that figurative meanings of idi-
oms were more salient for bilinguals dominant in English than for bilinguals domi-
nant in Spanish. Regression data for Spanish-dominant bilinguals also revealed
faster processing of literal than figurative meanings, in that significantly fewer
regressions were recorded to the idiom region when the idiom was used literally
than figuratively. Overall, the study supported the findings from Matlock and
Heredia (2002), in that literal activation was associated with bilinguals dominant in
Spanish, and hence less fluent in English, whereas figurative activation was shown
in bilinguals dominant in English.
Although limited, psycholinguistic studies into bilingual processing of other
figurative tropes have also pointed to language dominance as a factor modulating
figurative and literal meaning activation. In a cross-modal priming experiment with
fluent nonnative speakers of English, Heredia and Muñoz (2015) asked their partici-
pants to name visually displayed target words which were literally (pastry) or figu-
ratively (boxer) related to the metaphoric referential description (creampuff,
referring to the fighter who always lost in battles). The metaphoric referential
description was presented auditorily, where a larger discourse context was provided
leading to the sentence of interest, such as His friend replied “the creampuff didn’t
even show up, I can’t believe it!” Visual targets were displayed either at 0 or 1000 ms
after metaphoric reference (creampuff) offset. Results showed significant priming
for figuratively related targets already at 0 ms, suggesting that bilinguals can effort-
lessly access figurative meanings of metaphoric referential descriptions. More inter-
estingly, significant priming was found for literal-related targets at 1000 ms,
implying that literal meaning remains active even after the intended figurative inter-
pretation has been successfully computed. In a subsequent experiment, Spanish–
English bilinguals were recruited whose profile of language dominance varied. This
time, comparable literal and figurative activation was found at 0 ms and at the time
window of 300 ms employed for this experiment. In addition, activation at 300 ms
was substantially stronger than at 0 ms, both for literal and nonliteral meanings.
Heredia and Muñoz (2015) suggest that the differing results obtained in Experiments
1 and 2 can be explained within the framework of the GSH. While bilingual partici-
pants in Experiment 1 were all highly dominant in English, this was not the case for
Experiment 2 participants. Accordingly, figurative meanings might have been more
salient and available faster for Experiment 1 group, whereas bilinguals in Experiment
2, whose metaphorical competence was substantially lower, were more likely to

process metaphoric descriptions literally and activate both literal and figurative
meanings in parallel, regardless of the time course.
Further support for the role of language dominance in modulating priority of
either literal or figurative meanings comes from a recent eye-tracking study into
the processing of metaphoric reference by Spanish–English and English–Spanish
bilinguals (Heredia & Cieślicka, 2016). While Spanish-dominant bilinguals
obtained shorter first fixation durations on metaphoric references used literally
(creampuff referring to pastry) than figuratively (creampuff referring to weak
boxer), the results suggested that literal meanings might be more salient for less
proficient speakers, English-dominant and balanced bilinguals had comparable
reading times for both literal and figurative meanings, in line with the parallel-
processing view. Those differences were only true for early-processing measures,
such as first fixation duration, but did not extend to late-processing measures, such
as total reading time, where again literal meanings appeared easier to process than
figurative ones, regardless of bilingual’s language dominance. Heredia and
Cieślicka interpreted these results within the framework of Giora’s retention
hypothesis (Giora, 2002), under which meanings potentially useful for the evolv-
ing interpretation of the figurative expression are kept active for some by the lan-
guage processor.
The role of language dominance in bilingual metaphor comprehension is also
reported in Vaid, López, and Martinez’s (2015) interference effect study. Spanish–
English bilinguals were presented with English sentences that were either high-
typical examples of a category (e.g., Some flowers are daisies), low-typical category
examples (e.g., Some flowers are gladiolas), standard false items (e.g., Some flowers
are cobras), true metaphors (e.g., Some cats are detectives), or scrambled meta-
phors (e.g., Some cats are chains). The participants’ task was to decide, as quickly
as possible, if the presented sentences were literally true or false. Of issue was
whether, as demonstrated for monolinguals (Glucksberg et al., 1982), bilingual par-
ticipants would experience a metaphor interference effect and take longer to reject
as literally false the sentences that were metaphorically true (e.g., Some cats are
detectives). Taking longer to reject literally false statements which are metaphori-
cally true implies that figurative meaning is automatically and obligatorily activated
(cf. the figurative-first model). A robust metaphor interference effect was found in
bilinguals who were dominant in English and equally proficient in Spanish and
English. When tested in both languages, the effect was found only for Spanish and
only in the group that was equally proficient in both languages. Taken together,
those results imply that metaphors can be accessed directly in both languages of a
bilingual and reverberate the importance of language dominance in figurative pro-
cessing identified in earlier research.
It seems then that the issue of literal vs. figurative salience in L2 nonliteral pro-
cessing is not simply an either-or question, but it appears to be influenced by a
whole range of factors, only some of which have been identified so far. One of them
is language dominance, with speakers highly proficient and dominant in a given
language likely to process figurative expressions effortlessly and directly, whereas
speakers less fluent in a language (and nondominant bilinguals) likely to rely mostly
96 A.B. Cieślicka
on literal processing. Reliance on literal vs. figurative processing should perhaps be

viewed as a continuum, reflecting the evolving proficiency of the L2 user in ques-
tion. More likely, the literal vs. figurative processing is not so much a macro-level
or global processing strategy applied indiscriminately to all figurative expressions
in the language user’s repertoire, but rather a micro-level strategy adjusted individu-
ally to each and every lexical item in the learner’s ever-evolving mental lexicon.
The more familiarized the nonnative speaker becomes with a given figurative
expression, the less likely s/he is to rely on a literal analysis in the course of its
processing, as the expression’s figurative meaning becomes strongly encoded in the
mental lexicon and more easily available for automatic retrieval. Akin to the
syntagmatic-paradigmatic shift reported by Söderman (1993) for the evolving lexi-
cal network of a child and to the formal-semantic shift reported for patterns of lexi-
cal associations in L2 learners (Meara, 1980), the L2 learner might be viewed as
experiencing hundreds of micro-shifts in salience status for each newly acquired
formulaic expression, where literal-first strategy gradually gives way to the figura-
tive-first or parallel-processing strategy, as the learner’s metaphorical competence
grows and familiarity with a given expression strengthens.
In addition to dominance, an important factor likely to influence L2 literal and
figurative activation patterns is the type of figurative trope, as mentioned earlier.
While some expressions, such as literally plausible and highly transparent idiomatic
phrases might be more conducive to being activated literally and processed compo-
sitionally, others such as short and highly fixed formulas might become lexicalized
faster and more likely to be activated figuratively. Since bilingual research into figu-
rative tropes other than formulaic phrases such as idioms, phrasal verbs, or colloca-
tions is still conspicuously absent, it is difficult to speculate further, but this area of
research, if pursued, is likely to reveal many interesting insights. A promising
glimpse into such new possibilities is afforded by the latest research in the domain
of irony processing, which has been so far mostly unexplored for L2 learners. Recent
studies into L2 irony processing point to the relevance of attitude and evaluative
processing in the course of irony communication and comprehension (Bromberek-
Dyzman, 2015; Bromberek-Dyzman et al., 2010). To look at the processing of atti-
tudinal meaning, an emotive decision task was employed where participants were
presented with an evaluative comment, such as praise or criticism and needed to
judge whether the target comment expresses a favorable (positive valence) or unfa-
vorable (negative valence) emotion. In a self-paced emotion decision task,
Bromberek-Dyzman, Rataj, and Dylak (2010) presented Polish–English bilinguals
with short scenarios followed by ironically- or literally-intended statements (e.g.,
She is so nice) and had them decide if the comment was favorable or unfavorable.
Accuracy and response time turned out to be highly comparable, regardless of
whether participants performed the task in their dominant (Polish) or nondominant
(English) language. In addition, in both languages ironic utterances took longer to
process than literal ones, suggesting an increased challenge in comprehending this
figurative trope. In subsequent experiments, Bromberek-Dyzman (2015) replicated
those results, showing similar patterns of irony processing for native and nonnative
speakers and suggesting it might not so much be the literality or figurativeness per
se that pose a challenge in irony processing, but the valence (i.e., affective attitude)
which is crucial in comprehending ironic statements. On this view, attitudinal con-

tent (i.e., likes, dislikes) present in irony crucially influences the affective state of the
interlocutor and is a major driver of irony comprehension. In Bromberek-Dyzman’s
experiments, significant differences were found between positive and negative atti-
tude processing, such that positive valence (literal praise) resulted in processing
facilitation (faster responses and higher accuracy), whereas negative valence resulted
in processing inhibition (longer responses and lower accuracy), regardless of
whether the statement was literal (literal criticism) or ironic (critical irony). This line
of research thus suggests that valence emerges as more important in irony process-
ing than the linguistic distinction between literal and nonliteral meaning. Certainly
more complex and less conventionalized types of figurative language might depend
on a range of additional factors, linguistic and extralinguistic, for their successful
processing by nonnative language users, and here native vs. nonnative processing
patters might be less straightforwardly different (or similar).
Neurological Bases of Bilingual Nonliteral Processing
The question of a neurological basis for bilingual nonliteral processing is moti-

vated by numerous neuropsychological investigations in the monolingual litera-
ture, which have suggested the possibility that formulaic language is subserved by
different neurological circuits than nonformulaic language. Specifically, recent
studies from neurologically impaired patients suffering from Alzheimer’s disease
and Parkinson’s disease point to the relevance of the right hemisphere (RH) sub-
cortical system in mediating the processing of formulaic language (Van Lancker
Sidtis et al., 2015), supporting the long-debated RH is special theory (Anaki, Faust,
& Kravetz, 1998; Bottini et al., 1994; Brownell, Potter, Michelow, & Gardner,
1984; Brownell, Simpson, Bihrle, & Gardner, 1990; Champagne, Virbel,
Nespoulous, & Joanette, 2003; Van Lancker & Kempler, 1987; Winner & Gardner,
1977). On this view, the RH is uniquely specialized for the processing of meta-
phorical language, as opposed to the left hemisphere (LH), which mainly pro-
cesses literal language. However, a large number of subsequent studies with
healthy and clinical populations failed to show the unique role of the RH in figura-
tive language processing, instead demonstrating contributions from both hemi-
spheres (Arzouan, Goldstein, & Faust, 2007; Bambini, Gentili, Ricciardi,
Bertinetto, & Pietrini, 2011; Bambini, Bertini, Schaeken, Stella, & DiRusso, 2016;
Kasher, Batori, Soroker, Graves, & Zaidel, 2002; Nocentini et al., 2006; Uchiyama
et al., 2012) or superior importance of the LH in the comprehension of idiomatic
and metaphorical expressions (Lee & Dapretto, 2006; Papagno & Genoni, 2004;
Papagno, Tabossi, Colombo, & Zampetti, 2004; Rapp, Leube, Erb, Grodd, &
Kircher, 2004; Rapp, Leube, Erb, Grodd, & Kircher, 2007; Stringaris, Medford,
Giampietro, Brammer, & David, 2007). To reconcile these inconsistencies, it has
been suggested that the differential hemispheric asymmetries found in some stud-
ies reflect differences in the type of linguistic material they employed rather than
the difference between literal and nonliteral language processing per se (for
98 A.B. Cieślicka
example, Kasparian, 2013). Also, a consensus has emerged suggesting qualita-

tively distinct profiles of semantic activation in the two cerebral hemispheres and
their differential sensitivity to salient and nonsalient meanings, which might thus
be responsible for the conflicting results obtained. These views are captured by the
fine/coarse coding theory (FCT; Beeman, 1998; Jung-Beeman, 2005) and the GSH
(Giora, 1997, 1999, 2002, 2003).
The FCT focuses on cerebral asymmetries in response to context and qualitative
differences in the type of semantic processing the two hemispheres employ. While
the left hemisphere engages in a fine coding (i.e., strong activation of focused
semantic fields directly related to the input stimulus), the right hemisphere performs
a coarse coding, in that it receives input from more distantly related sources. Coarse
coding results in activation of large and diffuse semantic fields including informa-
tion only remotely related to the input being processed. Such a diffuse semantic
activation of coarsely related meanings is particularly suitable for processing com-
plex language stimuli, such as figurative expressions where distinct meanings have
to be maintained and considered for selection. Given the importance of the RH for
understanding more distant semantic relations, Faust et al. (2012) suggested that
difficulties experienced by nonnative language users with achieving full mastery of
L2 figurative competence might stem from a reduced ability of the RH to engage in
coarse semantic coding in a nonnative language (p. 229). In order to test for this
possibility, Faust et al. (2012) conducted a split visual presentation experiment with
Hebrew participants highly proficient in English who were presented with pairs of
either Hebrew or English targets words. The pairs were either strongly related (cat-
egorical associates such as palace-castle) or weakly related (nonassociated category
members such as bridge-tower). Based on the previous research into semantic pro-
cessing differences between the left and right hemispheres, the categorical associate
pairs were expected to primarily evoke the strong, fine coding processing in the left
hemisphere, whereas the nonassociated category members were likely to evoke the
weak, coarse coding mainly initiated in the right hemisphere. Briefly, in the split
visual presentation (also called divided visual field) technique, stimuli are presented
to the right visual field/left hemisphere (rvf/LH) and left visual field/right hemi-
sphere (lvf/RH). Typically, centrally presented primes are followed by laterally pre-
sented targets, in which participants make word/nonword lexical decisions. Lateral
presentation of a stimulus allows capturing hemisphere-specific computations in the
initial stages of language processing (see Coulson & Van Petten, 2007). As predicted
for native language processing, Faust et al. (2012) found larger priming effects for
strongly-related pairs presented in the rvf/LH than in the lvf/RH and larger priming
effects for weakly-related pairs in the lvf/RH than in the rvf/LH, supporting the
FCT. In addition, consistent with the assumption that coarse semantic coding is
weaker for nonnative speaker processing, when Hebrew speakers were presented
with English word pairs, priming effects were only reported for strongly-related
pairs and were comparable in both hemispheres. Faust et al. (2012) suggested that
the reduced coarse coding for nonnative language processing might explain the dif-
ficulties that L2 users experience with appreciating more complex linguistic expres-
sions such as figurative tropes, which are inherently ambiguous and call for richer
and wider semantic activation in order to be successfully understood.
Hemispheric differences in processing semantic information and captured by
the FCT are compatible with predictions of the GSH (Giora, 1997, 2003), under
which the left and right hemisphere involvement are determined by salience of
the language material being processed. Given that the figurative meaning of
highly conventionalized expressions such as idioms, proverbs, or familiar meta-
phors is salient and coded in the mental lexicon, its processing should primarily
engage the LH, which stores linguistic knowledge. Processing nonsalient figura-
tive expressions, on the other hand, like novel (poetic) metaphors or nonconven-
tionalized types of figurative language like sarcasm or irony, should evoke
increased engagement of the RH in order to compute more complex semantic
information and distant relations between the elements of an unfamiliar figura-
tive expression. In keeping with these predictions, Giora, Zaidel, Soroker, Batori,
and Kasher (2000) found differential effects of right and left brain lesions on
understanding salient conventional metaphors and nonsalient sarcasm in RH
damaged and LH damaged patients. Jointly, the FCT and GSH predict that,
because of its fine coding the LH should quickly activate a small set of salient
meanings, whereas the RH should show a processing advantage for activating a
wide range of nonsalient literal and metaphorical meanings, the claim largely
supported in the monolingual literature (Faust & Mashal, 2007; Giora et al.,
2000; Mashal, Faust, & Hendler, 2005; Mashal, Faust, Hendler, & Jung-Beeman,
2007; Oliveri, Romero, & Papagno, 2004; Schmidt, DeBuse, & Seger, 2007;
Stringaris et al., 2006). Do these predictions apply also to nonnative language
users? While the bilingual studies addressing this question are very few, they do
seem to suggest that cerebral asymmetries are present in bilingual figurative
processing as well, although they are not entirely consistent with the native
speaker processing patterns.
In a divided visual field study with Polish–English bilinguals, Cieślicka and
Heredia (2011) tested the possibility that patterns of literal and figurative facilitation
in both hemispheres might be different for nonnative figurative language process-
ing. Specifically, given that figurative meanings of L2 idiomatic expressions are
likely to be less well coded in the mental lexicon and thus less salient, they should
be processed mainly by the right hemisphere. In contrast, an rvf/LH advantage
should be obtained for processing literal meanings of L2 idioms, since they enjoy a
special salience status in nonnative figurative processing (cf. Cieślicka, 2006a,
2006b). Participants were presented with literally plausible Polish and English idi-
oms which were embedded in neutral (Quite typically, Matt had cold feet) or
figuratively-biasing sentences (Before the wedding, Matt had cold feet), followed by
target words for a lexical decision displayed in the rvf/LH or lvf/RH and related
either to the figurative (e.g., AFRAID) or literal (e.g., SOCKS) meaning of the
idiom. Targets were presented at 0, 300, or 800 ms after the end of the sentence. The
study showed predominance of literal meaning activation for nonnative processing,
regardless of the visual field, with literal-related targets primed in both hemispheres
and throughout the entire time-course, except at 300 ms where they were only facili-
100 A.B. Cieślicka
tated in the LH. In turn, figurative priming occurred solely in the lvf/RH and only at
the 300 ms time window, clearly pointing to the difficulty of processing figurative
meanings of L2 idioms. This delayed activation of L2 figurative meanings was true
even for idioms embedded in rich, figurative-biasing context. In contrast, figurative
meanings of L1 idiomatic expressions were significantly primed at idiom offset (at
0 ms), indicating the ease with which formulaic language is accessed in L1 and
providing support for the direct access view. As demonstrated in the metaphoric
reference eye-tracking study discussed earlier (e.g., Heredia & Cieślicka, 2016),
also here literal meanings for L1 idioms were still active after 800 ms, in line with
the retention hypothesis (Giora, 2002). The major finding in the study was that L2
figurative processing might mainly engage the RH on account of the challenge that
figurative expressions pose for nonnative language users.
Further support for the idea that figurative processing in nonnative speakers
mainly engages the RH was demonstrated by Mashal, Borodkin, Maliniak, and
Faust (2015), who looked into the processing of conventional metaphors by native
speakers of Hebrew and English–Hebrew bilinguals. Participants were presented
with pairs of words that constituted either literal (hunger strike), conventional meta-
phorical (sweet revenge) or unrelated phrases (coral exam) in the divided visual
field study, such that the first word in a pair was presented centrally and the second
laterally. Participants’ task was to decide if the words formed a meaningful phrase.
The task was performed both in English and Hebrew, with Hebrew both as partici-
pants’ L1 and L2, so as to allow within-language comparisons where the status of
the language differed (i.e., L1 vs. L2). In line with Mashal et al.’s predictions that
conventional L1 metaphors are mainly processed in the LH and L2 metaphorical
expressions will mainly engage the RH, Mashal et al. (2015) found faster responses
for L1 metaphorical words presented to the rvf/LH than lvf/RH. This finding
supports the GSH, under which L1 figurative meanings of conventional metaphors
are strongly coded in the lexicon and highly salient, hence mainly relying on the LH
activation. The opposite pattern was found for L2 metaphorical processing (i.e., an
lvf/RH advantage for L2 metaphorical pairs, supporting the view that metaphoric
meanings of L2 conventional metaphors were less salient and called for interpreta-
tion processes and coarse semantic coding which is the domain of the RH). In addi-
tion, L2 literal word pairs were responded to faster than metaphorical pairs when
presented in the rvf/LH, in line with the assumption that literal meanings of L2
metaphorical expressions are more salient and automatically available. Since they
get activated automatically on account of their salience, they need to be suppressed
in order for the metaphorical meaning to emerge. Suppression of the inappropriate
literal meaning incurs an extra processing cost, which is mainly reflected in the left
hemisphere (cf. Fogliata et al., 2007).
In sum, the limited existing studies into the neurological underpinnings of bilin-
gual figurative processing seem to provide support for Faust et al.’s (2012) hypoth-
esis that L2 users’ difficulty in figurative comprehension might be attributed to the
reduced ability of the RH to perform coarse semantic coding in a nonnative lan-
guage. As the L2 expressions are less well established in the lexicon and their figu-
rative meaning has to be computed rather than directly retrieved, they pose an extra
processing load on the RH, which is mainly engaged in comprehending less salient
and more complex semantic relations. This extra cost, coupled with the RH reduced
ability to successfully coarse code in nonnative semantic processing, results in
slower and more effortful comprehension of L2 figurative expressions. In addition,
since literal meanings of L2 figurative expressions are more salient, they become
quickly available and need to be subsequently inhibited as contextually irrelevant,
thus posing an additional challenge for the LH, which has been implicated in sup-
pressing incompatible ambiguous meanings.
L1 Effects in L2 Processing
Another essential determinant of bilingual figurative processing concerns the

degree of cross-language overlap or similarity across languages. While similar fig-
urative expressions have an identical meaning and a close word-for-word corre-
spondence across languages, different ones express a similar concept but vary
substantially in their lexical makeup. Research with second language learners has
shown that lexically and conceptually similar figurative expressions are easier to
comprehend than different ones (e.g., Charteris-Black, 2002; Cieślicka, 2006a;
Deignan, Gabryś, & Solska, 1997; Irujo, 1986; Liontas, 2002; Türker, 2016). As
similar figurative expressions are perceived to be more transparent, a number of
studies in L2 figurative processing have indeed shown a facilitating influence of
transparency on recognition and production (e.g., Irujo, 1986, 1993; Steinel,
Hulstijn, & Steinel, 2007; Yorio, 1989).
The role of L1 knowledge in L2 figurative processing was demonstrated in the
study by Harris, Tebbe, Leka, Garcia, and Erramouspe (1999), who looked at bilin-
gual memory for metaphors. They presented participants with concrete (Playful
monkeys are clowns) and abstract (Shameless ignorance is soft clay) metaphors and
similes. Following the auditory presentation, participants were given a cued recall
task. Results showed many language switches in the participants’ responses, that is,
cases where bilinguals recalled the originally presented figurative expression but
switched to another language. While Harris et al. (1999) suggest that bilingual par-
ticipants were put into the bilingual mode through the instructions and stimulus
materials which were presented in both languages (e.g., Grosjean, 1997), the study
clearly shows that L1 equivalents of L2 figurative expressions become activated in
the course of their processing.
A more recent study into the effects of cross-language similarity on bilingual
figurative comprehension manipulated whether the idiom-final word was code-
switched (translated to the participants’ L2) or intact (Titone et al., 2015). English–
French bilinguals were asked to perform a meaningfulness judgment task and
decide, as quickly as possible, if the presented sentences containing code-switched
(e.g., He played with feu (fire) and intact (He played with fire) idioms were mean-
ingful. The idioms varied in terms of cross-language similarity such that some idi-
oms had identical meaning and word-for-word translation overlap, or identical
102 A.B. Cieślicka
meaning and only one or two words overlapping, the same meaning with no shared
component words, or no equivalent meaning altogether. Results showed that
response times and accuracy were more affected by code-switches on idiom-final
words than on control literal sentences, indicating that a change in the idiom’s final
word might have interfered with the direct retrieval of the idiom. Moreover, partici-
pants’ response times became progressively faster in proportion to the increasing
degree of cross-language overlap, and this facilitation was true regardless of
whether the idiom was intact or code-switched. Titone et al. (2015) interpret those
results as supporting the constraint-based model (Libben & Titone, 2008) under
which factors such as cross-language similarity enhance the ease of L2 idiom pro-
cessing by encouraging direct retrieval of the idiom from the lexicon. Specifically,
a code-switched word in idioms with high cross-language overlap might act as a
cue for activating the nontarget language and allowing for faster retrieval of a holis-
tically stored idiomatic form.
A number of studies (e.g., Carrol & Conklin, 2014, 2015; Carrol, Conklin, &
Gyllstad, 2016) have employed a novel design to look into cross-linguistic interac-
tions in the course of processing multi-word idiomatic expressions by nonnative
language users. An ideal litmus test for exploring the role of L1 in L2 idiom process-
ing would be to see if presentation of an L1-translated idiom facilitates the process-
ing of the otherwise unknown idiom in L2. In this hypothetical scenario, for example,
a Polish learner of English is presented with the English translation of the Polish
idiom (Wiercić komuś dziurę w brzuchu: To drill a hole in somebody’s stomach,
meaning to nag somebody to do something). The question of interest is whether the
non-idiom in English drill a hole in somebody’s stomach will be treated and recog-
nized as an idiom by Polish–English bilinguals, simply because it is a utomatically
transferred from their native language and recognized as a multi-word expression in
their L2 on account of form congruency, or whether it will be accepted only if the
equivalent L2 form exists and is part of the nonnative speaker’s L2 linguistic reper-
toire. To answer these questions, Carrol and Conklin (2014) examined the eye-track-
ing data collected while their intermediate proficiency Chinese speakers of English
and English monolingual participants read English translations of Chinese idioms
(draw a snake and add feet) that clearly had no English counterpart and compared
the idiom reading times to control items, such as draw a snake an add hair, where
the final word had been changed. In addition to Chinese-translated idioms, the par-
ticipants saw the English idioms (e.g., spill the beans) and their control phrases
where the last word had been changed (e.g., spill the chips). While both the idiomatic
and control sentences were presented in rich figurative biasing context, without
knowledge of the Chinese idiom the English translations would make no sense and
would not be processed as an idiom. Based on this logic, a shorter reading time for
the last word of the translated Chinese idioms than for control phrases would be
indicative of cross-language transfer and the fact that Chinese–English participants
were treating Chinese-based idioms as legitimate phrases in English. Overall, the
findings showed processing advantage for English idioms over control phrases in the
group of English monolinguals. In contrast, the Chinese–English group showed no
processing advantage for English idioms over control phrases, which supports findings
from the earlier studies suggesting no advantage for nonnative speakers in the course
of processing formulaic sequences in L2. In addition, Chinese native speakers were
faster processing translated Chinese idioms than matched controls, the effect
obtained particularly in early reading measures, such as first fixation duration, and
suggestive of the presence of between-language activation effects.
A second experiment further investigated this question by employing literally
plausible idioms which were again either original English idioms (e.g., a piece of
cake) or Chinese-translated idioms (e.g., add oil and vinegar: embellish a story)
nonexistent in English. The English idioms were embedded in either the figurative
biasing (e.g., One of my hobbies is doing little jobs around the house. I find most
things I try are a piece of cake if you make sure you have the right tools before you
start) or literal basing context (e.g., Yesterday I was in the canteen at work and I was
very hungry. I really wanted to get a piece of cake for my lunch but I was good and
just had a sandwich). Likewise, Chinese-translated idioms were embedded in figu-
rative biasing (e.g., I have a friend who always exaggerates whenever he tells sto-
ries. The problem is he tends to add oil and vinegar so it’s hard to know whether or
not to believe what he says) or literal biasing context (e.g., I read a really simple
recipe for a salad dressing. You just chop up some garlic and then add oil and vin-
egar then you put it in the fridge until you need to use it). Results for native speakers
of English found equal reading times for English idioms, regardless of context, sug-
gesting availability of both literal and figurative meanings of well-known literally
plausible idioms, consistent with the parallel-processing model. On the other hand,
Chinese-translated idioms, which were obviously unfamiliar to native speakers of
English, were read significantly slower in figurative context, implying the difficulty
in understanding an expression that lacked a figurative concept representation.
Chinese native speakers read literal versions of English idiomatic phrases faster
than they did the figurative interpretations, providing support for literal salience in
the course of processing L2 idiomatic language. Interestingly, this literal over figu-
rative priority extended also to Chinese-translated idioms suggesting that, despite
cross-linguistic similarity, the figurative meaning of Chinese-based English idioms
was not automatically retrieved. Overall, there seems to be some degree of cross-
linguistic activation that allows priming at the lexical level; however, this priming
might not be sufficient to facilitate the retrieval of the idiom’s figurative meaning.
The overall results are explained by reference to the GSH. In line with this approach,
the noncanonical, translated Chinese form might have been less salient and there-
fore less easily accessible for direct retrieval.
In a subsequent study using the same rationale and comparable population,
Carrol and Conklin (2015) presented native English speakers and Chinese speakers
of English with initial fragments of either English idioms (spill the….beans) or
transliterated Chinese idioms (draw a snake and add…feet). The initial idiom frag-
ments served as a prime whereas the last idiom word was used as a target for the
lexical decision. Control phrases were prepared for both English (spill the chips)
and translated Chinese idioms (draw a snake and add hair), where the last word for
the lexical decision was carefully matched with the last word in respective idioms.
Results showed that native speakers of English processed English idioms faster
104 A.B. Cieślicka
than their matched controls, thus supporting previous research that demonstrated
native speaker advantage for processing formulaic over nonformulaic sequences.
This advantage did not extend to translated Chinese idioms, which were obviously
unfamiliar and so could not be recognized as formulaic sequences. In contrast,
Chinese–English bilinguals showed no difference in processing English idioms as
compared to their matched controls, again consistent with the previous research
showing lack of nonnative speaker advantage for processing L2 formulas. However,
Chinese-translated idioms were recognized faster than matched controls, even if
they were presented in an unfamiliar format (i.e., in English). The fact that Chinese
speakers responded faster to transliterated Chinese idioms is a very strong indica-
tor of cross-language priming and automatic activation of lexical-level translation
equivalents. The authors suggest that in addition to, or in place of the lexical-level
cross-language activation, there might have been a conceptual-level activation,
whereby the conceptual representation of the idiom meaning got activated through
its component parts. While the design of the study did not allow distinguishing
between the two options, it did demonstrate a robust role of cross-language interac-
tions in the course of bilingual figurative language processing.
Further demonstration of the substantial role that L1 plays in processing L2 figu-
rative language comes from a recent eye-tracking study by Carrol et al. (2016). In
the experiment, advanced Swedish speakers of English were presented with three
types of idioms: L2-only (English idioms with no Swedish equivalents), L1-only
(Swedish idioms translated into English), and idioms in English which had an iden-
tical or very similar form and meaning in both English and Swedish. The idioms
were matched with their literal control phrases where the first content word was
changed (break the ice vs. crack the ice). Participants’ eye movements were recorded
both for the whole idiom and for the last idiom word, as they read the idioms embed-
ded in short neutral contexts. Of interest was whether L1 knowledge would be acti-
vated in the course of online L2 idiom processing and whether idiom congruency
across languages would facilitate L2 idiom processing. In addition, since the study
employed advanced speakers of English, one purpose of the study was to address
the specific question of the privileged status of formulaic over nonformulaic pro-
cessing; that is, would high proficiency of the participants make a difference and
allow them to process L2 idioms faster? Overall, Swedish speakers showed process-
ing advantage for all three types of idioms over control literal phrases, suggesting
that advanced nonnative language users can indeed process idiomatic language
comparably to native speakers and faster than nonformulaic language. More impor-
tantly, late reading measures (such as total reading time and regressions) showed
faster processing of familiar Swedish-translated idioms than less familiar Swedish-
based idioms, suggesting that L1 knowledge did become activated in the course of
L2 figurative processing. Likewise, congruent idioms were processed faster by
Swedish participants if their familiarity in Swedish, rather than English was higher,
again implying that L1 knowledge is the main driving force of L2 idiom processing.
As expected, English speakers showed advantage for English idioms, regardless of
congruency, and longer processing times for transliterated Swedish idioms with
which they were obviously not familiar.
The activation of L1 knowledge in L2 figurative processing has also been dem-

onstrated in the ERP literature. In one study Moreno, Federmeier, and Kutas (2002)
measured brain responses as Spanish–English bilinguals varying in language domi-
nance read moderately constraining sentences (e.g., She put on her high-heeled…)
or highly constraining formulaic sentences (e.g., Out of sight out of…). The last
word was either the expected ending (shoes, mind), a code-switch (i.e., a translation
of the expected ending: zapatos, mente), or a within-language lexical switch (i.e., a
word related semantically to the expected ending: boots, brain). Of interest was
whether code-switches would elicit compatible amplitude of brain electrophysio-
logical responses as lexical switches. Consistent with previous findings, lexical
switches elicited an N400 response, both for idiomatic and non-idiomatic sentences,
suggesting semantic difficulty of integrating the unexpected word. In contrast, code-
switches did not elicit an increased N400 in formulaic sentences, suggesting that
processing an unexpected translation equivalent of an anticipated idiom-final word
is less challenging for the language comprehension system than processing an unex-
pected within-language synonym of an anticipated idiom completion.
While studies into L2 proverb processing are conspicuously lacking in the bilingual
figurative literature, an important insight into cross-language influences in memory for
L2 proverbs has been provided by Vaid and Martinez (2001). Spanish–English bilin-
guals were shown Spanish and English proverbs varying in terms of familiarity and
asked to either paraphrase or translate them. An incidental recognition task was admin-
istered, where both new and old (previously presented) proverbs were presented for the
old vs. new judgment. Half of the old proverbs were presented in the same language as
in the translation/paraphrase stage of the e xperiment and half in the other language. Of
issue was whether bilingual participants would mistakenly consider as old the prov-
erbs they had seen in another input language in the encoding phase of the experiment.
Such translation errors would point to the possibility that proverbs are analyzed in a
literal fashion and susceptible to cross-language interactions, rather than being holisti-
cally coded in the mental lexicon. The results showed a complex interplay of language
status and salience of the proverb as important factors modulating cross-language acti-
vation. Specifically, highly familiar proverbs presented in English (the participants’
dominant language) that were used for the translation task were recalled more cor-
rectly. In addition, low familiar proverbs originally presented in the bilinguals’ less
dominant language were remembered better. Clearly, in line with previous studies, L1
effects in L2 figurative processing are influenced by a number of factors having to do
both with the experimental task at hand and the nature of the linguistic material
employed. For example, in an eye-tracking experiment with Spanish–English bilin-
guals, Cieślicka and Heredia (2013) showed that the facilitatory effect of cross-lan-
guage similarity in L2 idiom processing might be affected by such idiom characteristics
as their transparency. In the study, idioms with a high L1-L2 overlap were actually
processed slower than different idioms, but those results were strongly modulated by
the degree of idiom transparency, such that opaque different idioms were easier to pro-
cess than opaque similar ones, whereas the reverse was true for transparent idioms,
with transparent different idioms taking longer to process to transparent similar ones.
106 A.B. Cieślicka
Summary and Conclusions
This chapter has attempted to provide state-of-the-art review of the ongoing

research in bilingual figurative processing. It started with an overview of the vari-
ous types of figurative language and the difficulties they pose for the language
processing mechanism due to their inherent ambiguity. The major models devel-
oped to account for L1 figurative processing were discussed next because of their
importance and wide-ranging influence in subsequent bilingual research. Four
main themes have emerged in bilingual figurative research. The first has to do with
the storage and retrieval format of figurative expressions and is generally referred
to as the holistic hypothesis. This research focuses on whether bilinguals store and
retrieve figurative expressions holistically and process them faster than control
literal sentences, and it has been motivated by the debate present in the monolin-
gual literature between supporters of the figurative-first view, on the one hand, and
parallel-processing view on the other. While monolingual-based research shows
pretty consistently that native language speakers enjoy a processing advantage for
formulaic over nonformulaic expressions, L2 research has provided mixed results.
The second major theme concerns priority of literal vs. figurative meaning in the
course of processing L2 figurative language. This research is mainly motivated by
the GSH developed in the monolingual literature and postulating faster retrieval of
salient (i.e., frequent, familiar, and conventional) meanings. The results obtained
so far from the very limited L2 figurative studies are mixed and seem to paint a
much more complex picture, where a variety of modulating factors need to be
taken into account. The third question has been motivated by monolingual cerebral
asymmetry research inspired by the fine coarse coding theory and the GSH. These
theories suggest semantic processing differences between the left and right hemi-
spheres, and their major tenets have been extended to account for nonnative pro-
cessing as well. While bilingual research in this domain is almost nonexistent, the
few studies available to date do suggest a differential cerebral involvement in the
course of L2 figurative language processing. Finally, the fourth major research
theme addresses the role of L1 knowledge in the course of bilingual figurative
processing. Here all the research so far seems to unanimously agree that L1 knowl-
edge plays a significant role, but more details are pending on when this role is
facilitatory and when inhibitory.
Overall, a few important conclusions can be drawn from this overview. First of
all, it appears unquestionable that nonnative speakers process figurative language
with more effort and possibly differently than native speakers, at least at some stage
of their L2 metaphorical competency. Second, a number of factors are likely to
influence the nature of this processing. One of them has to do with the type of figu-
rative expression. Highly fixed and short formulaic phrases such as binomials (thick
and thin), collocations (make peace), or phrasal verbs (look up) are likely to be
learned as rote phrases and retrieved as such from the mental lexicon. They might
be lexicalized faster and treated holistically, simply because trying to analyze them
compositionally is counterintuitive and defies cognitive economy, in line with
Kellerman’s (1983) transfer studies that showed unwillingness of Dutch learners of

English to infer metaphorical meanings using literal- and L1-based processing strat-
egies. In turn, longer figurative expressions such as idioms, metaphors and proverbs,
as well as figurative tropes whose interpretation calls for considering extralinguistic
context (irony, metonymy, sarcasm) might benefit from being processed composi-
tionally, and so their holistic retrieval is not necessarily a default L2 processing
strategy. Another important factor determining the storage and processing format of
L2 figurative language is the degree of L2 proficiency for L2 learners and degree of
dominance for bilingual speakers. While less proficient L2 learners and nondomi-
nant bilinguals are likely to mainly rely on literal analysis for L2 figurative process-
ing, with increasing proficiency, they gradually switch to the figurative-processing
strategy as figurative meaning becomes more salient. As mentioned earlier, this shift
should be viewed as a gradual micro-level process happening individually for each
and every figurative expression that becomes incorporated into the L2 user’s lin-
guistic repertoire, rather than as a global processing change affecting the entire
mental lexicon. Finally, the type of the experimental task and stimulus materials
used are likely to influence how nonnative language users process figurative lan-
guage. Studies explicitly referring to L1 knowledge, for example by using transliter-
ated L1 idioms presented in L2 or code-switches might encourage more
cross-language influences than studies conducted in a monolingual mode where no
reference to the learners’ L1 knowledge is explicitly made.
In view of all of those modulating factors, it might seem a daunting task to try
and fit the current findings into any of the existing figurative processing models. It
appears though that most research so far, at least with reference to formulaic expres-
sions such as idioms, fits the framework of the parallel processing view and the
hybrid models which view idioms simultaneously as compositional and noncompo-
sitional phrases. More broadly, following Carrol and Conklin (2014), the current L2
figurative data are consistent with the dual route model of language use (Van
Lancker Sidtis, 2004, 2012; Van Lancker Sidtis et al., 2015), which emphasizes
parallel availability of two processing routes: direct, for holistic retrieval of formu-
laic expressions, and compositional or analytic, employed for processing novel sen-
tences. Which of them will be relied on more heavily is contingent upon a number
of factors, the most important of which appears to be the language user’s familiarity
with the formulaic sequence. While both routes might be activated when processing
idioms by native and nonnative language users, the computation of the idiom’s figu-
rative meaning will be faster for native speakers, as L1 idioms are salient and highly
automatized in their mental lexicon. This does not preclude the possibility that lit-
eral meanings of idiom component words also become automatically available, but
they do not need to be relied on in order to understand the figurative meaning of the
idiom. Here individual differences between figurative tropes, such as their length or
degree of their literalness, transparency and compositionality might come to play a
role, making one route more accessible than another.
In turn, L2 idiom processing is mainly a function of learner’s familiarity with a
given L2 expression. With less familiar phrases, the learner will primarily rely on
108 A.B. Cieślicka
the compositional analysis, as reflected by studies showing primacy of literal over

figurative activation in the course of L2 formulaic processing. The more frequently
the L2 learner uses a given formulaic sequence, the more likely it is to become
automatized as a configuration of strongly related lexical items and hence retrieved
faster. This view hence posits no fundamental qualitative differences between the
way native and nonnative speakers process formulaic phrases. Both employ the
dual route, but reliance on one or another varies as a function of familiarity or
salience of the literal vs. figurative meaning and is possibly modulated by other
factors, such as characteristics of a specific formulaic phrase or type of the lan-
guage task performed. Given that L2 speakers’ figurative competence emerges
slowly, the compositional processing route is likely to be relied on more heavily.
Ultimately, with increasing proficiency, nonnative users will rely more on retriev-
ing the formulaic language directly from the mental lexicon, instead of analyzing it
compositionally, which will ensure cognitive economy and processing efficiency.
As argued by Christiansen and Chater (2016), given the severe limitations of
human memory, which they refer to as the now-or-never bottleneck, the language
system must rely on processing sequential information most efficiently so as to
avoid being deluged by subsequent incoming stimuli. This is possible by the chunk-
and-pass processing strategy under which the incoming material is rapidly pro-
cessed as overlearned and familiar chunks. Chunking formulaic sequences might
thus be viewed as part and parcel of the universal processing strategy employed by
the language processor to ensure efficient communication and overcome memory
overload.
To conclude, it seems that bilingual figurative research is still in its infancy and
considerably more work is needed to properly address some of the questions that
have been raised so far. In addition, it is apparent from the current overview that
some figurative tropes such as formulaic phrases (idioms, phrasal verbs) have been
consistently preferred in bilingual figurative research than others (proverbs, meta-
phor, metonymy, irony). In fact, studies into the L2 processing of the latter are
almost nonexistent. Until we know more about how bilinguals process these differ-
ent figurative tropes, any generalizations concerning bilingual figurative processing
need to be treated cautiously. What is the future of bilingual figurative language
research? One possibility is to broaden the scope of inquiry and borrow from the
disciplines of computational linguistics and natural language processing (e.g.,
Bortfeld, 2017) which offer such solutions as probabilistic approaches to modeling
learning. Bortfeld points out that the natural language processing framework might
help bridge the literal-figurative divide, as it considers such factors as communica-
tive goals, intention, context, and most importantly, the degree of shared knowledge
between speakers. This is a promising new path for bilingual figurative research and
it seems that, as long as we consider new ways of addressing old questions, we are
making steps in the right direction.
References
Abel, B. (2003). English idioms in the first language and second language lexicon: A dual repre-
sentation approach. Second Language Research, 19, 329–358.
Anaki, D., Faust, M., & Kravetz, S. (1998). Cerebral hemispheric asymmetries in processing lexi-
cal metaphors. Neuropsychologia, 36(7), 691–700.
Arzouan, Y., Goldstein, A., & Faust, M. (2007). Dynamics of hemispheric activity during metaphor
comprehension: Electrophysiological measures. NeuroImage, 36, 222–231.
Bambini, V., Bertini, C., Schaeken, W., Stella, A., & DiRusso, F. (2016). Disentangling metaphor
from context: An ERP study. Frontiers in Psychology, 7, 559. doi:10.3389/fpsyg.2016.00559
Bambini, V., Gentili, C., Ricciardi, E., Bertinetto, P. M., & Pietrini, P. (2011). Decomposing meta-
phor processing at the cognitive and neural level through functional magnetic resonance imag-
ing. Brain Research Bulletin, 86, 203–216. doi:10.1016/j.brainresbull.2011.07.015
Beeman, M. (1998). Coarse semantic coding and discourse comprehension. In M. Beeman &
C. Chiarello (Eds.), Right hemisphere language comprehension: Perspectives from cognitive
neuroscience (pp. 255–284). Mahwah, NJ: Lawrence Erlbaum Associates.
Blank, G. D. (1988). Metaphors in the lexicon. Metaphor and Symbolic Activity, 3(1), 21–36.
Bobrow, S. A., & Bell, S. M. (1973). On catching on to idiomatic expressions. Memory and
Cognition, 1, 342–346.
Bock, K., & Brewer, W. F. (1980). Comprehension and memory of the literal and figurative mean-
ing of proverbs. Journal of Psycholinguistic Research, 9(1), 59–71.
Boers, F., Demecheleer, M., & Eyckmans, J. (2004). Cultural variation as a variable in compre-
hending and remembering figurative idioms. European Journal of English Studies, 8, 375–388.
Boers, F., Eyckmans, J., & Stengers, H. (2007). Presenting figurative idioms with a touch of ety-
mology: More than mere mnemonics? Language Teaching Research, 11, 43–62.
Boers, F., & Lindstromberg, S. (Eds.). (2008). Cognitive linguistic approaches to teaching vocabu-
lary and phraseology. Berlin: Mouton de Gruyter.
Bortfeld, H. (2003). Comprehending idioms cross-linguistically. Experimental Psychology, 50(3),
217–230.
Bortfeld, H. (2017). Is Figurative Language the Final Frontier or a Pit Stop Along the Way?:
Bilingual Figurative Language Processing by Roberto R. Heredia and Anna B. Cieślicka.The
American Journal of Psychology, 130(2), 254–258.
Bottini, G., Corcoran, R., Sterzi, R., Paulesu, E., Schenone, P., Scarpa, P., et al. (1994). The role of
the right hemisphere in the interpretation of figurative aspects of language: A positron emission
tomography activation study. Brain, 117(6), 1241–1253.
Boulenger, V., Shtyrov, Y., & Pulvermüller, F. (2012). When do you grasp the idea? MEG evidence
for instantaneous idiom understanding. NeuroImage, 59, 3502–3513.
Brannon L. L. (1975). On the understanding of idiomatic expressions. Unpublished doctoral dis-
sertation. University of Texas, Austin.
Bromberek-Dyzman, K. (2015). Irony processing in L1 and L2: Same or different? In R. R.
Heredia & A. B. Cieślicka (Eds.), Bilingual figurative language processing (pp. 268–297).
New York: Cambridge University Press.
Bromberek-Dyzman, K., Rataj, K., & Dylak, J. (2010). Mentalizing in the second language: Is
irony online inferencing any different in L1 and L2? In I. Witczak-Plisiecka (Ed.), Pragmatic
perspectives on language and linguistics:Vol. 1. Speech actions in theory and applied studies
(pp. 197–216). Newcastle: Cambridge Scholars.
Brownell, H. H., Potter, H. H., Michelow, D., & Gardner, H. (1984). Sensitivity to lexical denota-
tion and connotation in brain-damaged patients: A double dissociation? Brain and Language,
22(2), 253–265.
Brownell, H. H., Simpson, T. L., Bihrle, A. M., & Gardner, H. (1990). Appreciation of metaphoric
alternative word meanings by left and right brain-damaged patients. Neuropsychologia, 28(4),
375–383.
110 A.B. Cieślicka
Cacciari, C., & Corradini, P. (2015). Literal analysis and idiom retrieval in ambiguous idioms pro-
cessing: A reading-time study. Journal of Cognitive Psychology, 27(7), 797–811.
Cacciari, C., & Glucksberg, S. (1991). Understanding idiomatic expressions: The contribution of
word meanings. In G. B. Simpson (Ed.), Understanding word and sentence (pp. 217–240).
Amsterdam: North-Holland.
Cacciari, C., & Tabossi, P. (1988). The comprehension of idioms. Journal of Memory and
Language, 27, 668–683.
Caillies, S., & Butcher, K. (2007). Processing of idiomatic expressions: Evidence for a new hybrid
view. Metaphor and Symbol, 22, 79–108.
Carrol, G., & Conklin, K. (2014). Eye-tracking multi-word units: Some methodological questions.
Journal of Eye Movement Research, 7(5), 5, 1–11.
Carrol, G., & Conklin, K. (2015). Cross language lexical priming extends to formulaic units:
Evidence from eyetracking suggests that this idea ‘has legs’. Bilingualism: Language and
Cognition. doi:10.1017/S1366728915000103
Carrol, G., Conklin, K., & Gyllstad, H. (2016). Found in translation: The influence of L1 on
the processing of idioms in L2. Studies in Second Language Acquisition. doi:10.1017/
S0272263115000492
Champagne, M., Virbel, J., Nespoulous, J.-L., & Joanette, Y. (2003). Impact of right hemi-
spheric damage on a hierarchy of complexity evidenced in young normal subjects. Brain and
Cognition, 53, 152–157.
Charteris-Black, J. (2002). Second language figurative proficiency: A comparative study of Malay
and English. Applied Linguistics, 23, 104–133.
Christiansen, M. H., & Chater, N. (2016). The Now-or-Never bottleneck: A fundamental constraint
on language. Behavioral and Brain Sciences, 39, e62. doi:10.1017/S0140525X1500031X
Cieślicka, A. (2006a). Literal salience in on-line processing of idiomatic expressions by L2 speak-
ers. Second Language Research, 22(2), 115–144.
Cieślicka, A. (2006b). On building castles on the sand, or exploring the issue of transfer in the
interpretation and production of L2 fixed expressions. In J. Arabski (Ed.), Cross-linguistic
influences in the second language lexicon (pp. 226–245). Clevedon: Multilingual Matters.
Cieślicka, A. (2007). Language experience and fixed expressions: Differences in the salience sta-
tus of literal and figurative meanings of L1 and L2 idioms. In M. Nenonen & S. Niemi (Eds.),
Collocations and Idioms 1: Papers from the First Nordic Conference on Syntactic Freezes,
Joensuu, Finland, May 19-20 (pp. 55–70). Joensuu: Joensuu University Press.
Cieślicka, A. (2010). Formulaic language in L2: Storage, retrieval and production of idioms by
second language learners. In M. Pütz & L. Sicola (Eds.), Cognitive processing in second lan-
guage acquisition: Inside the learner’s mind (pp. 149–168). Amsterdam/Philadelphia: John
Benjamins.
Cieślicka, A. B. (2015). Idiom acquisition and processing by second/foreign language learners.
In R. R. Heredia & A. B. Cieślicka (Eds.), Bilingual figurative language processing (pp. 208–
244). New York: Cambridge University Press.
Cieślicka, A. B., & Heredia, R. R. (2011). Hemispheric asymmetries in processing L1 and L2
idioms: Effects of salience and context. Brain and Language, 116, 136–150.
Cieślicka, A. B., & Heredia, R. R. (2013, May). The multiple determinants of eye movement pat-
terns in bilingual figurative processing. Paper presented at the 25th APS Annual Convention,
Washington, DC.
Cieślicka, A. B., Heredia, R. R., & Olivares, M. (2014). It’s all in the eyes: How language domi-
nance, salience, and context affect eye movements during idiomatic language processing. In
L. Aronin & M. Pawlak (Eds.), Essential topics in applied linguistics and multilingualism.
Studies in honor of David Singleton (pp. 21–42). Berlin: Springer.
Clark, H. H., & Lucy, P. (1975). Understanding what is said from what is meant: A study in con-
versationally conveyed requests. Journal of Verbal Learning and Verbal Behavior, 14, 56–72.
Colombo, L. (1993). The comprehension of ambiguous idioms in context. In C. Cacciari &
P. Tabossi (Eds.), Idioms: Processing, structure, and interpretation (pp. 163–200). Hillsdale,
NJ: Lawrence Erlbaum Associates.
Conklin, K., & Schmitt, N. (2008). Formulaic sequences: Are they processed more quickly than
nonformulaic language by native and non-native speakers? Applied Linguistics, 29, 72–89.
Conklin, K., & Schmitt, N. (2012). The processing of formulaic language. Annual Review of
Applied Linguistics, 32, 45–61. doi:10.1017/S0267190512000074
Cooper, T. C. (1999). Processing of idioms by L2 learners of English. TESOL Quarterly, 33(2),
233–262.
Coulson, S., & Van Petten, C. (2007). A special role for the right hemisphere in metaphor compre-
hension? ERP evidence from hemifield presentation. Brain Research, 1146, 128–145.
Cronk, B. C., Lima, S. D., & Schweigert, W. A. (1993). Idioms in sentences: Effects of frequency,
literalness, and familiarity. Journal of Psycholinguistic Research, 22(1), 59–81.
Cronk, B. C., & Schweigert, W. A. (1992). The comprehension of idioms: The effects of familiar-
ity, literalness, and usage. Applied PsychoLinguistics, 13, 131–146.
Cutter, M. G., Drieghe, D., & Liversedge, S. P. (2014). Preview benefit in English spaced com-
pounds. Journal of Experimental Psychology. Learning, Memory, and Cognition, 40(6), 1778–
1786. doi:10.1037/xlm0000013
Cutting, J. C., & Bock, K. (1997). That’s the way the cookie bounces: Syntactic and semantic com-
ponents of experimentally elicited idiom blends. Memory and Cognition, 25, 57–71.
Danesi, M. (1986). The role of metaphor in second language pedagogy. Rassegna Italiana Di
Linguistica Applicata, 18(3), 1–10.
Danesi, M. (1992). Metaphorical competence in second language acquisition and second language
learning: The neglected dimension. In J. Alatis (Ed.), Language, communication, and social
meaning (pp. 489–500). Washington, DC: Georgetown University Press.
De Grauwe, S., Swain, A., Holcomb, P. J., Ditman, T., & Kuperberg, G. R. (2010).
Electrophysiological insights into the processing of nominal metaphors. Neuropsychologia,
48(7), 1965–1984.
Deignan, A., Gabryś, D., & Solska, A. (1997). Teaching English metaphors using cross-linguistic
awareness-rasing activities. ELT Journal, 51, 352–360.
Erman, B., & Warren, B. (2000). The idiom principle and the open-choice principle. Text, 20, 29–62.
Faust, M., Ben-Artzi, E., & Vardi, N. (2012). Semantic processing in native and second lan-
guage: Evidence from hemispheric differences in fine and coarse semantic coding. Brain and
Language, 123, 228–233.
Faust, M., & Mashal, N. (2007). The role of the right cerebral hemisphere in processing novel
metaphoric expressions taken from poetry: A divided visual field study. Neuropsychologia, 45,
860–870.
Flores d’Arcais, G. B. (1993). The comprehension and semantic interpretation of idioms. In
C. Cacciari & P. Tabossi (Eds.), Idioms: Processing, structure, and interpretation (pp. 79–98).
Hillsdale, NJ: Lawrence Erlbaum Associates.
Fogliata, A., Rizzo, S., Reati, F., Miniussi, C., Oliveri, M., & Papagno, C. (2007). The time course
of idiom processing. Neuropsychologia, 45, 3215–3222.
Frisson, S., & Pickering, M. J. (1999). The processing of metonymy: Evidence from eye move-
ments. Journal of Experimental Psychology: Learning, Memory, and Cognition, 25(6),
1366–1383.
García, O., Cieślicka, A. B., & Heredia, R. R. (2015). Figurative language processing and meth-
odological considerations. In R. R. Heredia & A. B. Cieślicka (Eds.), Bilingual figurative lan-
guage processing (pp. 117–170). New York: Cambridge University Press.
Gentner, D., & Bowdle, B. F. (2001). Convention, form, and figurative language processing.
Metaphor and Symbol, 16(3&4), 223–247.
Gerrig, R., & Healy, A. (1983). Dual processes in metaphor understanding: Comprehension and
appreciation. Journal of Experimental Psychology: Learning, Memory, and Cognition, 9,
667–675.
Gibbs, R. W. (1980). Spilling the beans on understanding and memory for idioms in conversation.
Memory and Cognition, 8(2), 149–156.
Gibbs, R. W. (1981). Memory for requests in conversation. Journal of Verbal Learning and Verbal
Behavior, 20, 630–640.
112 A.B. Cieślicka
Gibbs, R. W. (1983). Do people always process the literal meanings of indirect requests? Journal
of Experimental Psychology: Learning, Memory, and Cognition, 9(3), 524–533.
Gibbs, R. W. (1985). On the process of understanding idioms. Journal of Psycholinguistic
Research, 14(5), 465–472.
Gibbs, R. W. (1986). On the psycholinguistics of sarcasm. Journal of Experimental Psychology:
General, 115(1), 3–15.
Gibbs, R. W. (1992). What do idioms really mean? Journal of Memory and Language, 31, 485–506.
Gibbs, R. W. (1994). The poetics of mind: Figurative thought, language, and understanding.
Gibbs, R. W. (1996). Why many concepts are metaphorical. Cognition, 61, 309–319.
Gibbs, R. W. (1998). The fight over metaphor in thought and language. In A. N. Katz, C. Cacciari,
R. W. Gibbs, & M. Turner (Eds.), Figurative language and thought (pp. 88–118). Oxford:
Oxford University Press.
Gibbs, R. W. (2001). Evaluating contemporary models of figurative language understanding.
Gibbs, R. W. (2002). A new look at literal meaning in understanding what is said and implicated.
Journal of Pragmatics, 34(4), 457–486.
Gibbs, R. W., & Nayak, N. P. (1989). Psycholinguistic studies on the syntactic behavior of idioms.
Cognitive Psychology, 21, 100–138.
Gibbs, R. W., Nayak, N. P., & Cutting, C. (1989). How to kick the bucket and not decompose:
Analyzability and idiom processing. Journal of Memory and Language, 28, 576–593.
Gildea, P., & Glucksberg, S. (1983). On understanding metaphor: The role of context. Journal of
Verbal Learning and Verbal Behavior, 22, 577–590.
Giora, R. (1991). On the cognitive aspects of joke. Journal of Pragmatics, 16, 465–485.
Giora, R. (1997). Understanding figurative and literal language: The graded salience hypothesis.
Cognitive Linguistics, 8(3), 183–206.
Giora, R. (1999). On the priority of salient meanings: Studies of literal and figurative language.
Giora, R. (2001). Foreword. Metaphor and Symbol, 15(3&4), 145–148.
Giora, R. (2002). Literal vs. figurative language. Different or equal? Journal of Pragmatics, 34(4),
487–506.
Giora, R. (2003). On our mind: Salience, context, and figurative language. Oxford: Oxford
University Press.
Giora, R., & Fein, O. (1999a). Irony, context and salience. Metaphor and Symbol, 14(4), 241–257.
Giora, R., & Fein, O. (1999b). Irony interpretation: The graded salience hypothesis. Humor, 12(4),
425–436.
Giora, R., & Fein, O. (1999c). On understanding familiar and less-familiar figurative language.
Giora, R., Fein, O., & Schwartz, T. (1998). Irony: Graded salience and indirect negation. Metaphor
and Symbol, 13, 83–101.
Giora, R., Zaidel, E., Soroker, N., Batori, G., & Kasher, A. (2000). Differential effects of right-
and left-hemisphere damage on understanding sarcasm and metaphor. Metaphor and Symbol,
15(1&2), 63–83.
Glass, A. L. (1983). The comprehension of idioms. Journal of Psycholinguistic Research, 12(4),
429–442.
Glucksberg, S. (1989). Metaphors in conversation: How are they understood? Why are they used?
Metaphor and Symbolic Activity, 4(3), 125–143.
Glucksberg, S. (1991). Beyond literal meanings: The psychology of allusion. Psychological
Science, 2(3), 146–152.
Glucksberg, S., Gildea, P., & Bookin, H. B. (1982). On understanding nonliteral speech: Can
people ignore metaphors? Journal of Verbal Learning and Verbal Behavior, 21, 85–98.
Glucksberg, S. (1993). Idiom meanings and allusional content. In C. Cacciari & P. Tabossi (Eds.),
Idioms: Processing, structure, and interpretation (pp. 3-26). Hillsdale, NJ: Lawrence Erlbaum
Associates.
Grice, H. P. (1975). Logic and conversation. In P. Cole & J. Morgan (Eds.), Syntax and semantics
3: Speech acts (pp. 41–58). New York: Academic.
Grosjean, F. (1997). Processing mixed language: Issues, findings, and models. In A. M. B. DeGroot
& J. F. Kroll (Eds.), Tutorials in bilingualism: Psycholinguistic perspectives (pp. 225–254).
Mahwah, NJ: Lawrence Erlbaum Associates.
Hamblin, J. L., & Gibbs, R. W. (1999). Why can't you kick the bucket as you slowly die: Verbs in
idiom comprehension. Journal of Psycholinguistic Research, 28(1), 25–39.
Harris, R. J. (1979). Memory for metaphors. Journal of Psycholinguistic Research, 8(1), 61–71.
Harris, R. J., Tebbe, M. R., Leka, G. E., Garcia, R. C., & Erramouspe, R. (1999). Monolingual and bilin-
gual memory for English and Spanish metaphors and similes. Metaphor and Symbol, 14(1), 1–16.
Heredia, R. R., & Cieślicka, A. B. (2014). Bilingual storage: Compound-coordinate and deriva-
tives. In R. R. Heredia & J. Altarriba (Eds.), Foundations of bilingual memory (pp. 11–40).
Berlin: Springer.
Heredia, R. R., & Cieślicka, A. B. (2016). Metaphoric reference: An eye movement analysis of
Spanish–English and English–Spanish bilingual readers. Frontiers in Psychology, 7, 439.
doi:10.3389/fpsyg.2016.00439
Heredia, R. R., & Muñoz, M. E. (2015). Metaphoric reference: A real-time analysis. In R. R.
Heredia & A. B. Cieślicka (Eds.), Bilingual figurative language processing (pp. 89–116).
Hillert, D., & Swinney, D. (2001). The processing of fixed expressions during sentence compre-
hension. In A. Cienki, B. J. Luka, & M. B. Smith (Eds.), Conceptual and discourse factors in
linguistic structure (pp. 107–121). Stanford, CA: CSLI.
Hoffman, R. (1980). Metaphor in science. In R. P. Honeck & R. R. Hoffman (Eds.), Cognition and
figurative language (pp. 393–418). Hillsdale, NJ: Lawrence Erlbaum Associates.
Howarth, P. (1998). The phraseology of learners’ academic writing. In A. Cowie (Ed.), Phraseology:
Theory, analysis and applications (pp. 161–186). Oxford: Oxford University Press.
Irujo, S. (1986). A piece of cake: Learning and teaching idioms. ELT Journal, 40, 236–242.
Irujo, S. (1993). Steering clear: Avoidance in the production of idioms. International Review of
Applied Linguistics, 21, 205–219.
Janus, R. A., & Bever, T. G. (1985). Processing of metaphoric language: An investigation of the
three-stage model of metaphor comprehension. Journal of Psycholinguistic Research, 14(5),
473–487.
Jiang, N., & Nekrasova, T. M. (2007). The processing of formulaic sequences by second language
speakers. Modern Language Journal, 91, 433–445.
Johnson-Laird, P. N. (1993). Foreword. In C. Cacciari & P. Tabossi (Eds.), Idioms: Processing,
structure, and interpretation (pp. VII–VIX). Hillsdale, NJ: Lawrence Erlbaum Associates.
Jung-Beeman, M. (2005). Bilateral brain processes for comprehending natural language. Trends in
Cognitive Sciences, 9(11), 512–518.
Kasher, A., Batori, G., Soroker, N., Graves, D., & Zaidel, E. (2002). Effects of right- and left-
hemisphere damage on understanding conversational implicatures. Brain and Language, 80(3),
510–535.
Kasparian, K. (2013). Hemispheric differences in figurative language processing: Contributions
of neuroimaging methods and challenges in reconciling current empirical findings. Journal of
Neurolinguistics, 26, 1–21.
Kathpalia, S., & Heah, L. (2011). Metaphorical competence in ESL student writing. RELC
Journal, 42(3), 273–290.
Katz, A., & Bowes, A. (2016). Embodiment in metaphor and (not?) in bilingual language. In
R. R. Heredia & A. B. Cieślicka (Eds.), Bilingual figurative language processing (pp. 3–27).
Kecskes, I. (2006). On my mind: Thoughts about salience, context and figurative language from a
second language perspective. Second Language Research, 22(2), 1–19.
Kecskes, I. (2016). Is the Idiom Principle blocked in bilingual L2 production? In R. R. Heredia
& A. B. Cieślicka (Eds.), Bilingual figurative language processing (pp. 28–52). New York:
Cambridge University Press.
114 A.B. Cieślicka
Kecskes, I., & Papp, T. (2000). Metaphorical competence in trilingual language production.
In J. Cenoz & U. Jessner (Eds.), English in Europe: The acquisition of a third language
(pp. 99–120). Clevedon: Multilingual Matters.
Kellerman, E. (1983). Now you see it, now you don’t. In S. Gass & L. Selinker (Eds.), Language
transfer in language learning (pp. 112–134). Rowley, MA: Newbury House.
Kemper, S. (1981). Comprehension and the interpretation of proverbs. Journal of Psycholinguistic
Research, 10(2), 179–198.
Keysar, B. (1989). On the functional equivalence of literal and metaphorical interpretations in
discourse. Journal of Memory and Language, 28, 375–385.
Konopka, A., & Bock, K. (2009). Lexical or syntactic control of sentence formulation? Structural
generalizations from idiom production. Cognitive Psychology, 58(1), 68–101.
Kövecses, Z., & Radden, G. (1998). Metonymy: Developing a cognitive linguistic view. Cognitive
Linguistics, 9(1), 37–78.
Kövecses, Z., Szelid, V., Nucz, E., Blanco-Carrión, O., Akkök, E. A., & Szabó, R. (2015).
Anger metaphors across languages: A cognitive linguistic perspective. In R. R. Heredia &
A. B. Cieślicka (Eds.), Bilingual figurative language processing (pp. 341–367). New York:
Kutas, M., & Delong, K. A. (2008). A sampler of event-related brain potential (ERP) analyses of
language processing. Brain Research in Language, 153–186.
Kutas, M., & Federmeier, K. D. (2000). Electrophysiology reveals semantic memory use in lan-
guage comprehension. Trends in Cognitive Sciences, 4(12), 463–470.
Kutas, M., & Federmeier, K. D. (2011). Thirty years and counting: Finding meaning in the
N400 component of the event-related brain potential (ERP). Annual Review of Psychology,
62, 621–647.
Kutas, M., & Hillyard, S. A. (1980a). Event-related potentials to semantically inappropriate and
surprisingly large words. Biological Psychology, 11, 99–116.
Kutas, M., & Hillyard, S. A. (1980b). Reading senseless sentences: Brain potentials reflect seman-
tic incongruity. Science, 207, 203–205.
Lai, V. T., Curran, T., & Menn, L. (2009). Comprehending conventional and novel metaphors: An
ERP study. Brain Research, 1284, 145–155.
Lakoff, G. (1987). Women, fire, and dangerous things. Chicago: Chicago University Press.
Lakoff, G. (1990). The invariance hypothesis: is abstract reasoning based on image-schemas?
Cognitive Linguistics, 1(1), 39–74.
Lakoff, G. (1993). The contemporary theory of metaphor. In A. Ortony (Ed.), Metaphor and
thought. New York: Cambridge University Press.
Lakoff, G. (2009). The neural theory of metaphor. Berkeley: University of California at Berkeley.
Lakoff, G., & Johnson, M. (1980). Metaphors we live by. Chicago: Chicago University Press.
Lakoff, G., & Turner, M. (1989). More than cool reason: A field guide to poetic metaphor. Chicago:
University of Chicago Press.
Laufer, B. (1989). A factor of difficulty in vocabulary learning: Deceptive transparency. AILA
Review, 6, 10–20.
Lazar, G. (1996). Using figurative language to expand students’ vocabulary. ELT Journal,
50, 43–51.
Lee, S. S., & Dapretto, M. (2006). Metaphorical vs. literal word meanings: fMRI evidence against
a selective role of the right hemisphere. NeuroImage, 29, 536–544.
Levin, S. R. (1979). Standard approaches to metaphor and a proposal for literary metaphor. In
A. Ortony (Ed.), Metaphor and thought (pp. 124–135). Cambridge: Cambridge University Press.
Libben, M. R., & Titone, D. A. (2008). The multidetermined nature of idiom processing. Memory
and Cognition, 36, 1103–1121.
Liontas, J. I. (2002). Context and idiom understanding in second languages. In S. H. Foster-
Cohen, T. Ruthenberg, & M.-L. Poschen (Eds.), EUROSLA Yearbook: Annual Conference
of the European Second Language Association (pp. 155–185). Amsterdam/Philadelphia:
John Benjamins.
Liontas, J. I. (2015). Straight from the horse’s mouth: Idiomaticity revisited. In R. R. Heredia &
A. B. Cieślicka (Eds.), Bilingual figurative language processing (pp. 301–340). New York:
Littlemore, J. (2001). Metaphoric competence: A language learning strength of students with a
holistic cognitive style? TESOL Quarterly, 35, 459–491.
Littlemore, J. (2010). Metaphoric competence in the first and second language: Similarities and
differences. In M. Pütz & L. Sicola (Eds.), Cognitive processing in second language acquisi-
tion: Inside the learner’s mind (pp. 293–316). Amsterdam/Philadelphia: John Benjamins.
Low, G. D. (1988). On teaching metaphor. Applied Linguistics, 9, 125–147.
Lowder, M. W., & Gordon, P. C. (2013). It’s hard to offend the college: Effects of sentence
structure on figurative-language processing. Journal of Experimental Psychology: Learning,
Memory, and Cognition, 39(4), 993–1011.
MacArthur, F. (2010). Metaphorical competence in EFL: Where are we and where should we be
going? A view from the language classroom. AILA Review, 23(19), 155–173.
Martinez, R., & Murphy, V. A. (2011). Effect of frequency and idiomaticity on second language
reading comprehension. TESOL Quarterly, 45, 267–290.
Mashal, N., Borodkin, K., Maliniak, O., & Faust, M. (2015). Hemispheric involvement in native
and non-native comprehension of conventional metaphors. Journal of Neurolinguistics, 35,
96–108.
Mashal, N., Faust, M., & Hendler, T. (2005). The role of the right hemisphere in processing non-
salient metaphorical meanings: Application of principle components analysis to fMRI data.
Neuropsychologia, 43, 2084–2100.
Mashal, N., Faust, M., Hendler, T., & Jung-Beeman, M. (2007). An fMRI investigation of the neu-
ral correlates underlying the processing of novel metaphoric expressions. Brain and Language,
100, 115–126.
Matlock, T., & Heredia, R. R. (2002). Understanding phrasal verbs in monolinguals and bilin-
guals. In R. R. Heredia & J. Altarriba (Eds.), Bilingual sentence processing (pp. 251–274).
Amsterdam: Elsevier.
McPartland-Fairman, P. (1989). The processing of phrasal verbs by native and nonnative speakers
of English. Unpublished doctoral dissertation. The City University of New York, NewYork.
Meara, P. (1980). Vocabulary acquisition: A neglected aspect of language learning. Language
Teaching and Linguistics Abstracts, 13, 221–246.
Moreno, E. M., Federmeier, K. D., & Kutas, M. (2002). Switching languages, switching palabras
(words): An electrophysiological study of code switching. Brain and Language, 80, 188–207.
Morgan, J. L. (1979). Observations on the pragmatics of metaphor. In A. Ortony (Ed.), Metaphor
and thought (pp. 136–147). Cambridge: Cambridge University Press.
Nocentini, U., Kahlaoui, K., Roberts, P., Eck, K., Giroux, F., Goulet, P., et al. (2006). Processing
of denotative and connotative alternative meaning of words after left- and right-hemispheric
lesion. Brain and Language, 99, 100–101.
Nunberg, G. (1978). The pragmatics of reference. Bloomington, IN: Indiana University Linguistics
Club.
Nunberg, G., Sag, I., & Wasow, T. (1994). Idioms. Language, 70, 491–538.
Oliveri, M., Romero, L., & Papagno, C. (2004). Left but not right temporal involvement in
opaque idiom comprehension: A repetitive transcranial magnetic stimulation study. Journal of
Cognitive Neuroscience, 16(5), 848–855.
Ortony, A., Schallert, D. L., Reynolds, R. E., & Antos, S. J. (1978). Interpreting metaphors and
idioms: Some effects of context on comprehension. Journal of Verbal Learning and Verbal
Behavior, 17, 465–477.
Papagno, C., & Genoni, A. (2004). The role of syntactic competence in idiom comprehension: A
study on aphasic patients. Journal of Neurolinguistics, 17, 371–382.
Papagno, C., Tabossi, P., Colombo, M. R., & Zampetti, P. (2004). Idiom comprehension in aphasic
patients. Brain and Language, 89, 226–234.
Paulmann, S., Ghareeb-Ali, Z., & Felser, C. (2015). Neurophysiological markers of phrasal
verb processing: Evidence from L1 and L2 speakers. In R. R. Heredia & A. B. Cieślicka
(Eds.), Bilingual figurative language processing (pp. 245–267). New York: Cambridge
University Press.
Peleg, O., Giora, R., & Fein, O. (2001). Salience and context effects: Two are better than one.
116 A.B. Cieślicka
Peterson, R. R., & Burgess, C. (1993). Syntactic and semantic processing during idiom compre-
hension: Neurolinguistic and psycholinguistic dissociations. In C. Cacciari & P. Tabossi (Eds.),
Idioms: Processing, structure, and interpretation (pp. 201–225). Hillsdale, NJ: Lawrence
Erlbaum Associates.
Pollio, H., Barlow, J., Fine, H., & Pollio, M. (1977). Psychology and the poetics of growth:
Figurative language in psychology, psychotherapy, and education. Hillsdale, NJ: Lawrence
Erlbaum Associates.
Pollio, H. R., Fabrizi, M. S., Sills, A., & Smith, M. K. (1984). Need metaphoric comprehension
take longer than literal comprehension? Journal of Psycholinguistic Research, 13(3), 195–214.
Rapp, A. M., Leube, D. T., Erb, M., Grodd, W., & Kircher, T. T. J. (2004). Neural correlates of
metaphor processing. Cognitive Brain Research, 20, 395–402.
Rapp, A. M., Leube, D. T., Erb, M., Grodd, W., & Kircher, T. T. J. (2007). Laterality in meta-
phor processing: Lack of evidence from functional magnetic resonance imagining for the right
hemisphere theory. Brain and Language, 100(2), 142–149.
Schmidt, G. L., DeBuse, C. J., & Seger, C. A. (2007). Right hemisphere metaphor processing?
Characterizing the lateralization of semantic processes. Brain and Language, 100, 127–141.
Schmitt, N., & Carter, R. (2004). Formulaic sequences in action: An introduction. In N. Schmitt
(Ed.), Formulaic sequences: Acquisition, processing, and use (pp. 1–22). Philadelphia, PA:
John Benjamins.
Schmitt, N., & Underwood, G. (2004). Exploring the processing of formulaic sequences through a
self-paced reading task. Formulaic sequences: Acquisition, processing and use, 173-189.
Schweigert, W. A., & Moates, D. R. (1988). Familiar idiom comprehension. Journal of
Psycholinguistic Research, 17(4), 281–296.
Searle, J. (1975). Indirect speech acts. In P. Cole & J. L. Morgan (Eds.), Syntax and semantics.
Speech acts (pp. 59–82). New York: Academic.
Searle, J. (1979). Metaphor. In A. Ortony (Ed.), Metaphor and thought (pp. 92–123). Cambridge:
Shinjo, M., & Myers, J. L. (1987). The role of context in metaphor comprehension. Journal of
Memory and Language, 26, 226–241.
Siyanova-Chanturia, A. (2015). On the ‘holistic’ nature of formulaic language. Corpus Linguistics
and Linuistic Theory. doi:10.1515/cllt-2014-0016
Siyanova-Chanturia, A., Conklin, K., & Schmitt, N. (2011). Adding more fuel to the fire: An
eyetracking study of idiom processing by native and nonnative speakers. Second Language
Research, 27, 251–272.
Siyanova-Chanturia, A., Conklin, K., & van Heuven, W. (2011). Seeing a phrase ‘time and again’
matters: The role of phrasal frequency in the processing of multi-word sequences. Journal of
Experimental Psychology: Language, Memory and Cognition, 37, 776–784.
Snider, N., & Arnon, I. (2012). A unified lexicon and grammar? Compositional and non-
compositional phrases in the lexicon. In S. Gries & D. Divjak (Eds.), Frequency effects in
language (pp. 127–163). Berlin: Mouton de Gruyter.
Söderman, T. (1993). Word associations of foreign language learners and native speakers- the phe-
nomenon of shift response type and its relevance for lexical development. In H. Ringbom (Ed.),
Near-native proficiency in English (pp. 91–182). Abo: Abo Akademi, English Department.
Sprenger, S. A., Levelt, W. J. M., & Kempen, G. (2006). Lexical access during the production of
idiomatic phrases. Journal of Memory and Language, 54, 161–184.
Steinel, M. P., Hulstijn, J. H., & Steinel, W. (2007). Second language idiom learning in a paired-
associate paradigm. Effects of direction of testing, idiom imageability, and idiom transparency.
Studies in Second Language Acquisition, 29(3), 449–484.
Stringaris, A. K., Medford, N., Giora, R., Giampietro, V. C., Brammer, M. J., & David, A. S.
(2006). How metaphors influence semantic relatedness judgments: The role of the right frontal
cortex. NeuroImage, 33, 784–793.
Stringaris, A. K., Medford, N. C., Giampietro, V., Brammer, M. J., & David, A. S. (2007). Deriving
meaning: Distinct neural mechanisms for metaphoric, literal, and nonmeaningful sentences.
Brain and Language, 100, 150–162.
Swinney, D. A., & Cutler, A. (1979). The access and processing of idiomatic expressions. Journal
of Verbal Learning and Verbal Behavior, 18, 523–534.
Tabossi, P., Fanari, R., & Wolf, K. (2008). Processing idiomatic expressions: Effects of semantic
compositionality. Journal of Experimental Psychology: Learning, Memory, and Cognition,
34, 313–327.
Tabossi, P., Wolf, K., & Koterle, S. (2009). Idiom syntax: Idiosyncratic or principled? Journal of
Tabossi, P., & Zardon, F. (1993). The activation of idiomatic meaning in spoken language compre-
hension. In C. Cacciari & P. Tabossi (Eds.), Idioms: Processing, structure, and interpretation
(pp. 145–162). Hillsdale, NJ: Lawrence Erlbaum Associates.
Tabossi, P., & Zardon, F. (1995). The activation of idiomatic meaning. In M. Everaert, E.-J. van der
Linden, A. Schenk, & R. Schreuder (Eds.), Idioms: Structural and psychological perspectives
(pp. 273–282). Hillsdale, NJ: Lawrence Erlbaum Associates.
Temple, J. G., & Honeck, R. P. (1999). Proverb comprehension: The primacy of literal meaning.
Journal of Psycholinguistic Research, 28(1), 41–70.
Titone, D., Columbus, G., Whitford, V., Mercier, J., & Libben, M. (2015). Contrasting bilingual
and monolingual idiom processing. In R. R. Heredia & A. B. Cieślicka (Eds.), Bilingual figura-
tive language processing (pp. 171–207). New York: Cambridge University Press.
Titone, D. A., & Connine, C. M. (1994). Comprehension of idiomatic expressions: Effects of
predictability and literality. Journal of Experimental Psychology: Learning, Memory, and
Cognition, 20, 1126–1138.
Titone, D. A., & Connine, C. M. (1999). On the compositional and noncompositional nature of
idiomatic expressions. Journal of Pragmatics, 31, 1655–1674.
Titone, D. A., & Libben, M. (2014). Time-dependent effects of decomposability, familiarity and
literal plausibility on idiom meaning activation: A cross-modal priming investigation. The
Mental Lexicon, 9(3), 473–496. doi:10.1075/ml.9.3.05tit
Türker, E. (2016). The role of L1 conceptual and linguistic knowledge and frequency in the acqui-
sition of L2 metaphorical expressions. Second Language Research, 32(1), 25–48.
Uchiyama, H. T., Saito, D. N., Tanabe, H. C., Harada, T., Seki, A., Ohno, K., et al. (2012).
Distinction between the literal and intended meanings of sentences: A functional magnetic
resonance imaging study of metaphor and sarcasm. Cortex, 48, 563–583.
Underwood, G., Schmitt, N., & Galpin, A. (2004). The eyes have it: An eye-movement study into
the processing of formulaic sequences. In N. Schmitt (Ed.), Formulaic sequences: Acquisition,
processing, and use (pp. 153–172). Amsterdam/Philadelphia: John Benjamins.
Vaid, J., López, B. G., & Martinez, F. E. (2015). Linking the figurative to the creative:
Bilinguals’ comprehension of metaphors, jokes, and remote associates. In R. R. Heredia
& A. B. Cieślicka (Eds.), Bilingual figurative language processing (pp. 53–86). New York:
Vaid, J., & Martinez, F. (2001). Figurative language and thought across languages: What trans-
fers? Poster presented at Third International Symposium on Bilingualism, University of the
West of England, Bristol, April 18, 2001.
Van de Voort, M. E. C., & Vonk, W. (1995). You don't die immediately when you kick an empty
bucket: A processing view on semantic and syntactic characteristics of idioms. In M. Everaert,
E.-J. Van der Linden, A. Schenk, & R. Schreuder (Eds.), Idioms: Structural and psychological
perspectives (pp. 283–300). Hillsdale, NJ: Lawrence Erlbaum Associates.
Van Lancker, D., & Kempler, D. (1987). Comprehension of familiar phrases by left- but not right-
hemisphere damaged patients. Brain and Language, 32(2), 265–277.
Van Lancker Sidtis, D. (2003). Auditory recognition of idioms by native and nonnative speakers of
English: It takes one to know one. Applied Psycholinguistics, 24, 45–57.
Van Lancker Sidtis, D. (2004). When novel sentences spoken or heard for the first time in the his-
tory of the universe are not enough (Cf. Pinker, 1995, p. 22): Toward a dual-process model of
language. International Journal of Language & Communication Disorders, 39(1), 1–44.
118 A.B. Cieślicka
Van Lancker Sidtis, D. (2012). Two-track mind: Formulaic and novel language support a dual-
process model. In M. Faust (Ed.), The handbook of the neuropsychology of language (pp. 342–
367). Chichester: Wiley-Blackwell.
Van Lancker Sidtis, D., Cameron, K., Bridges, K., & Sidtis, J. J. (2015). The formulaic schema in
the minds of two generations of native speakers. Ampersand, 2, 39–48.
Vespignani, F., Canal, P., Molinaro, N., Fonda, S., & Cacciari, C. (2009). Predictive mechanisms in
idiom comprehension. Journal of Cognitive Neuroscience, 22(8), 1682–1700.
Weiland, H., Bambini, V., & Schumacher, P. B. (2014). The role of literal meaning in figura-
tive language comprehension: Evidence from masked priming ERP. Frontiers in Human
Neuroscience. doi:10.3389/fnhum.2014.00583
Winner, E., & Gardner, H. (1977). The comprehension of metaphor in brain damaged patients.
Brain, 100, 717–729.
Wray, A. (2000). Formulaic sequences in second language teaching: Principle and practice.
Applied Linguistics, 21, 463–489.
Wray, A. (2002). Formulaic language and the lexicon. Cambridge: Cambridge University Press.
Wray, A. (2012). What do we (think we) know about formulaic language? An evaluation of
the current state of play. Annual Review of Applied Linguistics, 32, 231–254. doi:10.1017/
S026719051200013X
Yorio, C. A. (1989). Idiomaticity as an indicator of second language proficiency. In K. Hyltenstam
& L. K. Obler (Eds.), Bilingualism across the lifespan: Aspects of acquisition, maturity, and
loss (pp. 55–72). Cambridge: Cambridge University Press.
Zhang, H., Yang, Y., Gu, J., & Ji, F. (2013). ERP correlates of compositionality in Chinese idiom
comprehension. Journal of Neurolinguistics, 26, 89–112.
Chapter 5
The Cost of Being Bilingual: The Example
of Verbal Fluency and Switching
Jeanette Altarriba and Stephanie A. Kazanas
Contents
indings from Behavioral Measures
F 120
Age Effects Across Behavioral Measures 123
Insights from Neurocognitive Data 125
Switching and Executive Control 128
Future Directions and Conclusions 131
References 132
The bilingual experience affords some everyday challenges that may have cognitive
and neural implications. Recent experimental work has begun to bridge the gap
between what we can intuit and what we observe, from switching studies that mimic
a bilingual speaker’s need to select and control their linguistic output, to fluency mea-
sures that can better explain why lexical access limitations often appear, like smaller
vocabulary and less accurate performance. Recent investigations within this literature,
using more controlled methodology and a wide array of tasks, have begun to pinpoint
where bilinguals excel, as a function of higher proficiency and more frequent switch-
ing among their languages. As a result, we can now consider whether advantages and
disadvantages thought to be characteristic of monolingual and bilingual populations
are as widespread and reliable as originally considered. Findings from behavioral and
neuroimaging methods illustrate the complex influence of task effects and participant
J. Altarriba
Department of Psychology, University at Albany, State University of New York,
Albany, NY 12222, USA
S.A. Kazanas (*)
Department of Counseling and Psychology, Tennessee Technological University,
Box 5031, Cookeville, TN 38505, USA
e-mail: skazanas@tntech.edu

120 J. Altarriba and S.A. Kazanas
variables, both of which may also limit the generalizability of previous work. The cur-
rent chapter navigates through these recent findings, to demonstrate how new methods
can clarify previous inconsistencies within this literature.
Findings from Behavioral Measures
Numerous findings from work with behavioral tasks—measures of language-

switching, conversation ability, and so on—have noted bilingual deficits on verbal
fluency: measurements of verbal functioning that ask participants to list category
exemplars (e.g., fruits, words that begin with the letter m). Bialystok (2009) has
argued that bilinguals likely have a smaller vocabulary, more interference on lexical
decision tasks, a greater number of tip-of-the-tongue (i.e., retrieving phonological,
lexical, and semantic information about a word, without retrieving the word itself)
experiences, and slower performance on picture-naming tasks than monolinguals
(with some of these findings observed across children and adults). These tasks ask
participants to activate and retrieve words and concepts in one of their (perhaps
numerous) languages, while suppressing competing words and concepts from other
languages. How do researchers explain discrepancies between bilingual and mono-
lingual participants? For one, the connectionist approach or the view that informa-
tion is stored in a network of weights and activations would argue that bilinguals
have weaker links between words and concepts within each language (Michael &
Gollan, 2005), the result of having to manage more than one lexicon. Cross-
linguistic competition may also account for some of these deficits, with the every-
day need to activate context-appropriate languages and suppress others in real time
(Green, 1998). Findings from language-switching studies also reveal language-
specific deficits (for reviews of these findings see e.g., Altarriba & Basnight-Brown,
2009; Kroll, Bobb, Misra, & Guo, 2008). Onset age of second language acquisition,
in combination with issues of language proficiency and balanced/unbalanced bilin-
gualism are also likely factors (and discussed later on). The following sections
describe these bilingual deficits and limiting factors.
Language-Specific Deficits in Switching Tasks
Language-switching is widely studied by bilingual researchers, and has greatly

evolved over time. Early language-switching research asked participants to read aloud
passages of text: Bilinguals required more time to read passages that alternated
between languages, when compared with passages written in a single language
(Kolers, 1966). Researchers then examined the cost of reading individual words,
which also alternated between languages (e.g., Dalrymple-Alford, 1985, Experiment
2). Grainger and Beauvillain (1987) then began to test switching behaviors with stim-
uli including pronounceable English and French words and nonwords.
5 Fluency and Switching Effects 121
More recent investigations have found an asymmetry in these switch costs,

whereby there are larger switch costs when switching from naming pictures or num-
bers in a participant’s second language (L2) to their native language (L1), than vice
versa (e.g., Costa & Santesteban, 2004; Meuter & Allport, 1999; Peeters, Runnqvist,
Bertrand, & Grainger, 2013; Rojczyk, 2011; see also Linck, Schwieter, &
Sunderman, 2012 for a similar pattern of results with trilinguals). These findings are
important, as they indicate that switch costs cannot be reduced to a simple task-
switching mechanism. Instead, these switch costs likely reflect language-specific
control mechanisms, such as activation or inhibition of lexical representations spe-
cific to a language (Grainger, Midgley, & Holcomb, 2010). Both Meuter and Allport
(1999) and Costa and Santesteban (2004) have argued that these specific switch
costs occur when switching from a weaker language to a more dominant language.
This switch creates more interference because the control requirements from the
previous trial (suppressing the more dominant language) are more influential and
incur a greater cost than would a weaker language. Additional support for this con-
jecture arose in Costa and Santesteban’s (2004) study, which failed to replicate the
asymmetry with highly proficient bilinguals. This finding lends support to the notion
that language-specific deficits in performance are a function of proficiency and
experience with each language. Therefore, costs appear to be most visible among
unbalanced bilinguals.
Further support is provided by Schwieter and Sunderman (2008) and their
switching task with English–Spanish bilinguals. In their work, they assessed switch-
ing as a function of lexical robustness, rather than proficiency or fluency alone.
Lexical robustness, according to these authors, captures both the strength of the
representation and the ability to access the lexical term (p. 216). Thus, using a con-
tinuous measure of language acquisition allows for more precise comparisons
across bilingual performance. Their results support a selection by proficiency model
that locates a development shift that corresponds with the point in which L2 profi-
ciency no longer relies on inhibitory control mechanisms. The components of their
model better account for asymmetrical switching than pure self-reported proficiency
measures. They recommend these additional measures and analyses, which can pin-
point where along the fluency continuum a bilingual’s switch cost will become zero,
with matched performance across both languages. According to the authors, their
model can also account for previous findings, including the lack of asymmetry
reported by Costa and Santesteban (2004) in their work with highly proficient
bilinguals.
Fluency Effects Among Bilinguals
Though verbal fluency measures are quite numerous, data often demonstrate a
common theme: Bilingual performance is often at a disadvantage, relative to mono-
lingual performance. Earlier findings are pervasive in this regard (for a brief review
of these deficits, see Bialystok, 2009), but some more recent findings suggest that
other, more influential participant variables may account for group differences. A
discussion of performance on these measures follows.
Fluency measures often tap into lexical access (i.e., retrieval), with participants
instructed to generate items in a short period of time (e.g., a minute or two), according
to a set of instructions. Typically, the instructions include a cue, such as a single letter
that participants must use to begin each word, or a category that participants must use
to name exemplars. Sandoval, Gollan, Ferreira, and Salmon (2010) have found that
monolinguals typically do outperform bilinguals on these tasks, with bilinguals also
starting more slowly and generating fewer low frequency exemplars, but many more
cognates (Experiment 1). Furthermore, these effects are more prominent in nondomi-
nant languages, with more cross-language intrusions in nondominant language blocks,
but can also occur in both of a balanced bilinguals’ languages (Experiment 2).
In another study, Bialystok, Craik, and Luk (2008b) used several versions of the
verbal fluency task, comparing monolingual performance with high and low profi-
ciency bilinguals. For example, participants were asked to name animals and to
name words beginning with the letters F, A, and S. Their study also included a ver-
sion with more complex instructions that limited their responses (e.g., the instruc-
tions exclude verbs with different endings would prohibit run and ran responses
within the same trial). On the standard versions of the fluency tasks, monolinguals
outperformed bilinguals. However, when comparing letter fluency performance
across bilinguals and monolinguals, while controlling for effects of vocabulary
scores, high proficiency bilinguals outperformed low proficiency bilinguals and
monolinguals (though this effect was limited to letter fluency, only).
Thus, it appears that proficiency variables may account for some of these fluency
effects. The effects of proficiency have also been assessed with other types of flu-
ency tasks. In one of these, Pivneva, Palmer, and Titone (2012) had English–French
and French–English bilinguals engage in monologue and dialogue speech as they
provided instructions with maps. The experimenter attempted to recreate the
described route, to measure verbal fluency. While instructions were more clear in a
bilingual’s dominant language, greater proficiency helped to boost performance in
their nondominant language. Speaker nativeness improved as a function of profi-
ciency, demonstrating that proficiency can improve both objective fluency (i.e., rec-
reating a route) and perceptions of fluency.
In one additional task, Yim and Bialystok (2012) modified a fluency task by creat-
ing mixed and blocked conditions (with the mixed condition requiring active language-
switching). Their participants were highly proficient Cantonese–English bilinguals,
who were grouped according to the frequency of their natural, conversational code-
switching. In their fluency task, participants generated fewer items in the mixed condi-
tion, relative to the blocked condition, indicating that the language-switching mode
inhibited lexical access. In addition, across both mixed and blocked conditions, par-
ticipants generated more words in English and more words when they began the block
in English. Finally, and perhaps most crucial to this discussion, bilinguals who
engaged in more frequent code-switching during a conversation task performed better
on the fluency tasks. Together, these findings are quite important: While fluency dif-
ficulties are quite common in the bilingual literature, recent findings have shed light
on the ways bilinguals can compensate for lexical access difficulties; bilinguals may
switch more often to select the best lexical representation, or they may rely on a more
dominant language when the occasion calls for more precise language. One proposed
mechanism that can explain these behaviors, executive processing ability, is discussed
at the end of the current chapter.
Age Effects Across Behavioral Measures
Effects of Early Versus Late Bilingualism
Within this area of language-switching and fluency research, researchers have also
made comparisons between groups of bilinguals. For example, bilinguals can differ
according to onset age of second language acquisition, with some learning a second
language quite young (i.e., early bilinguals) and others when they are older (i.e., late
bilinguals). These particular types of comparisons have been assessed across a vari-
ety of tasks, in part to determine when (and, in some cases, how) differences begin
to emerge between bilinguals and monolinguals. Of course, their results can also be
interpreted in other ways, as participant variables may actually account for some of
the previously discussed findings. Several examples of these comparisons and tasks
follow.
In one of these studies, Proverbio, Roberto, and Alberto (2007, Experiment 1)
examined sentence processing across Italian monolinguals, early Slovenian–Italian
bilinguals, and professional Italian–English interpreters (akin to late bilinguals).
Participants rated sentence comprehension via keypress, pressing a yes or no button
according to whether each sentence made sense to them. The authors noted that,
overall, the bilinguals were slower to process sentences in their L1, relative to the
monolinguals and interpreters processing sentences in their L1. Upon further
inspection, the authors found an interesting, though subtle nuance to this particular
effect: When processing sentences in their L1, bilinguals were significantly faster
when responding with their left hand than their right, suggesting more right hemi-
sphere activation. Importantly, all participants reported right-handed manual prefer-
ence. Meanwhile, no hand effect was observed when bilinguals processed sentences
in their L2, nor were any group differences observed when the other participants
used their left hand, suggesting that early bilingualism may lead to some additional
functional organization within the right hemisphere, though this did not lead to
more efficient decision-making performance. These kinds of disadvantages afforded
to bilinguals have also been observed in other types of tasks. Pelham and Abrams
(2014) tested similar groups to those of Proverbio et al. (2007): English monolin-
guals, early Spanish–English bilinguals, and late English–Spanish bilinguals. Their
study included a picture-naming task and an attentional network task: a task that can
mimic inhibitory control, as participants must suppress interference from distract-
ing information These tasks can test both lexical access and executive function, as
participants deploy attention to some stimuli or an aspect of a stimulus while

ignoring other stimuli or aspects of the same stimulus. Within the dominant lan-
guage, monolinguals named faster than bilinguals, with no differences across early
and late bilinguals. Monolingual participants also named low frequency pictures
faster than the bilingual participants. These fluency deficits are quite common, as
discussed throughout the earlier section of this chapter. In the attentional network
task, bilinguals demonstrated smaller conflict effects and were better able to sup-
press distracting information than monolinguals. Again, early and late bilinguals
performed similarly. Thus, similar effects were observed across early and late bilin-
guals: disadvantages in lexical access and advantages in executive function, relative
to monolinguals (see Bialystok, Craik, & Luk, 2008a for a similar pattern of results
with other tasks measuring lexical access and executive control).
However, dissociations between early and late bilinguals are quite common.
Luk, de Sa, and Bialystok (2011) used a battery of tests and tasks to compare mono-
linguals, early bilinguals, and late bilinguals: a vocabulary test (i.e., Peabody Picture
Vocabulary Task; Dunn & Dunn, 1997), nonverbal intelligence test (i.e., Cattell
Cultural Fair Intelligence Test; Cattell, 1957), spatial working memory task (i.e.,
the Spatial Span Subtest from the Wechsler Memory Scale; Wechsler, 1997), and
flanker task (adapted from Bunge, Dudukovic, Thomason, Vaidya, & Gabrieli,
2002; Eriksen & Eriksen, 1974). Across these measures, early bilingualism was
associated with smaller costs on the flanker task and higher vocabulary scores. Late
bilinguals had the lowest vocabulary scores and, perhaps most important to the cur-
rent discussion, performed like monolinguals on tasks assessing executive control.
Thus, the age of active bilingualism—instead of merely bilingualism in general—
appears to be a better predictor for these effects and ultimately when a bilingual’s
cognitive performance will differ from monolinguals. Importantly, the lack of strong
commonalities across these findings necessitate further research into these bilingual
group comparisons.
Effects of Aging Across the Lifespan
Other researchers have examined the ways in which switching behaviors and per-
formance are affected by aging. In these studies, researchers attempt to match bilin-
guals on a wide variety of variables: language proficiency and onset age of L2
acquisition, and other factors such as socioeconomic status, education, neurological
and psychological evaluations, and so on. Then, comparisons in performance are
made on the basis of age, as they examine the ways in which young (usually early
20s), middle-aged (usually mid 40s), and older (usually 60s–70s) bilinguals differ.
The following discussion illustrates an important, yet surprising theme: Age differ-
ences and commonalities are not ubiquitous across all switching tasks.
With picture-naming tasks, several researchers have failed to find differences
across age groups. In one of these studies, Gollan and Ferreira (2009, Experiment
3) assessed picture-naming in Spanish–English bilinguals’ self-rated dominant and
nondominant languages and gave them the opportunity to name in whatever

language came to mind, provided that they switched approximately half of the time
(i.e., voluntary switching). Surprisingly, the authors observed many commonalities
across age groups and balanced/unbalanced bilingual groups. However, most impor-
tantly, voluntary switching was not predicted by age, indicating that age could not
fully explain why, or under what circumstances bilinguals decided to switch.
Instead, the authors note that issues with language balance—ensuring that young
and older bilinguals have similar levels of experience with each language—may be
a key factor in understanding these findings. Thus, matching language experience,
in addition to the variables listed previously, may also be necessary for these stud-
ies. Then again, unconfounding language experience does not necessarily result in
significant age effects. In a picture-naming study with Catalan–Spanish bilinguals,
Calabria, Branzi, Marne, Hernández, and Costa (2015) also failed to detect any age
differences in switch costs from Catalan to Spanish, and vice versa, nor were there
any significant correlations between picture-naming performance and aging.
Perhaps language control mechanisms required to perform picture-naming under
language-switching instructions are not affected by age?
On the other hand, age effects do persist with other types of tasks. For example,
Gollan, Sandoval, and Salmon (2011) detected greater verbal fluency in young bilin-
guals, relative to older Spanish–English bilinguals. The older bilinguals made fewer
correct responses and generated more intrusions (including, though rare, some cross-
language intrusions). This result contrasts some consistent findings indicating that
older bilinguals can display advantages in tasks related to inhibitory control (e.g.,
Bialystok, Martin, & Viswanathan, 2005; Salvatierra & Rosselli, 2011; but see
Weissberger, Wierenga, Bondi, & Gollan, 2012 for some findings indicating that older
bilinguals may be at a disadvantage with some switching tasks). In addition, more
recent findings have detected some hints of a bilingual advantage in verbal fluency
measures with older adults. In their experiment, Friesen, Luo, Luk, and Bialystok
(2015) found that bilinguals outperformed monolinguals, though this effect was
observed in the letter version of the task, only (comparisons with the category version
suggested the opposite pattern, though results were confounded by vocabulary scores).
Overall, it may be a bit premature to make any formal conclusions on age effects in
language switching; results appear to greatly differ across tasks and populations.
Insights from Neurocognitive Data
Advances in neuroimaging have increased our understanding of verbal fluency and

cognitive control (for recent reviews, see e.g., Abutalebi, 2008; Abutalebi & Green,
2007; Bialystok, 2009). For example, the neural basis of cognitive control is believed
to lie in the prefrontal cortex, a region which facilitates language production in the
presence of irrelevant information (as is required in switching tasks). Other areas
that are likely involved include the parietal cortex, anterior cingulate cortex, and
basal ganglia. The latter two areas have been shown to detect conflict between
competing responses (e.g., the competition involved in the Stroop task as partici-
pants are asked to name the color of a printed word when the color does not match
the meaning of the word; i.e., blue printed in red) and assist in language planning
(as reviewed by Abutalebi & Green, 2007). Event-related potentials (ERP) method-
ology has also been useful in this regard. Early work conducted by Jackson,
Swainson, Cunnington, and Jackson (2001) found that the N2 component—believed
to be related to inhibitory control processes—replicated behavioral work claiming
the dominant language requires more suppression inhibition. A more recent investi-
gation conducted by Misra, Guo, Bobb, and Kroll (2012) assessed picture-naming
with ERPs and found that inhibition can also persist across blocked naming tasks.
Additional insights from these methods are discussed further below.
Functional magnetic resonance imaging (fMRI) has also been adopted for its use
in this area of research. Combining behavioral paradigms with fMRI have led to
important discoveries surrounding the neural basis of language-switching and
switching tasks in general. In one early example, Hernandez, Dapretto, Mazziotta,
and Bookheimer (2001) assessed language-switching with actions and objects, in
part to determine whether differences in neural representations exist across a bilin-
gual’s languages. Within- and between-language-switching, as well as monolingual
naming, were assessed with early Spanish–English bilinguals, highly proficient in
both languages. With behavioral data, Hernandez et al. observed faster and more
accurate naming in English. The fMRI recorded activation in Broca’s area when
naming in a single language and there were no differences in activation between
each language. However, during the switching task, increased activation was
observed in the right dorsolateral prefrontal cortex and left inferior frontal gyrus,
areas associated with increased executive functioning. Importantly, switching is not
always accompanied by activation in the right dorsolateral prefrontal cortex (e.g.,
Price, Green, & Von Studnitz, 1999). Instead, switching task difficulty likely moder-
ates activation of executive control areas. Together, these authors argue that fluency
and overall practice with language-switching are also important factors.
How can we be certain that activation in these areas has any relevance to bilin-
gualism, and not merely general switching behaviors? Rodríguez-Pujadas et al.
(2013) recently compared color-shape switching (assessing a set of colored shapes
according to their color or shape) across Spanish monolinguals and highly proficient
Catalan–Spanish bilinguals. Their behavioral data did not find any differences in
response time, accuracy, or switch costs across their two groups. However, their
fMRI results found increased activity in the left inferior frontal gyrus—replicating
the fMRI activation reported by Hernandez et al. (2001)—and the left caudate, an
additional area associated with language control. Additional support for differences
between monolinguals and bilinguals has been observed in semantic categorization
tasks in which participants define a series of nouns as either living or nonliving, also
utilizing fMRI methodology. Coderre, Smith, van Heuven, and Horwitz (2016)
detected functional overlap in the left inferior frontal gyrus, an area related to both
language and executive control with Spanish–English bilinguals, but not in English
monolinguals. These data can be interpreted in many ways. For one, monolinguals
could be more efficient: Less activation can be likened to simpler processing. Without
any differences in response time or accuracy, it is difficult to argue against this line
of reasoning. Nevertheless, it is also possible that language control areas can be acti-
vated without any costs in performance, as would be the case with simple tasks or
those that are so well-practiced that they can run more efficiently and automatically.
Thus, it appears that there is some monolingual advantage with respect to general
switching behaviors that cannot be explained with purely behavioral methods.
Despite these concerns, the most informative experimental designs within this
paradigm may still be those that combine behavioral and imaging methods.
Additionally, sometimes the behavioral and imaging data appear to be at odds with
previous findings and, at other times, they are at odds with each other. Thus, collect-
ing data simultaneously can be quite informative. For example, Perani et al. (2003)
assessed highly proficient Spanish–Catalan bilinguals’ verbal fluency while record-
ing fMRI data. Their behavioral data indicated that verbal fluency was matched
across Spanish and Catalan. However, their neural data did detect differences, with
significantly less brain activation during lexical retrieval in their second language.
The authors reasoned that increased exposure to a language will moderate the extent
of neural activation in a verbal fluency task: Since participants reported more regu-
lar exposure to their native language, repeated exposure strengthened the neocorti-
cal connections that often require some amount of increased left prefrontal cortex
engagement. Thus, despite their high proficiency in both Spanish and Catalan, along
with a very early L2 age of acquisition, lexical retrieval in their second language
incurred some neural cost. Again, these results mirror those of Rodríguez-Pujadas
et al. (2013) who detected differences in performance with neural data, but not
behavioral data.
Neural data have also been collected during a simultaneous interpreting task,
though this is a less common paradigm than others. In one of these studies, Hervais-
Adelman, Moser-Mercer, Michel, and Golestani (2015) had multilinguals engage in
simultaneous interpreting, shadowing, and listening. Shadowing (i.e., simultaneous
repetition, in which participants begin repeating after the onset of a sentence) and
listening tasks were created for meaningful fMRI comparisons, though all areas
activated during shadowing were also activated during simultaneous interpreting.
Nevertheless, the authors noted several areas that were specific to simultaneous
interpreting: the supplementary motor area, anterior cingulate cortex, caudate
nuclei, and left anterior insula. These results are important because they highlight
how additional areas are recruited during difficult tasks: areas responsible for
initiating complex motor actions, monitoring outcomes for errors, maximizing mul-
tilingual language control, and preparing for the attentional demands of this particu-
lar task. Moreover, this pattern of activation can provide useful insight into the ways
that novice interpreters—who likely found language-switching quite complex and
challenging—differ from more experienced interpreters. Together, this small com-
pilation of results from neural data helps to illustrate the complexity inherent in
language-switching research.
Switching and Executive Control
Support for a Bilingual Advantage in Executive Processing
Given what we have discovered about language-switching—the relationship between

everyday switching and fluency task performance, how it is measured in experimental
contexts, and how it varies across tasks and populations—researchers are now turning
toward more applied questions. What can switching behaviors and switching perfor-
mance tell us more generally: Is language-switching similar to task-switching? Does
practice with switching afford some advantages to those who switch often, relative to
those who have fewer opportunities to engage in switching? What can language-
switching research inform us, regarding bilingualism, that we cannot glean from stud-
ies with monolingual participants? Very generally, findings gathered from this
particular set of research questions have been very fruitful. Many of them support the
existence of some general cognitive advantages, perhaps a function of the central
executive and its regulation of task demands (for recent reviews, see e.g., Green &
Abutalebi, 2013; Hilchey & Klein, 2011). Green and Abutalebi (2013) have hypoth-
esized that executive control may involve a series of control processes that are differ-
entially activated by single and dual language contexts, as well as switching contexts:
goal maintenance, conflict monitoring, interference suppression, and so on. The most
current debate rests with what these kinds of advantages mean, particularly, how, and
under what circumstances, language-switching relates to executive control.
In one recent example, Festman, Rodriguez-Fornells, and Münte (2010) tested
executive control among late Russian–German bilinguals with tasks including the
Tower of Hanoi, Go/noGo, and divided attention: This set of tasks taps skills related
to problem-solving, attention, and inhibitory control. The authors used a picture-
naming task to separate switchers (those with weak language control abilities, who
could not maintain single language use) from nonswitchers. The nonswitchers per-
formed better on the executive control tasks: They required fewer moves to reach
the target in the Tower of Hanoi and responded faster and more accurately in the Go/
noGo and divided attention tasks. Nonswitchers also outperformed switchers on the
intelligence test and verbal fluency measures. Thus, Festman et al. (2010) reasoned
that there is a strong relationship between language control abilities and executive
control abilities; bilinguals who were better able to suppress irrelevant lexical infor-
mation while picture-naming were also able to monitor irrelevant task information.
Of course, this notion of dividing the bilingual experience into dichotomous switcher
and nonswitcher groups may be problematic, given its more continuous nature. In
spite of that conjecture, Prior and Gollan (2011) found some similar results with
their bilingual participants, specifically those who reported very infrequent switch-
ing in daily conversation. Those who switched less often performed more slowly in
a task-switching paradigm than monolinguals and more frequent language switch-
ers. Verreyt, Woumans, Vandelanotte, Szmalec, and Duyck (2016) have also reported
this executive processing advantage with switchers using Simon and flanker tasks.
However, these comparisons are a bit tentative, as their bilinguals were performing
different tasks (and, in different languages) than Festman et al.’s (2010) bilinguals.
Despite the significance of these results, the link between language-switching

and executive control may have its limitations (see e.g., Bobb & Wodniecka, 2013
for additional insight regarding the tenuous relationship between switching and
inhibition). In addition, the magnitude of any bilingual advantage appears to depend
on a variety of important factors. For one, task demands may affect the advantage.
Findings from low monitoring and high monitoring versions of the attentional net-
work task indicate that bilinguals excel on the more difficult high monitoring task
(with more variable congruent/incongruent trials), with faster performance than
monolinguals (Costa, Hernández, Costa-Faidella, & Sebastián-Gallés, 2009). In
addition, balanced and more experienced bilinguals also garner a greater benefit
than unbalanced bilinguals (e.g., Blumenfeld & Marian, 2013; Woumans, Ceuleers,
Van der Linden, Szmalec, & Duyck, 2015). In one example, Blumenfeld and Marian
(2013) compared high and low proficiency English–Spanish bilinguals on several
measures, with the most crucial being a word recognition task. During this task,
participants heard an English word (e.g., pool) and had to locate its matching image
in an array that also contained cross-linguistic competitor images (e.g., a pulgar, or
thumb) and fillers. Eye-tracking measures indicated decreased competitor activa-
tion during the later stages of word recognition, particularly among the high profi-
ciency bilinguals (who also performed better on a Stroop task than the low
proficiency bilinguals). Woumans et al. have also observed the largest advantages in
executive processing with their balanced bilinguals and interpreters, as they outper-
formed unbalanced bilinguals on both the Simon task and attentional network task.
Verbal fluency measures were also correlated with Simon task performance, but
again, these effects were most prominent among balanced bilinguals. Similar effects
have been described by other researchers, comparing Simon task performance and
working memory task performance (e.g., Linck, Hoshino, & Kroll, 2008; Poarch &
van Hell, 2012; Soliman, 2014; Tse & Altarriba, 2014).
Across these experiments, researchers have detected a strong relationship
between bilingual proficiency, nontarget language suppression, and conflict resolu-
tion. Importantly, these studies have included a wide array of tasks and populations,
while generating similar findings (Poarch & van Hell, 2012; Soliman, 2014; with
Tse & Altarriba, 2014 in fact, administering these tasks to young children). Thus, it
seems that proficiency may enhance the monitoring and conflict resolution aspects
of executive control abilities. As a result, these enhancements may be most visible
during difficult tasks, as was the case in the high monitoring condition in the atten-
tional network task (Costa et al., 2009).
Disputing the Bilingual Advantage in Executive Processing
Given these limitations, Paap and his colleagues have largely argued against any siz-
able bilingual advantage, oftentimes debating its very existence (see Paap, 2014,
2017; Paap, Johnson, & Sawi, 2015 for recent reviews of this literature). Citing
issues with extant work, Paap reasons that bilingual advantages in the literature,
including some works described in the current chapter, may be due to researcher
biases, small sample sizes, too-narrow demographic data, and scant neural data.
More worrisome are the potential publication biases within this area of research,
including the increased likelihood of publishing work supporting a bilingual advan-
tage, but not a null effect or bilingual disadvantage (de Bruin, Treccani, & Della Sala,
2015a, 2015b; but see Bialystok, Kroll, Green, MacWhinney, & Craik, 2015 for a
counterargument). Several of the more problematic findings for a bilingual advan-
tage are discussed further below (for an additional, brief review, see Jared, 2015).
Null effects can be particularly difficult to publish (i.e., the file-drawer problem).
When they are published, and within a hot topic, their findings are often heavily
scrutinized. Several recent papers deserve some special consideration. In the first of
these, von Bastian, Souza, and Gade (2016) tested four hypotheses underlying cog-
nitive advantages afforded to bilinguals: inhibitory control, conflict monitoring,
shifting, and general cognitive performance. Their tasks included the Simon, flanker,
and Stroop tasks, color-shape switching, task-switching, working memory, reason-
ing, and several others. von Bastian et al. failed to find any evidence for a bilingual
benefit. In fact, higher levels of parents’ education were a stronger predictor for
cognitive ability and task performance than degree of bilingualism. Their work did
have its limitations, most notably the lack of a monolingual group for a true baseline
comparison with their bilinguals.
Similar findings have been observed with both children and adults, even with a
control group. In their longitudinal design, Woumans, Surmont, Struys, and Duyck
(2016) studied preschool children: Half were enrolled in a monolingual program
and half in a bilingual program. Both groups were matched on age, gender, socio-
economic status, verbal fluency, reasoning, and Simon task performance. After a
year, both groups improved, equally, on verbal fluency measures. Bilingual children
saw some improvement in reasoning ability, but no improvement on the Simon task
(there was, in fact, a speed-accuracy tradeoff). In one final example, Scaltriti,
Peressotti, and Miozzo (2017) compared Venetian–Italian bilinguals (Venetian
being an Italian dialect) and Italian monolinguals on verbal fluency, flanker, and
Stroop task performance. The authors did not detect any bilingual advantages across
these measures. Is dialect-switching not as cognitively demanding as other, more
disparate languages? Is the bilingual advantage a bit misleading, if only special
populations receive a boost in performance?
These findings illustrate a common theme: Tasks purported to measure executive
processing benefits in other studies at times do not appear to measure any benefit at
all. Moreover, bilinguals sometimes show larger Simon effects, as well as greater
antisaccade (an additional task requiring response inhibition) and Simon response
time, indicative of bilingual disadvantages (Paap & Greenberg, 2013; Paap & Sawi,
2014). Perhaps the most troubling finding across these studies was initially detected
by Paap and Sawi (2014), whose large battery of executive functioning tasks led to
a combination of null effects, bilingual advantages and disadvantages, and margin-
ally significant results. Across their findings, measures of executive processing per-
formance (e.g., the Simon and flanker tasks) were not correlated. Should researchers
continue to use these tasks to measure executive processing? Are these tasks the
proper baseline for language-switching performance? And, what do they truly

measure?
In addition to these issues with convergent validity, Paap (2017) has further
advised researchers to limit any explicit comparisons across participants only to
samples closely matched on culture, socioeconomic status, immigrant status, and
other related experiences that could confound advantages observed across bilin-
guals and monolinguals, variable language proficiency, and so on. In addition, exec-
utive function is very likely a heritable trait, and additional differences in ability are
likely a combination of everyday tasks and activities, as well as special interests.
Thus, the most prudent researcher would want to carefully consider these recent
controversies: from selecting the proper switching tasks and fluency measures, to
controlling for the multitude of relevant participant variables.
Future Directions and Conclusions
As summarized in the current chapter, bilingual performance on fluency measures

and switching tasks depends on a number of factors. These factors include, but are
not limited to: the tasks, paradigms, and bilinguals completing each task. The tasks
appear to vary on what they are purported to measure: proficiency, executive pro-
cessing, lexical robustness, and so on. In terms of paradigm differences, behavioral
and neuroimaging data are often inconsistent, and sometimes only marginally sig-
nificant. Lastly, participant variables have had the greatest impact, with performance
differing as a function of age, onset age of second language acquisition, proficiency,
balanced bilingualism, switching frequency, and more. Truly, both the subtle and
bold differences across studies can be numerous, demonstrating the continuous
need for additional, carefully controlled research that considers how previous work
can be updated and improved.
A recent issue of the journal Cortex has provided extensive advice to those
invested in this area of research. According to Li and Grant (2015), while Paap and
colleagues have provided an important and necessary critique of the extant bilingual
literature, additional work is needed that examines the causal link and factors under-
lying the bilingual experience and cognitive control. Others have urged future
research to consider additional moderators that affect bilingual performance on
these fluency measures and switching tasks, recommending longitudinal designs
and additional norming for these measures and tasks (e.g., Woumans & Duyck,
2015). Anecdotally, we can understand and appreciate that bilinguals must continu-
ously monitor their speech and select the appropriate lexical representations, given
their surroundings and current linguistic context. However, whether this continuous
task affords them other advantages and how these advantages are gained, remains
largely unknown. They recommend longitudinal designs, even those that are merely
short-term (e.g., across an introductory language course) to track these advantages
and their development over time (for similar commentary, see Gold, 2015; Woumans
& Duyck, 2015).
Others, like Vaughn, Greene, Nuñez, and Hernandez (2015) are a bit hesitant to
encourage future research investigating bilingual advantages and disadvantages,
particularly with neuroimaging data. Instead, Vaughn et al. argue that it is not neces-
sarily important to compare behavioral data and neuroimaging data with these pop-
ulations, as the causal link may not lie in these data. They point to recent findings in
genetic research and language ability that could better explain the development of
cognitive abilities including switching and control processes (e.g., Hernandez,
Greene, Vaughn, Francis, & Grigorenko, 2015; Stelzel, Basten, Montag, Reuter, &
Fiebach, 2010).
Finally, each of these researchers has concluded that this literature is both impor-
tant and meaningful, in both a basic and applied sense. To continue moving forward,
as a field, van Heuven and Coderre (2015) have urged researchers to be mindful that
more sophisticated statistical approaches will provided a better understanding of
complex data, including data from neuroimaging techniques. Indeed, assessing con-
tinuous participant variables (e.g., proficiency measures, and language usage),
switching frequency, and other variables are also becoming more necessary within
this literature. Incorporating additional measures and tasks seems most important,
given inconsistencies across tasks and issues with convergent validity. Perhaps
assessing higher-level abilities, such as those related to working memory, can pro-
vide a better representation of a bilingual advantage (e.g., Tse & Altarriba, 2014).
References
Psychologica, 128, 466–478.
Abutalebi, J., & Green, D. (2007). Bilingual language production: The neurocognition of language
representation and control. Journal of Neurolinguistics, 20, 242–275.
Altarriba, J., & Basnight-Brown, D. M. (2009). Empirical approaches to the study of code-
switching in sentential contexts. In L. Isurin, D. Winford, & K. de Bot (Eds.), Cross-disciplinary
approaches to code-switching (pp. 3–25). Amsterdam: John Benjamins.
Bialystok, E. (2009). Bilingualism: The good, the bad, and the indifferent. Bilingualism: Language
and Cognition, 12(1), 3–11.
Bialystok, E., Craik, F. I. M., & Luk, G. (2008a). Cognitive control and lexical access in younger
and older bilinguals. Journal of Experimental Psychology: Learning, Memory, and Cognition,
34(4), 859–873.
Bialystok, E., Craik, F. I. M., & Luk, G. (2008b). Lexical access in bilinguals: Effects of vocabu-
lary size and executive control. Journal of Neurolinguistics, 21, 522–538.
Bialystok, E., Kroll, J. F., Green, D. W., MacWhinney, B., & Craik, F. I. M. (2015). Publication bias
and the validity of evidence: What’s the connection? Psychological Science, 26(6), 944–946.
Bialystok, E., Martin, M. M., & Viswanathan, M. (2005). Bilingualism across the lifespan: The rise
and fall of inhibitory control. International Journal of Bilingualism, 9(1), 69–85.
Blumenfeld, H. K., & Marian, V. (2013). Parallel language activation and cognitive control dur-
ing spoken word recognition in bilinguals. Journal of Cognitive Psychology, 25(5), 547–567.
Bobb, S. C., & Wodniecka, Z. (2013). Language switching in picture naming: What asymmetric
switch costs (do not) tell us about inhibition in bilingual speech planning. Journal of Cognitive
Psychology, 25(5), 568–585.
Bunge, S. A., Dudukovic, N. M., Thomason, M. E., Vaidya, C. J., & Gabrieli, J. D. (2002).
Immature frontal lobe contributions to cognitive control in children: Evidence from fMRI.
Neuron, 33, 301–311.
Calabria, M., Branzi, F. M., Marne, P., Hernández, M., & Costa, A. (2015). Age-related effects
over bilingual language control and executive control. Bilingualism: Language and Cognition,
18(1), 65–78.
Cattell, R. B. (1957). Culture fair intelligence test, a measure of “g”: Scale 3, forms A and B
(high school pupils and adults of superior intelligence). Savoy, IL: Institute for Personality and
Ability Testing.
Coderre, E. L., Smith, J. F., van Heuven, W. J. B., & Horwitz, B. (2016). The functional over-
lap of executive control and language processing in bilinguals. Bilingualism: Language and
Cognition, 19(3), 471–488.
Costa, A., Hernández, M., Costa-Faidella, J., & Sebastián-Gallés, N. (2009). On the bilingual
advantage in conflict processing: Now you see it, now you don’t. Cognition, 113, 135–149.
Costa, A., & Santesteban, M. (2004). Lexical access in bilingual speech production: Evidence
from language switching in highly proficient bilinguals and L2 learners. Journal of Memory
and Language, 50, 491–511.
Dalrymple-Alford, E. C. (1985). Language switching during bilingual reading. British Journal of
Psychology, 76(1), 111–122.
de Bruin, A., Treccani, B., & Della Sala, S. (2015a). Cognitive advantage in bilingualism: An
example of publication bias? Psychological Science, 26(1), 99–107.
de Bruin, A., Treccani, B., & Della Sala, S. (2015b). The connection is in the data: We should
consider them all. Psychological Science, 26(6), 947–949.
Dunn, L. M., & Dunn, L. M. (1997). Peabody picture vocabulary test (3rd ed.). Bloomington, IN:
Pearson Assessments.
Eriksen, B. A., & Eriksen, C. W. (1974). Effects of noise letters upon the identification of a target
letter in a nonsearch task. Perception & Psychophysics, 16, 143–149.
Festman, J., Rodriguez-Fornells, A., & Münte, T. F. (2010). Individual differences in control
of language interference in late bilinguals are mainly related to general executive abilities.
Behavioral and Brain Functions, 6, 5.
Friesen, D. C., Luo, L., Luk, G., & Bialystok, E. (2015). Proficiency and control in verbal fluency
performance across the lifespan for monolinguals and bilinguals. Language, Cognition, and
Neuroscience, 30(3), 238–250.
Gold, B. T. (2015). Executive control, brain aging and bilingualism. Cortex, 73, 369–370.
Gollan, T. H., & Ferreira, V. S. (2009). Should I stay or should I switch? A cost-benefit analy-
sis of voluntary language switching in young and aging bilinguals. Journal of Experimental
Psychology: Learning, Memory, and Cognition, 35(3), 640–665.
Gollan, T. H., Sandoval, T., & Salmon, D. P. (2011). Cross-language intrusion errors in aging bilin-
guals reveal the link between executive control and language selection. Psychological Science,
22(9), 1155–1164.
Grainger, J., & Beauvillain, C. (1987). Language blocking and lexical access in bilinguals.
Quarterly Journal of Experimental Psychology, 39(2), 295–319.
Grainger, J., Midgley, K., & Holcomb, P. J. (2010). Re-thinking the bilingual interactive-activation
model from a developmental perspective (BIA-d). In M. Kail & M. Hickmann (Eds.), Language
acquisition across linguistic and cognitive systems (pp. 267–284). New York: Benjamins.
Green, D. W. (1998). Mental control of the bilingual lexico-semantic system. Bilingualism:
Language and Cognition, 1, 67–81.
Green, D. W., & Abutalebi, J. (2013). Language control in bilinguals: The adaptive control hypoth-
esis. Journal of Cognitive Psychology, 25(5), 515–530.
Hernandez, A. E., Dapretto, M., Mazziotta, J., & Bookheimer, S. (2001). Language switching
and language representation in Spanish-English bilinguals: An fMRI study. NeuroImage, 14,
510–520.
Hernandez, A. E., Greene, M. R., Vaughn, K. A., Francis, D. J., & Grigorenko, E. L. (2015).
Beyond the bilingual advantage: The potential role of genes and environment on the develop-
ment of cognitive control. Journal of Neurolinguistics, 35, 109–119.
Hervais-Adelman, A., Moser-Mercer, B., Michel, C. M., & Golestani, N. (2015). fMRI of simul-
taneous interpretation reveals the neural basis of extreme language control. Cerebral Cortex,
25(12), 4727–4739.
Hilchey, M. D., & Klein, R. M. (2011). Are there bilingual advantages on nonlinguistic interfer-
ence tasks? Implications for the plasticity of executive control processes. Psychonomic Bulletin
& Review, 18(4), 625–658.
Jackson, G. M., Swainson, R., Cunnington, R., & Jackson, S. R. (2001). ERP correlates of execu-
tive control during repeated language switching. Bilingualism: Language and Cognition, 4,
169–178.
Jared, D. (2015). What is the theory? Cortex, 73, 361–363.
Kolers, P. A. (1966). Reading and talking bilingually. American Journal of Psychology, 79(3),
347–376.
Kroll, J. F., Bobb, S. C., Misra, M., & Guo, T. (2008). Language selection in bilingual speech:
Evidence for inhibitory processes. Acta Psychologica, 128, 416–430.
Li, P., & Grant, A. (2015). Identifying the causal link: Two approaches toward understanding the
relationship between bilingualism and cognitive control. Cortex, 73, 358–360.
Linck, J. A., Hoshino, N., & Kroll, J. F. (2008). Cross-language lexical processes and inhibitory
control. Mental Lexicon, 3(3), 349–374.
Linck, J. A., Schwieter, J. W., & Sunderman, G. (2012). Inhibitory control predicts language
switching performance in trilingual speech production. Bilingualism: Language and Cognition,
15(3), 651–662.
Luk, G., de Sa, E., & Bialystok, E. (2011). Is there a relation between onset age of bilingualism and
enhancement of cognitive control? Bilingualism: Language and Cognition, 14(4), 588–595.
Meuter, R. F. I., & Allport, A. (1999). Bilingual language switching in naming: Asymmetrical costs
of language selection. Journal of Memory and Language, 40, 25–40.
Michael, E. B., & Gollan, T. H. (2005). Being and becoming bilingual: Individual differences and
consequences for language production. In J. F. Kroll & A. M. B. de Groot (Eds.), Handbook of
bilingualism: Psycholinguistic approaches (pp. 389–407). New York: Oxford University Press.
Misra, M., Guo, T., Bobb, S. C., & Kroll, J. F. (2012). When bilinguals choose a single word to
speak: Electrophysiological evidence for inhibition of the native language. Journal of Memory
and Language, 67, 224–237.
Paap, K. R. (2014). The role of componential analysis, categorical hypothesizing, replicability
and confirmation bias in testing for bilingual advantages in executive functioning. Journal of
Cognitive Psychology, 26(3), 242–255.
Paap, K. R. (2017). Bilingualism and executive functioning. In J. Altarriba & R. R. Heredia (Eds.),
An introduction to bilingualism: Principles and processes (2nd ed.). New York: Taylor &
Francis Group.
Paap, K. R., & Greenberg, Z. I. (2013). There is no coherent evidence for a bilingual advantage in
executive processing. Cognitive Psychology, 66, 232–258.
Paap, K. R., Johnson, H. A., & Sawi, O. (2015). Bilingual advantages in executive functioning
either do not exist or are restricted to very specific and undetermined circumstances. Cortex,
69, 265–278.
Paap, K. R., & Sawi, O. (2014). Bilingual advantages in executive functioning: Problems in
convergent validity, discriminant validity, and the identification of the theoretical constructs.
Frontiers in Psychology, 5, 962.
Peeters, D., Runnqvist, E., Bertrand, D., & Grainger, G. (2013). Asymmetrical switch costs in
bilingual language production induced by reading words. Journal of Experimental Psychology:
Learning, Memory, and Cognition, 40(1), 284–292.
Pelham, S. D., & Abrams, L. (2014). Cognitive advantages and disadvantages in early and late
bilinguals. Journal of Experimental Psychology: Learning, Memory, and Cognition, 40(2),
313–325.
Perani, D., Abutalebi, J., Paulesu, E., Brambati, S., Scifo, P., Cappa, S. F., et al. (2003). The role
of age of acquisition and language usage in early, high-proficient bilinguals: An fMRI study
during verbal fluency. Human Brain Mapping, 19, 170–182.
Pivneva, I., Palmer, C., & Titone, D. (2012). Inhibitory control and L2 proficiency modulate
bilingual language production: Evidence from spontaneous monologue and dialogue speech.
Frontiers in Psychology, 3, 57.
Poarch, G. J., & van Hell, J. G. (2012). Executive functions and inhibitory control in multilin-
gual children: Evidence from second-language learners, bilinguals, and trilinguals. Journal of
Experimental Child Psychology, 113, 535–551.
Price, C. J., Green, D. W., & Von Studnitz, R. (1999). A functional imaging study of translation and
language switching. Brain, 122, 2221–2235.
Prior, A., & Gollan, T. H. (2011). Good language-switchers are good task-switchers: Evidence from
Spanish-English and Mandarin-English bilinguals. Journal of International Neuropsychological
Society, 17, 682–691.
Proverbio, A. M., Roberto, A., & Alberto, Z. (2007). The organization of multiple languages in
polyglots: Interference or independence? Journal of Neurolinguistics, 20, 25–49.
Rodríguez-Pujadas, A., Sanjuán, A., Ventura-Campos, N., Román, P., Martin, C., Barceló, F., et al.
(2013). Bilinguals use language-control brain areas more than monolinguals to perform non-
linguistic switching tasks. PLoS One, 8(9), e73028.
Rojczyk, A. (2011). Time-limited verbal fluency task with Polish-English unbalanced bilinguals.
In J. Arabski & A. Wojtaszek (Eds.), Individual learner differences in SLA (pp. 210–225).
Tonawanda, NY: Multilingual Matters.
Salvatierra, J. L., & Rosselli, M. (2011). The effect of bilingualism and age on inhibitory control.
International Journal of Bilingualism, 15(1), 26–37.
Sandoval, T. C., Gollan, T. H., Ferreira, V. S., & Salmon, D. P. (2010). What causes the bilingual
disadvantage in verbal fluency? The dual-task analogy. Bilingualism: Language and Cognition,
13(2), 231–252.
Scaltriti, M., Peressotti, F., & Miozzo, M. (2017). Bilingual advantage and language switch:
What’s the linkage? Bilingualism: Language and Cognition, 20(1), 80–97.
Schwieter, J. W., & Sunderman, G. (2008). Language switching in bilingual speech production.
The Mental Lexicon, 3(2), 214–238.
Soliman, A. M. (2014). Bilingual advantages of working memory revisited: A latent variable
examination. Learning and Individual Differences, 32, 168–177.
Stelzel, C., Basten, U., Montag, C., Reuter, M., & Fiebach, C. J. (2010). Frontostriatal involve-
ment in task switching depends on genetic differences in D2 receptor activity. Journal of
Neuroscience, 30(42), 14205–14212.
Tse, C.-S., & Altarriba, J. (2014). The relationship between language proficiency and attentional
control in Cantonese-English bilingual children: Evidence from Simon, Simon switching, and
working memory tasks. Frontiers in Psychology, 5, 954.
van Heuven, W. J. B., & Coderre, E. L. (2015). A call for sophisticated statistical approaches and
neuroimaging techniques to study the bilingual advantage. Cortex, 73, 330–331.
Vaughn, K. A., Greene, M. R., Nuñez, A. I. R., & Hernandez, A. E. (2015). The importance of
neuroscience in understanding bilingual cognitive control. Cortex, 73, 373–374.
Verreyt, N., Woumans, E., Vandelanotte, D., Szmalec, A., & Duyck, W. (2016). The influence
of language-switching experience on the bilingual executive control advantage. Bilingualism:
Language and Cognition, 19(1), 181–190.
von Bastian, C. C., Souza, A. S., & Gade, M. (2016). No evidence for bilingual cognitive advan-
tages: A test of four hypotheses. Journal of Experimental Psychology: General, 145(2),
246–258.
Wechsler, D. (1997). Spatial span subtest of Wechsler Memory Scale–Third Edition (WMS-III).
San Antonio, TX: The Psychological Corporation.
Weissberger, G. H., Wierenga, C. E., Bondi, M. W., & Gollan, T. H. (2012). Partially overlapping
mechanisms of language and task control in young and older bilinguals. Psychology and Aging,
27(4), 959–974.
Woumans, E., Ceuleers, E., Van der Linden, L., Szmalec, A., & Duyck, W. (2015). Verbal and
nonverbal cognitive control in bilinguals and interpreters. Journal of Experimental Psychology:
Learning, Memory, & Cognition, 41(5), 1579–1586.
Woumans, E., & Duyck, W. (2015). The bilingual advantage debate: Moving toward different
methods for verifying its existence. Cortex, 73, 356–357.
Woumans, E., Surmont, J., Struys, E., & Duyck, W. (2016). The longitudinal effect of bilingual
immersion schooling on cognitive control and intelligence. Language Learning, 66(2), 76–91.
Yim, O., & Bialystok, E. (2012). Degree of conversational code-switching enhances verbal task
switching in Cantonese-English bilinguals. Bilingualism: Language and Cognition, 15(4),
873–883.
Chapter 6
Bilingual Episodic Memory Processes
Roberto R. Heredia and Elva A. García
Creo que venimos a este mundo nada más para aprender y sin
pensarlo tarde o temprano alguien … [nos] lo hará entender
(Marco Antonio Solis).
Contents
Introduction 137
Bilingual Memory 138
Encoding 139
Retrieval 148
Conclusion 153
References 154
Introduction
Knowing that dogs, like humans, possess an episodic-like memory system that
allows them to encode and recall information (Fugazza, Pogány, & Miklósi, 2016)
is factual information. Knowing that killer whales and Asian elephants, like humans,
have grandmas (Lahdenperä, Mar, & Lumma, 2016) is also factual scientific knowl-
edge. This knowledge is part of one’s semantic memory. Now consider the anecdote
in which the first author remembers the day he drank a liter of Charanda (a rum-like
alcoholic beverage) that his grandfather carelessly misplaced under the bed. He
wakes up in a hospital bed with his mother and sister by his side, very happy that he
was alive. He remembers being terrified and frightened that he would be scolded
and punished. This personal and highly vivid (e.g., head spinning, bright lights,
smell of alcohol) experience is an episodic memory involving a specific event and
time in his life (he was about 8 years old!). This episodic memory, as argued by
Tulving (1985), is characterized by the subjective feeling of I remember as opposed
to I know that describes facts and general knowledge (i.e., semantic memory) that is
R.R. Heredia (*) • E.A. García

e-mail: rheredia@tamiu.edu

138 R.R. Heredia and E.A. García
independent of personal experiences or temporal learning. Notice that it could be

said that during the 2016 presidential election I remember not believing my ears and
eyes, but it would be odd to remark that I knew I could not believe my ears and eyes.
Thus, episodic memory is a system that encodes, represents (stores), and retrieves
information. It is part of a system that interacts with semantic memory and long-
term memory (Roediger, 1990).
Although bilingual storage (i.e., whether bilinguals organize their languages in
one or two memory systems) has been one of the most debated issues over the last
40 years or so (e.g., Basnigh-Brown, 2014; Heredia, 2008), this paper is about bilin-
gual memory processes. We specifically look at some of the most influential pro-
cessing approaches to encoding (e.g., levels of processing, the generation,
distributed-practice and repetition effects) and retrieval (e.g., encoding specificity
principle and transfer appropriate processing) from the monolingual episodic
memory literature to gauge their generalizability to bilingual episodic memory and
assess possible interaction between memory and language effects. Our purpose is to
critically review and evaluate these classic memory effects, in relation to bilingual
episodic memory, and when appropriate, suggest research areas that would benefit
from further empirical inquiry. Early bilingual long-term models (e.g., bilingual
dual coding theory), and other theoretical frameworks of bilingual memory repre-
sentations are discussed, and when possible, they are related to the empirical find-
ings being discussed.
Bilingual Memory
In the typical bilingual episodic memory experimental paradigm, bilingual speakers

learn a series of word pairs (e.g., synonyms) in the first language (L1; e.g., casa-
hogar), second language (L2; e.g., house-home), and translations from L1–L2 (e.g.,
casa-house), and L2–L1 (e.g., house-casa). Depending on the experimental manip-
ulation, whether the memory task is incidental (where participants are unaware that
their memory will be tested), or intentional (where participants are aware that their
memory will be measured; e.g., Cieślicka, Morales, Heredia, 2013; Craik &
Lockhart, 1972; Heredia, Cieślicka, & Altarriba, 2016), after the encoding phase,
participants recall as many words as possible, or on a recognition task, determine if
the presented word was studied (old) or nonstudied (new). The dependent measure
is accuracy of retrieval, and depending on the theoretical question, reaction response
times (RTs) as a function of transfer (see for example, Francis & Gutiérrez, 2012).
Other memory tasks might include cued recall in which participants are provided
with a cue (e.g., casa__) and asked to remember the studied target (e.g., synonym
hogar or translation: house).
Typically, L1–L1 (casa-hogar) and L2–L2 (house-home) language conditions
(henceforth within-language), and L1–L2 and L2–L1 (henceforth between-
language) conditions, are combined for a statistical comparison of within- vs.
between-language conditions (but see Kroll & Stewart, 1994), among other factors
6 Bilingual Episodic Memory Processes 139
of interest (e.g., language proficiency or whether participants learned the L2 early

or late in life). Lack of statistical differences between the two language conditions
has been interpreted as supporting the view that bilinguals organize their two lan-
guages in one general memory system—a language-independent memory system.
Statistical differences across the within- vs. between-language conditions are typi-
cally interpreted as supporting separate stores or language-dependent systems
(Durgunoğlu & Roediger, 1987; Heredia, 2008, Heredia & Cieślicka, 2014). The
underlying assumption being that between-language conditions are distinctive and
encoded in multiple ways (one in each language) that involve language-specific
associative links resulting in better retrieval. This assumption is supported by the
classic finding that studying word categories in one language (e.g., Study List 1: cat,
dog then Study List 2: horse cow) and then tested in the same language (e.g., Recall
List 2 items) produces massive effects of proactive interference or the process in
which older/previous information (List 1) interferes forward in time with the recol-
lection of the newly learned (List 2) material (e.g., Coggin & Wickens, 1971; see
Heredia, 2008 for a review). In this case, retrieval from List 2 is significantly
affected. However, a change of language (e.g., Spanish) even if the same categories
are held constant (e.g., Study List 2: caballo, vaca), leads to release from proactive
interference, thus enhancing memory retrieval. Moreover, models such as Paivio’s
(2014) bilingual dual coding theory (DCT) provide special status to translation
equivalents as opposed to L1 or L2 words. Accordingly, [t]he language indepen-
dence hypothesis of bilingual DCT implies further that L1 and L2 translation equiv-
alents … should have an additive memory effect … such that each language code
contributes equally to their additive effect (p. 50). Briefly, Bilingual DCT hypothe-
sizes a memory structure in which each bilingual language is represented in sepa-
rate but interdependent language specific memories further interconnected to an
image system specializing on nonverbal objects (e.g., pictures) and events (Paivio,
2014). In its monolingual version, this model has been highly influential. However,
its theoretical impact in bilingual memory has been limited (see for example,
Cieślicka et al., 2013; Heredia, 2008; Jared, Pei Yun Poh, & Paivio, 2013). We now
turn to perhaps one of the most influential theories of episodic memory emphasizing
mental processes and depth of analysis.
Encoding
Levels of Processing
Levels of processing or depth-of-processing (Craik & Lockhart, 1972; Craik &

Tulving, 1975; see also Roediger, Gallo, & Geraci, 2002 for a review) is the theo-
retical approach wherein deep or meaningful processing of the to-be-learned infor-
mation leads to better retrieval than shallow or surface processing. Although it has
been shown that memory retrieval maybe a function of congruency between
encoding and retrieval in regards to contextual cues and processing demands as

posed by encoding specificity and transfer appropriate processing (see discussion
below), depth-of-processing effects are robust especially under traditional memory
tasks such as free recall and recognition known to tap conceptually driven processes
(Roediger, 2008). In the typical experiment, during the encoding phase, participants
are asked to analyze the physical structure of a word (e.g., Is the word in uppercase?
TABLE/table), determine the phonemic properties (e.g., Does the word rhymes with
WEIGHT? crate/market), or the word’s semantic properties (e.g., Is the word a type
of FISH? shark/heaven) or sentence questions (e.g., Would the word fit the sentence:
He met a_____ in the street? friend/cloud. Case and rhyming would be at the shal-
low, and categorization and sentence questions would be at the deep levels of pro-
cessing, respectively. In the now classic study, Craik and Tulving (1975) showed, as
predicted by levels of processing, that regardless of time, tasks that involved deeper
levels of processing during encoding lead to higher memory retrieval than shallow
processing. Accordingly, depth-of-processing leads to better remembering because
it creates a unique or distinct episodic memory trace that is different from other
memory traces or stimuli, and through elaboration it emphasizes meaning similari-
ties and relationships among interconnected concepts (Bradshaw & Anderson,
1982). Moreover, from an empirical point of view, of special interest is the distinc-
tion between incidental (unplanned learning that is potentially devoid of strategic
processing such as expectations or active rehearsal) and intentional tasks that
involve active and strategic processing. From a bilingual memory standpoint, this
distinction is critical, since it is more interesting to inspect the bilingual’s cognitive
inner workings in relation to ‘what’ and ‘how’ information is encoded without con-
scious awareness.
What are the effects of memory and language and depth-of-processing?
Surprisingly enough, only a handful of studies have addressed this issue. In one of
the first studies to directly address the effects of bilingualism and levels of process-
ing, Ayçiçegi-Dinn and Caldwell-Harris (2009; see also Vaid, 1988) had Turkish–
English (i.e., L1–L2) bilinguals assigned to three deep processing tasks: rate the
word for emotional intensity, a word association task (provide as many word associ-
ates as possible in 10 s), and an L1–L2 translation task that has been hypothesized
as involving semantic processing and mental effort, as well as an L2–L1 translation
that required less mental effort and thus less semantic processing (e.g., Kroll &
Stewart, 1994); and a shallow processing task (e.g., how many letters contain a
closed circle? [p, q, o, a, b, d, g]?). Stimuli were presented in English and Turkish
and included, positive (mother), negative (war), neutral (window), and taboo words
(breast). An incidental free recall task was administered, and neutral words were
subtracted from the other words types to create a recall difference score.
Task-specific analysis demonstrated that for the emotion-intensity rating task,
emotion memory effects were confined to the L1, namely, taboo and positive words.
No language differences were found for the letter counting task. However, a word
type main effect revealed higher retrieval for taboo words, followed by positive and
negative words. Although the word type effect for the translation task followed a
similar pattern as the letter counting task, a significant interaction revealed that
English, the L2, had stronger emotion-memory effects for taboo and positive words,
suggesting that an L2 is capable of demonstrating emotional-memory effects as
strong as L1 effects. This result is also theoretically interesting because, contrary to
the predictions, L2–L1 translations provided better retrieval. An alternative possi-
bility is that L2–L1 translation directions are indeed sensitive to conceptual/seman-
tic processing. Otherwise, ease of translation, as posed by Ayçiçegi-Dinn and
Caldwell-Harris, would have led to shallow-like processing and less retrieval. In
regards to the word association task, only word type effects were found, resembling
those of the letter counting and emotion-intensity tasks.
Figure 6.1 provides an overall summary of Ayçiçegi-Dinn and Caldwell-Harris’
(2009) results. Inspection of the graph suggests that depth-of-processing might have
had an effect on remembering, where counting letters leads to decrease in recall for
both languages, relative to the other tasks. Of the so-called deep processing, only
the emotion-intensity rating and the translation task revealed language differences.
What is clear from these results is that word type effects are robust, where retrieval
of taboo > positive > [negative = neutral] words. What is not clear is the effect of the
orienting task and whether the higher order interaction reported was driven by taboo
words. Future work in this area might consider utilizing properly controlled and
timed tasks such as those used by Craik and Tulving (1975), and reviewed here, that
have been shown to exhibit shallow (e.g., a structural task) and deep semantic pro-
cessing properties. Of particular interest is to determine if/when L1–L2 translations
require more mental effort, are more conceptually or semantically driven than L2–
0.65
0.6
Taboo-T
Taboo-E
0.55
Negative-T
0.5 Negative-E
Positive-T
0.45 Positive-E
Neutral-T
0.4 Neutral-E
Recall Percentage
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
Emotion Translation Association Count
Orienting Task
Fig. 6.1 Recall percentage as a function of orienting task and word type (adapted from Ayçiçegi-
Dinn & Caldwell-Harris, 2009, Fig. 2, p. 296)
L1 translations hypothesized to be effortless and more likely to exhibit shallow-

processing properties.
Francis and Gutiérrez (2012) compared L1 and L2 recognition between Spanish–
English bilinguals whose L1 (Spanish) or L2 (English) was the dominant language.
During the encoding conditions, participants were exposed to Spanish and English
concrete words. For the shallow conditions, participants indicated the number of
vowels in each word by pressing the number correct on a keyboard; for the deep
task, participants determined whether the object that the word referred to was a
natural or manufactured. A recognition task was administered after the enconding
conditions. Results are summarized in Fig. 6.2, where E–E is equivalent to English
being the dominant language and the task language or language of presentation; we
refer to this condition as the English dominant condition. E–S denotes English as
the dominant language and Spanish as the language of presentation, and the nondo-
mimant language, and so on. As can be seen from Fig. 6.2, all language conditions
benefitted equally from the deep processing task. Deep processing improved mem-
ory recognition relative to shallow processing. However, the less nondominant con-
ditions (E–S, S–E) revealed a significant advantage in the shallow processing
condition than the dominant language conditions (E–E, S–S).
Figure 6.3 summarizes the results for the response time (RT) to correct responses
on the recognition task. Levels of processing had an affect on RT responses, with
faster responses for words processed under deep than under shallow conditions.
Based on the reliable three-way interaction and the MSE reported, follow-up com-
parison using a least significant difference (LSD = 78.0, p = 0.05, two-tailed test),
shows that the only difference in the deep encoding condition was between S–S and
90.0 87.5
87 86.9
87.5 86
85.0
82.5
Deep
Percentage Yes Responses
80.0
77.5 Shallow
75.0
72.5
70.0 67.5 68.1
67.5
65.0
62.5 61.1
60.0 59
57.5
55.0
52.5
50.0
E–E E–S S–S S–E
Language dominance and task language
Fig. 6.2 Percentage recognitiom as a function of language dominance and task language and
levels of processing (Adapted from Francis & Gutiérrez, 2012, Table 1, p. 500)
1275 1250
1250
1224
1225
1200
1175
Response Time (in ms)
1150
1125 1110
1100
1075 1050 Deep
1050 Shallow
1023
1025 1009
1000 975
975
950 941
925
900
E–E E–S S–S S –E
Language Dominance and Task Language
Fig. 6.3 Response time as a function of language dominance and task language and levels of
processing (adapted from Francis & Gutiérrez, 2012, Table 1, p. 500)
S–E; it is not clear as to why these differences in RT between these two conditions.
For the shallow encoding condition, the significantly large RTs’s for the E–E and
S-S language conditions were essentially driving the higher order interaction.
Overall, the data reported by Francis and Gutiérrez (2012) provide a clear dem-
onstration of the interaction of language (i.e., bilingualism) and levels of processing
(but see Francis, Fernandez, & Bjork, 2013). Although the authors provide a reason-
able theoretical explanation for their findings, we note three important issues that
might have influenced Francis and Gutiérrez’ (2012) results. First, it is unclear as to
the classification of the bilinguals’ languages into dominant vs. nondominant lan-
guages. Accordingly, language dominance was based on self-reported relative pro-
ficiency (p. 498) and the percentage of each language used over the preceding
month, without specifying the aspect of language assessed (e.g., reading, writing,
spoken; see also Francis et al., 2013). However, we do agree that language domi-
nance is a much more theoretically interesting bilingual classification than the tra-
ditional L1 vs. L2. In fact, Dunn and Fox Tree’s (2009) bilingual dominance scale
has been utilized in some of our work (Heredia & Cieślicka, 2016) to more objec-
tively discern between the bilinguals’ dominant language (L1 or L2), and whether
both languages are equally dominant (i.e., balanced bilinguals). For example, given
the similar patterns in the language conditions across Figs. 6.2 and 6.3, it could very
well be argued that the bilinguals being described are likely to be balanced bilin-
guals. Second, it is not at all clear as to why word frequency was not manipulated or
controlled for, given that one of the authors’ arguments involved word frequency as
a tenable factor explaining bilingual memory recognition. Word age-of-acquisition
(Kuperman, Stadthagen-Gonzalez, & Brysbaert, 2012) might be another factor to
consider to better understand the influence of word level effects in bilingual memory.
Third, it is also unclear as to why the encoding tasks were so different. For the shal-
low task, participants indicated the number of vowels by pressing the appropriate
number on the keyboard; for the deep task, participants were required to press n for
natural or m for manufactured object. Perhaps, the shallow task could have been
made more comparable by simply asking participants to determine if the presented
word (YES/NO) appeared in uppercase. It is possible that the extra step in locating
the appropriate number of vowels on the keyboard introduced additional cognitive
demands on the bilingual’s attentional resources.
Overall, levels of processing provide a simple and powerful theoretical approach
to further investigate the interaction of language (bilingualism, multilingualism)
and memory. As argued before, it has been posed that L1–L2 translation directions
take additional mental effort and that they are sensitive to conceptually/semantically
driven processes. That is, are L1–L2 translations more likely to benefit from a deep
processing orienting task than L2–L1 translations that are hypothesized to be auto-
matic and effortless? (cf. Kroll & Stewart, 1994; see also Francis et al., 2013).
Clearly, these questions could be empirically and directly addressed using this very
promising classic memory approach to learning and memory. We now turn to the
generation effect, another classic memory phenomenon that can be seen as an exten-
sion of levels of processing, and that has received limited attention in the bilingual
memory arena.
The Generation Effect
The generation effect is the general memory phenomenon in which a target word
that is generated/produced is significantly better remembered later than if the same
word was read (e.g., Bertsch, Pesta, Wiscott, & Mcdaniel, 2007; MacLeod, Pottruff,
Forrin, & Masson, 2012; Slamecka & Katsaiti, 1987). Although different theoretical
formulations have been proposed to account for the generation effect (e.g., greater
effort, generated items steal rehearsal from others), the current account (i.e., multi-
factor view) holds that generation (during encoding) increments item distinctive-
ness by emphasizing item-specific processing, and that generation differentiates
items from one another (MacLeod et al., 2012, p. 6934). In general, the generation
effect appears to be a robust and consistent effect (Bertsch et al., 2007). In the typi-
cal monolingual task, participants are presented with a word pair (e.g., attorney-
lawyer) in which they simply read both words aloud (read condition), or a stimulus
is followed by a cue (e.g., attorney-l) that the participant uses to generate the target
word (e.g., lawyer), in this case a synonym. The bilingual counterpart is essentially
the same except the read/generate conditions include translation equivalents for the
read (e.g., attorney-abogado) and generate (e.g., attorney-a) conditions. Although
the generation effect failed to materialize under conditions in which participants
expected a memory task (intentional learning), and other methodological issues
such as utilizing a between subjects design for the task condition, and other issues
related to the participants language proficiency (e.g., Slamecka & Katsaiti, 1987;
see also Durgunoğlu & Roediger, 1987), other studies (e.g., O’Neill, Roy, &
Trembaly, 1993) have indeed demonstrated a bilingual (translation-based) genera-
tion effect in recall and recognition. In one experiment, O’Neill et al. (1993) had
French–English bilinguals read and copy word targets in English or French (repeti-
tion condition), read translation equivalents and copy the second word (translation
condition), and read a word in one language and write its translation equivalent
(translation-generate condition). Results revealed no differences in recall and recog-
nition between repetition and translation condition. However, the translation-
generate condition clearly enhanced both recall and recognition. Additional studies
demonstrated the translation-based generation effect to be robust under incidental
task conditions, but elusive under intentional experimental manipulations. Basi,
Thomas, and Wang (1997) report similar findings in relation to bilingual generation
effects and intentional vs. incidental tasks. Interestingly, Basi et al. report larger
generation effects for compound (learning both L1 and L2 before 10 years of age)
as opposed to coordinate (learning the L2 after 10 years of age in a different setting
than the L1) bilinguals (cf. Heredia & Cieślicka, 2014). According to Basi et al.,
these differences among these bilinguals were possibly due to greater differences in
allocation of attention to read than to generate targets (cf. Francis & Gutiérrez,
2012).
Though the consistency of the generation effect extends to bilingualism, and
particularly to a translation-based generation effect, clearly this area of research
would benefit from further investigating differences between intentional and inci-
dental tasks controlling for such factors as language dominance and bilingual types
(e.g., compound vs. coordinate), as suggested by Basi et al. (1997). Again, we can-
not emphasize enough the importance of implementing more objective means to
assess language dominance and/or language proficiency. Moreover, the generation
effect might prove fruitful in determining possible enhancing encoding properties
between L1–L2 and L2–L1 translation directions. We now turn to the distributed-
practice and repetition effects.
The Distributed-Practice Effect
The distributed-practice or spacing effect is the general finding in which distributed

or spaced learning presentations lead to better retention than massed presentations
(e.g., Cepeda, Pashler, Vul, Wixted, & Rohrer, 2006).
Figure 6.4 shows a sample of a partial bilingual experimental list manipulating
lag or spacing between different bilingual conditions (Glanzer & Duarte, 1971;
Heredia & McLaghlin, 1992; Paivio, Clark, & Lambert, 1988). A lag 0 SE denotes
a Spanish–English translation (e.g., gis-chalk) with no intervening items in between.
A lag 0 would correspond to a massed presentation where all information is pre-
sented at once or separated by zero intervening items—in this case, gis is immedi-
ately followed by chalk. At the applied level, massed presentations would be
equivalent to cramming for an exam in bilingual cognitive psychology, for example,

the night before. A lag 1 EE denotes an English–English repetition (e.g., milk-milk)
with one intervening Spanish item (e.g., huella) in between. A lag 5 ES represents
an English–Spanish translation with 5 intervening items between the word pair.
Other important conditions to note are the Nonrep E denoting an English item (e.g.,
desk) that appears only once, or is not repeated. The nonrepeated condition is
included as a comparison to the repeated items presented at the different lags, and
as a measurement of the repetition effect where multiple exposure to a to-be-learned
stimulus, during the learning phase, increases remembering, relative to a nonre-
peated stimulus. All within- and between-language conditions appeared in all
experimental manipulations (i.e., lag, repetition, nonrepeated). Using materials
similar to those described in Fig. 6.4, after the encoding phase, Spanish–English
bilinguals in Glanzer and Duarte’s (1971) experiment were given a free recall task.
Results are summarized in Fig. 6.5.
Figure 6.5 reveals three noteworthy effects: (1) the repetition effect, (2) the distrib-
uted practice effect, where spaced repetitions are remembered better than massed rep-
etitions (0 distance); and (3) greater recall for between-language conditions than
within-language conditions up to distance three. By distance five, recall for both lan-
guage conditions converged. These bilingual memory effects were further replicated
by Heredia and McLaghlin (1992) and further expanded by Paivio et al. (1988) show-
ing similar trends for concrete and abstract words, in which case, concrete words were
better recalled than abstract words (i.e., the concreteness effect; Kaushanskaya &
Rechtzigel, 2012; see also, De Groot, 1992) for both language conditions. Another
Fig. 6.4 Sample of an

experimental list as a
function of distributed-
practice, repetition,
within- and between-
languages (adapted from
Heredia & McLaghlin,
1992)
Fig. 6.5 Proportion of words recalled as a function of between- and within-language and distance
(adapted from Glanzer & Duarte, 1971, Fig. 1, p. 628)
important aspect to point out, in relation to the recurring issue of bilingual storage, is
that depending on whether we look at short or long distances, one would find support
for either hypothesis. That is, recall for short distances (e.g., 0–2) would support a
two-memory store model (i.e., reliable differences for between- vs. within-language
conditions), and long distances (e.g., 4–5) would support a one-memory store repre-
sentation for bilinguals (i.e., no reliable differences for between- vs. within-language
conditions). Given this disparity between short and long distances, it would be highly
informative if these same bilingual language effects prevailed under controlled
(spaced) timing conditions (e.g., 0, 1, 2, 5 s). Empirically, these results would be sug-
gestive that in addition to controlling for the type of memory task (see discussion on
transfer appropriate processing below), exposure time during the encoding phase
should be controlled as well. For example, inspection of item exposure in previous
bilingual memory experiments is indeed highly variable.
Thus, it appears that these classic memory learning/encoding effects show … an
empirical regularity, an established functional relation, one that holds widely (ide-
ally, universally) across manipulations of other variables [such as language and
bilingualism] (Roediger, 2008, p. 227). Clearly, more empirical work in bilingual-
ism is needed to further establish these empirical regularities by manipulating such
factors as word frequency, orienting tasks (deep vs. shallow processing), whether
the task is incidental vs. intentional, and carefully controlling for the bilingual’s
language fluency or dominance. So far, if we were to name one of the most pressing
issues in bilingual research, without a doubt, it would be the lack of consistency in
measuring language proficiency and language dominance. Let us now turn into the
fascinating domain of memory retrieval and bilingualism and address issues per-
taining to how bilinguals recover information from their two languages, and whether
some information in the bilingual’s world is language-dependent.
Retrieval
As an introduction, consider a bilingual speaker describing how to best prepare a

hamburger: para preparar una buena hamburguesa, necesitamos … pickles (to pre-
pare a good hamburger, we need … pickles; the ellipsis is included as suggestive of
a time related pause). Although the uttered bilingual string of words would be a
perfectly grammatical sentence (e.g., Heredia & Altarriba, 2001; Lederberg, &
Morales, 1985; Poplack, 1980) and would necessitate no further explanation, for
our purposes, it would be suggestive of a retrieval failure in which the Spanish
equivalent pepinillo is not accessible or unrecoverable (i.e., an accessibility issue)
due to such factors as more exposure to an English environment (e.g., community
and school), and the Spanish equivalent being of a very low word frequency/famil-
iarity in Spanish (see for example, Kreiner & Dagani 2015; Pureza, Soares, &
Comesaña, 2016; see also Maruyama & Ginzburg, 2016). Alternatively, it could
also be possible that pepinillos is not part of the bilingual’s vocabulary repertoire, in
which case, it would be an availability issue (Tulving & Pearlston, 1966). Thus, a
retrieval failure occurs when the information is available (i.e., it is present in the
memory system), but it cannot be recovered or is not accessible at a particular time.
Or simply put, the information is not present; that particular lexical item is not part
of the bilingual’s semantic memory. Indeed, this is why it is critical that bilingual
researchers assure that their bilingual participants are familiar with the stimuli being
manipulated (by norming their experimental stimuli with their respective bilingual
population in terms of word frequency and familiarity, among other variables).
Moreover, consider the subjective experience in which the smell of onions and
cilantro brings up memories about Sundays (usually around 9:00 AM) when the
first author’s mother would take him to the Mercado Municipal (Main City Market).
Or the Submarinos Marinela (Twinkie The Kid snacks!) that reminds him of grandpa
(the same one in the Charanda autobiographical memory!) who used to bring those
magnificent treats from the big city. The smell of onion and cilantro, and the snacks
are retrieval cues or cues that allow him to access and make available those very
specific memories about his younger years.
Now consider the story by Li (2017), a native Chinese speaker that grew up in
China and later in life became an English writer. Li writes,
Over the years, my brain has banished Chinese. I dream in English. I talk to myself in
English. And memories—not only those about America but also those about China; not only
those carried with me but also those archived with the wish to forget—are sorted in English
(para. 21).
Why is Li (2017) unable to remember her younger memories in the L1, even in
the presence of what would be considered strong retrieval cues (e.g., black rotary
telephone, Soviet-like buildings)? Li explains: … To be orphaned from my native
language felt, and still feels, like a crucial decision … (para. 21). My abandonment
of my first language is personal, so deeply personal that I resist any interpretation
… (para. 11). What is more interesting of Li’s account is that she finds herself trans-
lating her originally acquired memories in Chinese from English to Chinese. Are
Li’s memories in L1 completely forgotten? In this section we review bilingual lan-

guage effects and encoding specificity, and the impact of transfer appropriate pro-
cessing (e.g., Morris, Bransford, & Franks, 1977) in bilingual memory storage. We
specifically suggest how transfer appropriate processing can serve as a viable mem-
ory paradigm to contribute to a better understanding of forgetting and bilingualism
(e.g., Isurin, 2000; Isurin and McDonald, 2001; Levy, McVeigh, Marful, &
Anderson, 2007; Peynircioğlu & Göksen-Erelçin, 1987; but see Runnqvist & Costa,
2012), and whether forgetting or suppression of a language can be induced inten-
tionally (i.e., motivated forgetting; e.g., Hockley, Ahmad, & Nicholson, 2016)), as
in the case of Li (2017).
Encoding Specificity
According to Tulving and Thomson (1973), in its broadest form, the encoding spec-
ificity principle … asserts that only that can be retrieved that has been stored, and
that how it can be retrieved depends on how it was stored (p. 359). Moreover,
[W]hat is stored about the occurrence of a word in an experimental list is the information
about the specific encoding of that word in that context in that situation. This information
may or may not include the relation that the target word has with some other word in the
semantic system. If it does, that other word may be an effective retrieval cue. If it does not,
the other word cannot provide access to the stored information because its relation to the
target is not stored. Thus, the effectiveness of retrieval cues depends on the properties of the
trace of the word event in the episodic system (p. 359).
Thus, retrieval will be better if the context during retrieval (cue A’) matches the
context of encoding (cue A), as shown in Fig. 6.6; a Match between retrieval and
encoding (cue AA’ or BB’) will lead to better memory retrieval than a Mismatch
(cue AB’ or BA’).
So, according to encoding specificity, recall of the target word COLD would be
substantially better if the retrieval cue ground (which is a weak cue) is used during
the encoding phase (a match in Fig. 6.6) than if the retrieval cue hot (which is a
strong cue based on its associative properties to COLD; a mismatch in Fig. 6.6) is
used. Godden and Baddeley (1975), in a now classic demonstration of context-
dependent memory (a corollary of encoding specificity), had participants learn
words on land (dry) or underwater (wet); they were then asked to recall the learned
Fig. 6.6 Encoding
specificity as a function of
encoding and retrieval
(adapted from Tulving,
1983, p. 220)
words on land or underwater. As predicted, recall was significantly better when the
context of retrieval and the context of encoding matched (dry-dry, wet-wet), than
when they did not match (dry-wet, wet-dry): dry-dry > dry-wet and wet-wet > wet-
dry. These results in conjunction with those by transfer appropriate processing,
another manifestation of encoding specificity and described below, argued against
the levels of processing tenets suggesting, for example, that deep processing was not
always better than shallow processing during retrieval. Accordingly, shallow pro-
cessing could be as effective, as long as the learning-and-retrieval conditions
involved the same contextual properties or memorial processes (Roediger, 2008).
How is encoding specificity related to bilingual storage and remembering? Is
there any evidence of language-dependent memory? That, is it possible that infor-
mation learned in one language (L1 or L2) can only be accessed via the language of
encoding or learning, for a bilingual speaker? As Tulving and Thomson (1973)
rightly suggest, if the information was learned, it can and should be retrieved if the
appropriate encoding context (i.e., language or cues) is reinstated during the remem-
bering process.
Marian and colleagues (e.g., Francis et al., 2013; Marian & Fausey 2006; Marian
& Neisser, 2000; Schrauf, & Rubin, 2000; cf. Schwanberg, 2010) have demon-
strated language-dependent memory effects in which, for example, retrieval is sig-
nificantly better if the language of study/encoding matches the language of retrieval.
In Marian and Fausey’s experiment, Spanish–English bilinguals were better in
remembering information about stories they read, but only if the language of instruc-
tion (e.g., Spanish) matched the language of testing (e.g., Spanish). This general
finding was true for Spanish and English conditions, but only for balanced bilin-
guals. Although Marian and colleagues provide a clear demonstration of the gener-
alizability of encoding specificity to language and bilingualism, in particular, it
would be more impressive if, for example, it is shown that certain behavioral aspects
or memories learned earlier in life, and possibly in another country, can be triggered
by the language of encoding. Such demonstration must include specific instances
that are directly tied (i.e., automatic) to the language/culture that could be verifiable,
such as for example, reciting verses of the L1 and L2’s national anthems, prayers or
the different ways in which the national flags are saluted. Marian and Kaushanskaya
(2007) provide such demonstration. In their study, Mandarin–English bilinguals,
who were more proficient in Mandarin, responded to questions pertaining to every-
day knowledge that had been learned in their respective languages (and countries)
and had one possible correct response in each language (e.g., names of lakes, actors,
historical facts). Overall, Marian and Kaushanskaya’s results revealed that memo-
ries learned in Mandarin were remembered better if tested in Mandarin, and like-
wise for the English language condition. A language mismatch in the learning and
retrieval conditions lead to a decrease in memory recall accuracy. More impressive,
however, was the finding showing that when asked to Name a statue of someone
standing with a raised arm while looking into the distance, bilinguals responded
The Statue of Mao, if asked in Chinese, and the Statue of Liberty, if asked in English.
These results contrast with Li’s (2017) account described above where all her mem-
ories from China were remembered in English and not Chinese, the language of
encoding, even in conditions in which encoding cues were available to her (e.g.,
dreams, family, buildings in China). Li’s case may very well be a case of retroactive
interference (Isurin, 2000; Isurin and McDonald, 2001) in which the learning of the
new language (L2) interferes (back in time) with the retrieval of the old information.
Alternatively, it could be a case of motivated forgetting in which L1 is intentionally
suppressed due to traumatic experiences. In any case, encoding specificity would
predict that exploring such traumatic experiences L1 would more likely restore L1
accessibility, provided that affect-related issues are resolved (see for example,
Schwanberg, 2010).
Indeed, encoding specificity is a simple and powerful tool to investigate the theo-
retical possibility that memory for coordinate bilinguals (i.e., learning L2 after
10 years of age in a different environment than the L2) is more likely to be stored in
a language-dependent fashion than compound bilinguals whose languages are
learned simultaneously. In other words, is it possible to distinguish between
language-dependent and language-independent (i.e., language-free) memories?
That is, to what extent is language context guiding or driving memory access as
opposed to language-free retrieval cues (e.g., pictures, smells, melody)? The scent
of an onion, for example, might trigger an image of a remote super market that was
experienced in one’s L1. Thus, how is this language-free retrieval cue accessing that
information from the L1? Indeed, these are issues awaiting further research. We
now discuss the impact of transfer appropriate processing and the resolution of
bilingual memory storage.
Transfer Appropriate Processing
Transfer appropriate processing, another formulation of encoding specificity, pro-

poses that retrieval is a function of the properties (i.e., processes, procedures) of the
task, and whether encoding information is appropriately transferred to performance
on the retrieval test (Roediger, 2008). In direct contrast to Craik and Tulving (1975),
Morris et al. (1977) went on to show that when both encoding and retrieval tasks
tapped similar processes (elaborative or phonemic: AA’ or BB’ match in Fig. 6.6), the
likelihood of retrieval was high. A mismatch (AB’ or BA’ in Fig. 6.6) between encod-
ing and retrieval led to a significant decrease in retrieval, especially for the meaning-
based elaborative condition. One of the best demonstrations of transfer appropriate
processing and task demands is provided by Weldon and Roediger’s (1987) classic
findings. Using a free recall task, Weldon and Roediger replicated the picture superi-
ority effect wherein pictures are remembered better than words. However, when a
word-fragment completion task was used, words outperformed pictures, thus revers-
ing the picture superiority effect. Briefly, in a typical word-fragment completion task,
after studying words and pictures, participants are presented with word-fragments
(e.g., _yr_mi_ for pyramid) and fragmented pictures and their task is to solve the frag-
ments by completing the word or picture fragment. Why these differences between
the two tasks? Differences in memory performance were taken to suggest that each
task drew on different memory processes. Free recall or direct memory tasks (e.g.,
recognition, cued-recall) that required deliberate or conscious awareness of past
events were hypothesized to tap elaborative or conceptually/semantically driven-pro-
cesses; word-fragment completion or indirect memory tasks (e.g., word stem comple-
tion, lexical decision) that required unconscious recollection tapped perceptual or
data-driven processes (see for example, Roediger, 2008; Weldon, Roediger, & Challis,
1989). How are these results related to bilingual memory?
In one of the most significant demonstrations in the bilingual memory literature,
Durgunoğlu and Roediger (1987) were able to show retrieval differences in bilin-
gual memory as a function of type of task. Results from free recall and recognition
revealed retrieval patterns consistent with language-independent views (i.e., a single
memory for both languages), and results from the word-fragment completion task
revealed patterns similar to those proposed by language-dependent views (i.e., two
separate memories for each language). This pattern of results was replicated by
Heredia and McLaghlin (1992) shown clear retrieval differences between a free
recall task and a word-fragment identification task. Results from the free recall task
exhibited the distributed-practice effect, discussed above (se Fig. 6.5); for the word-
fragment identification task, the distributed-practiced effect was manifested, but
only for the language at test that matched the language at encoding (i.e., language-
dependent effects), in this case, English, the L2. Overall, transfer appropriate pro-
cessing was able to account for the various contradictory bilingual experimental
results showing sometimes support for the interdependent (shared) or independent
(two stores) hypotheses. Direct memory tasks (e.g., free recall, recognition) were
more likely to support the shared memory view; indirect memory tasks (e.g., word-
fragment completion, lexical decision) supported the two-memory view. In other
words, support for either hypothesis was a function of task and processing demands
(see also Fernandes, Wammes, & Hsiao, 2013).
Transfer appropriate processing is a simple and powerful paradigm to further
investigate bilingual memory. As an empirical tool, it could be used to design sound
and interpretable experiments in such a way that language is held constant during
both the encoding and retrieval conditions. This is critical for recognition and other
tasks such as word-fragment completion. In some of our current work, for example,
we are investigating the extent to which bilingual speakers might forget/suppress
information from one language intentionally. Using an incidental learning approach,
bilinguals are asked to read a series of Spanish words. After exposure, they are sim-
ply told that the words they read were simply practice trials and that they should
forget them and clear their mind. Then they are exposed to English words that are
translation equivalents of the Spanish words read in the first list. After completion
of the English list, participants are given an English recognition task. Critically, and
consistent with transfer appropriate processing, the recognition task is in English.
This bilingual condition would be equivalent to a Spanish–English translation (i.e.,
List 1: read casa; List 2: read house; test recognition of house). The English–Spanish
bilingual condition follows the same procedure. At issue is whether List 1 (casa for
Spanish–English) facilitates the recognition (response time and percentage remem-
bered) of the English word relative to a control condition (i.e., read house; recognize
house; see for example, López & Young, 1974). We ask whether bilinguals are able
to intentionally suppress information from one language, and whether bilingual
retrieval is language-independent (see for example, Lauro & Schwartz, 2017).
Moreover, by matching the learning and retrieval conditions and independently
varying language conditions (Spanish–English vs. English–Spanish), we will be
able to discern bilingual processing differences, if any, between Spanish–English
and English–Spanish (cf. Francis et al., 2013).
Conclusion
Although storage (i.e., how bilinguals represent languages) is a topic that must be
addressed in any discussion of bilingual memory, this chapter is about encoding
(learning) and retrieval in episodic memory. Our overall purpose was to provide a
general overview of some of the classic encoding (e.g., levels of processing, the gen-
eration effect, the distributed-practice and the repetition effect) and retrieval (e.g.,
encoding specificity, and transfer appropriate processing) approaches to episodic
memory and gauge their universality (i.e., generalizability) to bilingualism. As
reviewed here, these classic encoding and retrieval effects interact with language and
generalize to bilingualism. Although limited, the bilingual research findings reviewed
here suggest comparable mechanisms underlying the bilingual and monolingual epi-
sodic memory system. In relation to encoding, bilingual memory improved by fac-
tors (e.g., deep-processing, mental elaboration, spacing practice, and repetition)
known to enhance retrieval. In relation to retrieval, bilingual memory appears lan-
guage-dependent and sensitive to conceptually- and data-driven processes. There is a
general tendency for between-language (i.e., translation) conditions to enhance
memory retrieval, relative to within-language (monolingual) conditions. Whether
translation direction (Spanish–English vs. English–Spanish) involves differential
processing as posed by some bilinguals models (e.g., Kroll & Stewart, 1994), remains
to be seen; it is an open question that can be directly addressed by systematically
manipulating depth-of-processing, for example, and considering task demands.
Considering the limitations of early bilingual research and the lack of bilingual
norms (e.g., word frequency, familiarity, concreteness), the classification of bilinguals
(e.g., dominant, compound vs. coordinate, proficiency) seems to be a common denomi-
nator between early and current research. In our view, Dunn and Fox Tree’s (2009) is a
valid language dominance scale with discriminant validity to distinguish between lan-
guage dominance and balanced bilinguals (i.e., equal dominance across both lan-
guages). Moreover, the importance of stimuli norming among the population of interest,
even with existing published bilingual norms, cannot be stressed enough. In addition to
controlling for word frequency, in our own work, we also norm our stimuli in terms of
subjective familiarity, since Spanish words, for example, that are high in word fre-
quency, may not be familiar to bilinguals in the respective population of interest.
Although the bilingual storage question has concerned us for the last 40 years or so, the
bilingual episodic memory is an open field waiting for us to ask the right question(s).
Acknowledgement The authors are grateful to Catherine L. Caldwell-Harris for comments and
suggestions on an earlier version of this paper.
References
Ayçiçegi-Dinn, A., & Caldwell-Harris, C. L. (2009). Emotion-memory effects in bilingual speak-
ers: A levels-of-processing approach. Bilingualism: Language and Cognition, 12(3), 291–
3030. doi:10.1017/S1366728909990125
Basi, R. K., Thomas, M. H., & Wang, A. Y. (1997). Bilingual generation effect: Variations in
participant bilingual type and list type. The Journal of General Psychology, 124(2), 216–222.
Basnigh-Brown, D. (2014). Models of lexical access and bilingualism. In R. R. Heredia &
J. Altarriba (Eds.), Foundations of bilingual memory (pp. 85–107). New York: Springer.
Bertsch, S., Pesta, B. J., Wiscott, R., & Mcdaniel, M. A. (2007). The generation effect: A meta-
analytic review. Memory & Cognition, 35(2), 201–210.
Bradshaw, G. G., & Anderson, J. R. (1982). Elaborative encoding as an explanation of levels of
processing. Journal of Verbal Learning and Verbal Behavior, 21, 165–174.
Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., & Rohrer, D. (2006). Distributed practice in
verbal recall tasks: A review and quantitative synthesis. Psychological Bulletin, 132, 354–380.
doi:10.1037/0033-2909.132.3.354
Cieślicka, A. B., Morales, J., & Heredia, R. R. (2013). Dual-coding theory revisited: Concreteness
and picture superiority effects in bilingual incidental memory tasks. In E. Mańczak-Wohlfeld
& M. Jodłowiec (Eds.), Exploring the microcosm of language teaching and learning: A fest-
schrift on the occasion of 70th birthday of Professor Anna Niżegorodcew (pp. 23–39). Kraków,
Polland: Jagiellonian University Press.
Coggin, J., & Wickens, D. D. (1971). Proactive interference and language change in short-term
memory. Journal of Verbal Learning and Verbal Behavior, 10, 453–458.
Craik, F. I. M., & Lockhart, R. S. (1972). Levels of processing: A framework for memory
research. Journal of Verbal Learning & Verbal Behavior., 11(6), 671–684. doi:10.1016/
S0022-5371(72)80001-X
Craik, F. I. M., & Tulving, E. (1975). Depth of processing and the retention of words in episodic
memory. Journal of Experimental Psychology: General, 104, 268–294.
De Groot, A. M. B. (1992). Determinants of word translation. Journal of Experimental Psychology:
Learning, Memory, and Cognition, 18, 1001–1018.
Dunn, A. L., & Fox Tree, J. E. (2009). A quick, gradient bilingual dominance scale. Bilingualism:
Language and Cognition, 12, 273–289. doi:10.1017/S1366728909990113
Durgunoğlu, A. Y., & Roediger, H. L. (1987). Test differences in accessing bilingual memory.
Journal of Memory and Language, 26, 377–391.
Francis, W., Fernandez, N. P., & Bjork, R. A. (2013). Conceptual and non-conceptual repetition
priming in category exemplar generation: Evidence from bilinguals. Memory, 18(7), 787–798.
http://dx.doi.org/10.1080/09658211.2010.511234
Francis, W., & Gutiérrez, M. (2012). Bilingual recognition memory: Stronger performance but
weaker levels-of-processing effects in the less fluent language. Memory & Cognition, 40, 496–
503. doi:10.3758/s13421-011-0163-3
Fugazza, C., Pogány, A., & Miklósi, A. (2016). Recall of others’ actions after incidental encod-
ing reveals episodic-lie memory in dogs. Current Biology, 25, 1–5. dx.doi.org/10.1016/j.
cub.2016.09.057
Godden, D. R., & Baddeley, A. D. (1975). Context-dependent memory in two natural environ-
ments: On land and underwater. British Journal of Psychology, 66(3), 325–331.
Glanzer, M., & Duarte, A. (1971). Repetition between and within languages in free recall. Journal
of Verbal Learning and Verbal Behavior, 10, 625–630.
Fernandes, M. A., Wammes, J. D., & Hsiao, J. H. (2013). Representation of linguistic informa-
tion determines its susceptibility to memory interference. Brain Sciences, 3, 1244–1260.
doi:10.3390/brainsci3031244
Heredia, R. R. (2008). Mental model of bilingual memory. In J. Altarriba & R. R. Heredia (Eds.),
An introduction to bilingualism: Principles and processes (pp. 39–67). New York: Psychology
Press.
Heredia, R. R., & Altarriba, J. (2001). Why do bilinguals code-switch? Current Directions in
Psychological Science, 10(5), 164–168.
Heredia, R. R., & Cieślicka, A. B. (2014). Bilingual storage: Compound-coordinate and deri-
vates. In R. R. Heredia & J. Altarriba (Eds.), Foundations of bilingual memory (pp. 11–40).
New York: Springer.
Heredia, R. R., & Cieślicka, A. B. (2016). Metaphoric reference: An eye movement analysis
of Spanish-English and English-Spanish bilingual readers. Frontiers in Psychology, 7, 439.
https://doi.org/10.3389/fpsyg.2016.00439
Heredia, R. R., Cieślicka, A. B., & Altarriba, J. (2016). Introduction to bilingual research meth-
ods. In R. R. Heredia, & A. B. Cieślicka (Eds.), The bilingual mind and brain series: Vol.
1. Methods in bilingual reading comprehension research (pp. 1-9). New York: Springer. doi:
10.1007/978-1-4939-2993-1_1
Heredia, R. R., & McLaghlin, B. (1992). Bilingual memory revisited. In R. J. Harris (Ed.),
Advances in psychology: Cognitive processing in bilinguals (Vol. 83, pp. 91–103). New York:
Elsevier.
Hockley, W. E., Ahmad, F. N., & Nicholson, R. (2016). Intentional and incidental encoding of item
and associative information in the directed forgetting procedure. Memory & Cognition, 44,
220–228. doi:10.3758/s13421-015-0557-8
Isurin, L. (2000). Deserted island or a child’s first language forgetting. Bilingualism: Language
and Cognition, 3(2), 151–166.
Isurin, L., & McDonald, J. L. (2001). Retroactive interference from translation equivalents:
Implications for first language forgetting. Memory & Cognition, 29(2), 312–339.
Jared, D., Pei Yun Poh, R., & Paivio, A. (2013). L1 and L2 picture naming in Mandarin-English
bilinguals. A test of bilingual coding theory. Bilingualism: Language and Cognition, 16(2),
383–396. https://doi.org/10.1017/S1366728912000685
Lauro, J., & Schwartz, A. I. (2017). Bilingual non-selective lexical access in sentence contexts:
A meta-analytic review. Journal of Memory and Language, 92, 217–233. http://dx.doi.
org/10.1016/j.jml.2016.06.010
Kreiner, H., & Dagani, T. (2015). Tip-of-the-tongue in a second language: The effects of brief first-
language exposure and long-term use. Cognition, 137, 106–114.
Kaushanskaya, M., & Rechtzigel, K. (2012). Concreteness effects in bilingual and monolingual
word learning. Psychonomic Bullutin Review, 19, 935–941. doi:10.3758/s13423-012-0271-5
Kroll, J. F., & Stewart, E. (1994). Category interference in translation and picture naming:
Evidence for asymmetric connections between bilingual memory representations. Journal of
Kuperman, V., Stadthagen-Gonzalez, H., & Brysbaert, M. (2012). Age-of-acquisition ratings
for 30,000 English words. Behavioral Research Methods, 44(4), 978–990. doi:10.3758/
s13428-012-0210-4
Lahdenperä, M., Mar, K. U., & Lumma, V. (2016). Nearby grandmother enhances calf survival and
reproduction in Asian elephants. Scientific Reports, 6, 27213. doi:10.1038/srep27213
Lederberg, A. R., & Morales, C. (1985). Code switching by bilinguals: Evidence against a third
grammar. Journal of Psycholinguistic Research, 14(2), 113–136.
Levy, B. J., McVeigh, N. D., Marful, A., & Anderson, M. C. (2007). Inhibiting your native language:
The role of retrieval-induce forgetting during second-language acquisition. Psychological
Science, 18(1), 29–34.
Li, Y. (2017). To speak is to blunder: Choosing to renounce a mother tongue. In The new Yorker.
Retrieved from http://www.newyorker.com/magazine/2017/01/02/to-speak-is-to-blunder
López, M., & Young, R. K. (1974). The linguistic interdependence of bilinguals. Journal of
Experimental Psychology, 102(6), 981–983.
MacLeod, C. M., Pottruff, M. M., Forrin, N. D., & Masson, M. E. J. (2012). The next generation:
The value of reminding. Memory & Cognition, 40, 693–702. doi:10.3758/s13421-012-0182-8
Marian, V., & Fausey, C. M. (2006). Language-dependent memory in bilingual memory. Applied
Cognitive Psychology, 20(8), 1025–1047. doi:10.1002/acp.1242
Marian, V., & Kaushanskaya, M. (2007). Language content guides memory content. Psychonomic
Bulletin & Review, 14(5), 925–933.
Marian, V., & Neisser, U. (2000). Language-dependent recall of autobiographical memories.
Journal of Experimental Psychology, 129(3), 361–368.
Maruyama, T., & Ginzburg, J. (2016). Self addressed questions and filled pauses: A cross-linguistic
investigation. Journal of Psycholinguistic Research, 1–18. doi:10.1007/s10936-016-9468-5
Morris, C. D., Bransford, J. D., & Franks, J. J. (1977). Levels of processing versus transfer appro-
priate processing. Journal of Verbal Learning and Verbal Behavior, 16, 519–533.
O’Neill, W., Roy, L., & Trembaly, R. (1993). A translation-based generation effect in bilingual
recall and recognition. Memory & Cognition, 21(4), 488–495.
Paivio, A., Clark, J. M., & Lambert, W. E. (1988). Bilingual dual-coding theory and semantic
repetition effects on recall. Journal of Experimental Psychology: Learning, Memory, and
Cognition, 14(1), 163–172.
Paivio, A. (2014). Bilingual dual coding theory and memory. In R. R. Heredia & J. Altarriba (Eds.),
Foundations of bilingual memory (pp. 41.62–41.40). New York: Springer.
Peynircioğlu, Z. F., & Göksen-Erelçin, F. (1987). Part-set cuing across languages: Evidence
for both word- and concept-mediated inhibition depending on language dominance. Acta
Pureza, R., Soares, A. P., & Comesaña, P. (2016). Cognate status, syllable position and word length
on bilingual tip-of-the-tongue states induction and resolution. Bilingualism: Language and
Cognition, 19(3), 533–549. doi:10.1017/S1366728915000206
Poplack, S. (1980). Sometimes i‘ll st art a sentence in Spanish y termino en Español: Toward a
typology of code-switching. Linguistics, 18, 581–618.
Roediger, H. L. (2008). Relativity of remembering: Why the laws of memory vanished. Annual
Review of Psychology, 59(22), 225–254.
Roediger, H. L. (1990). Implicit memory: Retention without remembering. American Psychologist,
45, 1043–1056.
Roediger, H. L., Gallo, D. A., & Geraci, L. (2002). Processing approaches to cognition: The impe-
tus from the levels-of-processing framework. Memory, 10(5/6), 319–332.
Runnqvist, E., & Costa, A. (2012). Is retrieval induced forgetting behind the bilingual disadvantage
in word production? Bilingualism: Language and Cognition, 15(2), 365–377.
Schrauf, R. W., & Rubin, D. C. (2000). Internal languages of retrieval: The bilingual encoding of
memories for the personal past. Memory & Cognition, 28(4), 616–623.
Schwanberg, J., & S. (2010). Does language of retrieval affect the remembering of trauma? Journal
of Trauma & Dissociation, 11, 44–56. doi:10.1080/152997309031435505uJ2Tr9nD-
Slamecka, N. J., & Katsaiti, L. T. (1987). The generation effect as an artifact of selective displaced
rehearsal. Journal of Memory and Language, 26, 589–627.
Tulving, E. (1983). Elements of episodic memory. New York: Oxford Press.
Tulving, E. (1985). Memory and consciousness. Canadian Psychology, 26, 1–12.
Tulving, E., & Pearlston, Z. (1966). Availability versus accessibility of information for memory
words. Journal of Verbal Learning and Verbal Behavior, 5, 381–391.
Tulving, E., & Thomson, D. M. (1973). Encoding specificity and retrieval processes in episodic
memory. Psychological Review, 80(5), 352–373.
Vaid, J. (1988). Bilingual memory representation: A further test of dual coding theory. Canadian
Journal of Psychology, 42(1), 84–90.
Weldon, M. S., & Roediger, H. L. (1987). Altering retrieval demands reverses the picture superior-
ity effect. Memory & Cognition, 15(4), 269–280.
Weldon, M. S., Roediger, H. L., & Challis, B. H. (1989). The properties of retrieval cues constrain
the picture superiority effect. Memory & Cognition, 17(1), 95–105.
Chapter 7
Active–Passive Bilingualism
and the Functional Distance Between L1
and L2 as Viewed Within One Unifying
Theoretical Framework
Michael Sharwood Smith
Contents
Introduction 157
The MOGUL Framework 158
Two Implementations of the Framework 174
Crosslinguistic Influence in Bilingual Representation and Performance 178
Conclusion 183
References 184
Introduction
The Modular Online Growth and Use of Language (MOGUL) represents an ongo-
ing project that has been in operation since 2000. The first publication presenting
the framework was Truscott and Sharwood Smith (2004). Since that time it has been
discussed, illustrated and implemented in various ways in a range of publications
(e.g., Sharwood Smith & Truscott, 2005, 2006, 2010, 2014; Truscott & Sharwood
Smith, 2011; Truscott, 2015a, 2015b). In this particular chapter it will be used to
shed light on two aspects of bilingualism, that alongside such variables as language
proficiency and age of acquisition are relevant for any investigation into bilingual
ability, here using the term bilingual to include multilingual. The first of these is the
active–passive distinction: bilinguals may show understanding of a given language
but not necessary speak it: this may characterize their performance either all the
time or in particular situations. The active–passive distinction is relevant for the way
in which a person’s performance is used to characterize their bilingual ability. The
M. Sharwood Smith (*)
The University of Edinburgh, Edinburgh, UK
Heriot-Watt University, Edinburgh, UK
e-mail: msharwood@blueyonder.co.uk

158 M. Sharwood Smith
second topic to be discussed is the way in which different language systems

possessed by the bilingual work both interactively and independently in perfor-
mance and this particularly with regard to how closely or distantly they happen to
be related to one another.
The discussion will accordingly proceed in the following way: starting, in the
second sections, with as succinct account as possible of the MOGUL framework.
This will at least have to be sufficiently elaborated to shed light, in the space avail-
able, on the two phenomena chosen for this chapter but leaving out extraneous
detail, for example the embedding of the framework in current thinking in cognitive
science, a topic that is dealt with elsewhere including book-length publications
(Sharwood Smith & Truscott, 2014; Sharwood Smith, 2017; Truscott, 2015a). The
third section will then apply the framework with respect to the two topics chosen for
discussion.
The MOGUL Framework
Theoretical Frameworks and Theories
The MOGUL framework is just that, a framework. Although it is certainly theoreti-

cal, it is not a theory. This means that, when implemented, it requires further elabo-
ration. This must be selected from one or other of the theoretical approaches
developed within relevant areas of cognitive science as long as they are compatible
with the basic architecture of the framework. The most comprehensive account of
the framework to date is still Sharwood Smith and Truscott (2014).
It is important to note at the outset that the framework is an account of mental
(psychological) functioning. For example, it is not about the brain: mind and brain are
two separate levels of description. At the same time, the idea behind the framework is
that it should be as neurally plausible as possible so that these two levels can be use-
fully related. In this, it should ideally serve researchers on both sides of this particular
fence equally well and thus facilitate cooperation between the two. It is also not about
social phenomena although indirectly these play a part in the shape of the internalized
world that the individual creates by interacting with the world outside.
An Overview of the MOGUL Framework
The framework may be thought of as a skeleton account of the mind’s basic process-
ing and knowledge-formation mechanisms seen as a system of systems (see Fig. 7.1).
The central focus of interest has always been on clarifying the place and role of
language in the mind as a whole. And, with regard to language, the strengths of the
framework are that it permits a detailed analysis of not just the language-specific
systems that linguists focus on but importantly the crosstalk with all the other
7 Active–Passive Bilingualism and the Functional Distance Between L1 and L2… 159
Fig. 7.1 Language processing and storage: the inner network
cognitive systems involved in language processing (as well as with the motor
system). The claim here is that, without involving these other mental systems, it is
impossible to give a complete account of how, for example, different languages in
the bilingual or multilingual speaker are kept distinct in the mind both in terms of
their storage as well as how they are used during online performance.
The standard focus of theoretical linguistics and so the theories and frameworks
used with this field are concerned with the properties underlying linguistic struc-
ture. To the extent that linguists assume that linguistic structures are also mental
properties and not just abstractions, they will accordingly treat the different ways in
which linguistic properties combine as having to do with the nature of mental rep-
resentations, in this case linguistic ones. The traditional concern of psycholinguists,
by way of contrast, is not on linguistic structural properties themselves but on how
they are manifested in real time performance, i.e., in online processing. Although
this is no longer true, for quite some time psycholinguists have focused on the sim-
plest structural units such as the word and the syllable because of their amenability
to experimentation. However, this neat division of tasks between theoretical linguis-
tics and psycholinguistics breaks down when wider, interdisciplinary perspectives
are adopted.
Although regularly presented as a processing approach, MOGUL does cover
both processing and representation: in other words, processing and representation
are defined and handled within one unified framework. Within this perspective, like
the brain itself, the mind is composed of independent systems which can be recruited
to solve millions of different tasks at different times or in parallel and which each
handle a particular aspect of storage and processing that is handled by none of the
others. In other words, the mind has a modular architecture. It is composed of a
network of modular systems each of which has an identical basic design. Within this
network, none of these systems can be described as domain-general, which means
there is no central processor acting as a kind of command and control center.
Coordination and collaboration between the systems does naturally take place but it
is manifested in different ways depending upon the task in hand. The closest thing
that most of us would see as a system that purports to plan and supervise different
types of mental activity is our conscious selves. The conscious self, however, repre-
sents an extremely small part of the whole. Most of what goes on in the mind is
subconscious and completely inaccessible to the conscious self. Our assumed men-
tal commander-in-chief may have the illusion of being in total control but mani-
festly is not. Coordination and collaboration in the mind is mostly carried out
without his or her knowledge.
Modules: The Basic Design
The mind’s expert systems as displayed in Fig. 7.1, are each unique meaning that a
module will perform tasks that no other one can. Nevertheless they still have a basic
design in common. So, for example, the visual system that encodes and stores visual
experience, the conceptual system that handles abstract meanings and the phonological
system that handles speech sound (as opposed to generic sound) all consist of a proces-
sor and a store. This is the basic design. The processor handles, and can handle only
those elements that make up its own store. The store can be thought of as a dedicated
memory where the unique type of structures that make up that store are located. The
processor manipulates these structures according to its own unique principles. In this
way, visual structures—also called representations—are organized in a way that is dif-
ferent from structures in all other stores. Again, at the neural level, this is like the brain’s
responses to visual stimuli, which consist of patterns that are unique to vision and can
be identified as such. This unique type of organization is, in the MOGUL framework,
equivalent to the different codes in which structures in different modules are written.
Linguists will be most familiar with the way in which syntactic principles differ from
phonological principles: each may be thought of as a constituting a code creating struc-
tures that have unique and identifiable characteristics. More will be said about the way
modules function internally but first, we turn to the question of how modules do not
exist in splendid isolation but form a network.
Interfaces and Indices
Modules are of little or no use if they cannot collaborate with other modules. For
this, they need a way of connecting. The pathways between modules are their inter-
faces which form associations between given structures in one store with those in
another. Interfaces are different from the connections that bind elements within
modules. Those module-internal connections that associate structural items inside a

particular store are created according to the principles of that module. Take syntac-
tic principles, for example; syntactic features may combine to form a complex rep-
resentation such as a type of noun phrase that is regularly manifested in the linguistic
input to which an individual is exposed. They therefore reflect the grammar of a
particular language dictating for instance whether there are articles and, if so, where
articles and adjectives should be placed with respect to the head noun. In this way,
single elements are bound together to form more complex representations all writ-
ten in syntactic code. The code is defined according to the preferred syntactic the-
ory. The same goes for any other module. Interfaces, however, have a different
connecting function than those principles governing the internal combination of
elements within a module’s store. Interfaces associate and coactivate structures
across different stores.
Although virtually every module is involved in some way or another in the pro-
duction and comprehension of language, the interfaces that most obviously concern
us here are those that manage associations between the two modules that handle
linguistic structures (see discussion below), that is (a) the phonological module and
the (b) (morpho-)syntactic module plus systems closely involved in language pro-
duction and comprehension, namely (c) the conceptual system that handles mean-
ing and certain perceptual modules, in particular (d) the auditory module that
handles generic sound and e) the visual module since language is also processed in
the visual mode. So, for example, a given phonological structure in the phonological
store is paired via the connecting interface with a given auditory structure in the
auditory store. This means, for example, that when a specific auditory representa-
tional structure is activated in response to the incoming acoustic stimuli that happen
to be carrying speech patterns, this interface coactivates a specific phonological
structure in the phonological store. In this way, the incoming sound can be inter-
preted both as generic sound and also as speech. This is always assuming that the
hearer currently has the appropriate phonological knowledge in place to handle it;
otherwise, the phonological module may attempt, but will be unable to make sense
of its auditory input. This inner network for language with interfaces (the thick
arrows) connecting memory stores is represented in Fig. 7.1. Each store contains
only the structures belonging to its module.
The question arises as to what mechanism allows an interface to associate and
coactivate particular items in a different store rather than to operate at random.
Interface connections work according to an indexing system. An index is a common
identifying tag placed on two structures (representations) that have been matched
by an interface. Using the number 123 for the index, for instance, and the two-way
arrow (⇔) to indicate the interface, the pairing of an auditory structure (e.g., the
sound made when pronouncing a vowel) with a particular phonological structure
would be represented thus (see also Fig. 7.1):
AS123 ⇔ PS123

This system of indexing, matching and coactivating items across (not within)
systems is adopted from Jackendoff’s model (e.g., Jackendoff, 1987, 1997, pp. 89).
Note that indices used elsewhere in generative linguistics to match up or bind par-
ticular types of linguistic structure work differently. Indices here, in the MOGUL
framework, act as tags that mark an association between representations in adjacent
modules. In this particular example indices are created as a result of the early
encounters with a particular vowel sound. In some cases, exposure is not required:
an indexed association may be already in place as part of our biological starter pack-
age, the one that all members of the same species possess at birth to optimize their
survival chances. Behavior resulting from these pre-assigned associations is com-
monly referred to using the words instinct and instinctive. These terms are used for
example to describe a baby’s involuntary responses such as grasping and putting out
his/her arms when tipped forward suddenly (i.e., the parachute reflex) but also
instinctive preferences and dislikes making him/her avoid certain things without
having any prior experience, particularly important for the young of other species
who need to survive in the wild (see Sharwood Smith, 2017).
The Core Language System
What is known nowadays in generative circles as the language faculty narrowly

defined (LFN) is dominated by an autonomous syntactic system (Fitch, Hauser, &
Noam, 2005). By contrast, in Jackendoff’s modular architecture, LFN consists of
not one but two autonomous systems, covering the domains, respectively, of pho-
nology and syntax; a third system heavily involved in language processing, the con-
ceptual module, lies outside the LFN. In fact, both phonetic and conceptual-semantic
structure, although they owe their existence to interaction with phonology and syn-
tax, are in fact formed according to, respectively, auditory and conceptual princi-
ples, both involving modules that operate with more than just language- related
sounds and meanings.
One way of designating the combination of two specifically linguistic mod-
ules in the framework is the core language system. This means that what we
recognize consciously as a word, or lexical item, for example, is actually a com-
posite of different types of structures, some of which are specifically linguistic
(i.e., handled within the LFN, and others which are handled outside the core such
as auditory and conceptual structures). The composition of spoken word there-
fore breaks down into a multiple association of structures between, respectively,
the phonological, morphosyntactic and conceptual systems. A chain of associ-
ated structures in built d uring processing. In other words, a phonological struc-
ture (PS) is coindexed with a (morpho)syntactic structure (SS). This in turn is
coindexed with a conceptual structure (CS). The chain is longer than that in fact.
The original speech input is first encoded as a generic sound representation (i.e.,
as auditory structure [AS] before being fed into the phonological system and
matched with a PS. In the case of speech output, a further processing stage is
involved, namely the motor system and a c orresponding motor structure [MoS]).
Each of these representations, now associated by means a common index, are
written in a different mutually incomprehensible code and cannot be merged
with one another into one single structure. So, a word, our familiar (lexical) unit,
is actually represented as the following complex representation (the core lan-
guage structures shown in boldface): AS ⇔ PS ⇔ SS ⇔ CS. Put another way, the
word is from a processing point of view, minimally composed of a coactivated
chain of these four structures. Excluded from consideration here are motor,
affective and other perceptual associations with a given word. If they are taken
into account, our linear chain of structures becomes an even more complex net-
work. Viewed more abstractly, as in a classic theoretical linguistic account, it is
the product of four separate sets of formation rules (reflecting the principles of
each processor) with each representation linked by correspondence rules (in pro-
cessing terms implemented by interface operations) determining in each case
which representation should correspond to which. This is just another more
abstract way of viewing the same thing. Recall that the conceptual formation
rules (aka principles guiding the conceptual processor) are not specifically lin-
guistic but are part of the conceptual-semantic system that lies outside the LFN
and by virtue of that are part of the language faculty broadly defined of LFB. The
same goes for the auditory formation rules and the visual formation rules that
handle, respectively, the sounds and the visual signs of language and which also
handle the auditory and visual representations that are not associated with the
core language system as well. In this way, a Jackendoff-style architecture as
exemplified by the MOGUL framework can in principle be expressed either in
processing terms or in purely representational terms depending on one’s focus of
interest. Naturally in the framework adopted here the two perspectives are treated
together, as two side of the same coin.
One characteristic of the core language system that is particularly important for
multilingualism is the way it handles all its input without creating or recognizing
any explicit distinctions between particular languages, dialects, varieties, registers
and accents. In processing terms, it deals blindly with any input that comes its way
as long as it can be processed as language. This means that all phonological repre-
sentations wherever they come from are stored and processed in the phonological
module and the same goes for syntax. There is no language identification tagging
within the core language system. The consistent coactivation of structures that
belong to a given language, for example, is taken care of elsewhere. With regard to
the idea of a (mental) bilingual (mental) lexicon, this means that a French word and
a Japanese word, for example, are not distinguished with regard to two of their com-
ponent parts. In other words, their PS and SS are not marked as belonging to any
particular language system. This of course leaves open, for the moment, the ques-
tion of how consistent language selection takes place during performance but in any
case it must be triggered outside the core language system. This will be important
especially with respect to the second topic to be discussed below (i.e., to the role of
the distance existing between the particular language systems that are possessed by
a given bilingual or multilingual).
Language Outside the Core
Language in its totality is seen as a product of a collaboration between the two mod-
ules in the core language system and many other modules; these other modules
specialize in tasks that include those involved in language processing but which are
still not uniquely linguistic. In passive language use for example, either the auditory
or visual (or both) systems are fully engaged, while the motor system plays a sup-
porting role. In active language use, however, the motor systems are fully engaged
and collaborate with the auditory and visual systems.
Meaning and Value
Two modular systems are particularly important for the management of language
systems during performance, namely (1) the conceptual system which stores
abstract meanings and writes them in conceptual code and one which has not yet
been mentioned: (2) the affective system one function of which is to attribute values
(via its interfaces) to other structures in other modules.
Although commonly associated with consciously experienced and identifiable
emotions like fear, anger, happiness, sadness and disgust, the affective system is
much more than that since it is responsible for the way we assess or appraise, posi-
tively or negatively, all our current sensations and all the knowledge that we have
stored across the network of modular system the mind is composed of. Positive and
negative affective structures (AfS) are basic elements in this system. In this sense it
is useful to think of it as a value system rather than simply an emotional one. Take
the concept of a particular language for example. This will be an amalgam of vari-
ous conceptual features including those expressing how we conceive of a language
and features associated with a given language community and probably countries
where that is spoken. Let us call the conceptual structure representing the abstract
notion Spanish as follows: CSSPANISH. Let us also assume that the individual con-
cerned has a positive view of anything and anyone associated with this particular
concept. This could change at any moment, but let us assume it is positive for the
time being. This means it is associated with (in terms of the framework architecture:
co-indexed) an affective structure representing positive value. Leaving questions of
strength aside, this association is represented as structures from two modules co-
indexed via a common interface, the affective-conceptual interface—with a shared
index, here chosen randomly as 111 thus:
CSSPANISH111 ⇔ AfSPOSITIVE111

This chain of two coactivated structures could be expanded in different ways.
This might include the sound (i.e., AS of the Spanish greeting hola!), which since in
this case the listener, without yet understanding its meaning, is already able to rec-
ognize the sound as speech and more specifically as Spanish-sounding speech. This
AS will be matched with a reasonably well specified but not necessarily native-like
PS. There might also be some provisional default syntactic structure like N(oun) to
be amended later as the individual acquires the syntax of hola. There will as yet be
no conceptual structure representing the meaning, which has not yet been acquired,
but there will already be two associated conceptual structures within the conceptual
store representing, (a) the notion language, and (b) the concept Spanish. The most
obviously associated structures, all co-indexed with our index 111, would therefore
look like this, taking into account that the initial h in hola is not pronounced in
Spanish.:
ASola111 ⇔ PS/ ola /111 ⇔ SSN ( oun )111 ⇔ CSLANGUAGE111 ⇔ CSSPANISH111

⇔ AfSPOSITIVE111

This also shows that, as mentioned earlier, with an increasing number of modular
system collaborating, the notion of a chain quickly becomes a misleading way of
describing the associations. It is a network of co-indexed representations that can be
activated together, one possible example of which is shown in Fig. 7.2. In this case,
the positive affective structure being linked directly to CSSPANISH.
What this positive AfS does is effectively boost the network associated with the
meaning Spanish language, i.e., the conceptual representation on the right combin-
ing two separate conceptual structures Spanish and language. The promotes net-
works that contain Spanish phonological and syntactic structures. Simply put,
Spanish processing may now win out against any available competing alternative.
So far, the listener is however only recognizing the sound as a Spanish word and a
noun, without actually knowing what it actually means. Figure 7.2 shows that no CS
has yet been activated that contains its meaning. If, later, this listener was later able
to expand the AS/PS/SS chain with a correct meaning creating a new CS the net-
work would include the additional 111 association created via the PS/CS interface
thus:
SSN ( oun )111 ⇔ CSINFORMAL GREETING111

Here we have a syntactic structure which binds together various nominal proper-
ties (not shown here) to form a noun (N) and the interface (⇔) which has associated
this noun, by means of the 111 index, with a conceptual representation expressing
the meaning of hola. The simultaneous coactivation of all the other structures
including CSSPANISH and AfSPOSITIVE enables a positive association across the whole
network displayed in Fig. 7.2. The reader may imagine various permutations of this
chain, or rather network, depending upon the circumstances and state of the indi-
vidual’s current knowledge.
Note that coindexation does not involve a merging or combination of informa-
tion across modules: it simply enables coactivation. The contribution of each indi-
vidual module remains separate. For example, when our ears pick up a bit of
speech, a network of collaborating modular systems is set to work, typically with

a lot of parallel to-and-fro activity as our minds try to find the best-fit between all
the various collaborating systems. At the end of it all, via spreading activation,
what finally pops into our conscious mind is the meaning part, the contribution of
the conceptual system. In actual fact we are also aware separately of the accom-
panying speech sound, the contribution of the auditory system. This is because
perceptual activities, being important for our survival in the physical world, natu-
rally involve very high levels of activation. This can triggers awareness. A key
feature of conscious awareness is precisely intense activation igniting a burst of
synchronized activity across many different systems (see for example Truscott,
2015b; see also Crick & Koch, 2003). As a result, perceptual structures such as AS
gain access to our conscious awareness making us aware of the sound of speech.
In addition, the content of conceptual structures, when given similar high activa-
tion levels, also gains access to consciousness. So, given the availability of all
associated structures, we are able, virtually simultaneously, to be aware of the
speech sound and of its meaning which is actually activated milliseconds later.
Even though we consciously experience them as simultaneous, the two experi-
ences essentially remain separate. The sound is not the meaning and the meaning
is not the sound and if we don’t yet have the meaning we can still be aware of the
sound. More will be said about consciousness below but all the processing activity
between sound and meaning, carried out without requiring intense levels of acti-
vation, remains mercifully inaccessible to us. Life would be intolerable if we were
aware of even a small fragment of all ongoing mental activity. Even the boost that
the Spanish PS and SS get as a result of the positive affective structure displayed
in Fig. 7.2 is not nearly sufficient for us to become aware the phonological and
syntactic processing going on. Note finally that we are even unaware of the
detailed auditory and conceptual processing that underlies the sound and the
meaning of which we are aware.
In sum, apart from activating a chain or network of systems, none of which have
a structure that can be translated into the code of the next in line, interfaces between
systems are limited in what they can do. They can activate two structures with the
same index and they can, given the operation of a collaborating affective system,
transmit a boost to the level of activation of all associated structures. Interfaces can-
not merge all the structures into one single complex unit encoded in the same way.
Fig. 7.2 An activated network of structures

Affect is pretty well all pervasive, assigning and adjusting the values associated
with the many mental structures of different types in the various modules that make
up the complete system that constitutes what we call the mind. Affective boosts to
the activation levels of any associated structure will have a variety of effects on the
individual at the time. Motivation to learn, for example, would be explained using
networks of associations involving affective structures. Value associations including
strong positive and strong negative values will always have this boosting effect and
this in turn, according to the particular composition of the currently activated net-
work will result in either attraction or avoidance behavior. The raised activation
levels of all associated structures will be accordingly rendered more competitive,
which is the topic of the next section. The value and hence the competitive advan-
tage of given structures that are currently activated can change at any moment with
the changing (internal and external) circumstances. When new structures with their
own associated values come into play, this can affect those already activated and
cause shifts in behavior. In any case, this can certainly explain a whole range of
bilingual phenomena including, for example, the degree of influence one language
has on the other and code-switches back and forth between languages in mid-
conversation (Truscott & Sharwood Smith, 2016).
The Motor System
Another system outside the core language system but which is regularly in use dur-
ing language processing is the motor system. This is the mind’s software controlling
the physical motor system. The principal interest from the framework perspective is
the voluntary (as opposed to involuntary, autonomic) motor system. As language
develops in the individual be it a first, second or other language, initially the focus
will be on trying to comprehend the language. This is especially striking in first
language acquisition where children classically go through a receptive period before
first producing speech by which time they have clearly already acquired some of the
language system already. As the child gradually gains control of the appropriate
speech mechanisms, the relevant motor structures (MoS) associated with the organs
of speech get interfaced with the auditory and phonological structures of particular
speech sounds and speech production can begin. This is one account of the differ-
ence between passive and active use of language where active use is constrained by
what one might call a lack of motor programming or deficient programming, a topic
that will be further discussed below.
The Auditory and Visual Systems
In one way or other, the auditory and visual systems are both implicated in language
processing, that is, in both reading and written modes, and of course, during sign lan-
guage communication. Like the conceptual and motor systems, they are involved in
nonlinguistic activity as well and in fact their primary function is not language-related.
The auditory system deals with all kinds of sounds. So, the processing of acous-
tic stimuli always starts with the auditory system regardless of whether the stimuli
are speech-related or not. The phonological system is geared always to, as it were,
listen to the activity in the auditory system, that is, to be more specific, to whatever
is currently active in the working memory of the auditory system. This the phono-
logical system will automatically attempt to treat as potential input and will attempt
to match activated auditory structures with PSs. Our modules are geared to respond
to anything that they are exposed to but may not necessarily be able to handle every
input by coactivating structures from within their own stores, i.e., not all sound can
be turned into speech structures. The same goes for the visual system. Visual signs
that relate to language, written text and other language signs, are first processed by
the visual system as generic visual structures before any linguistic structures are
associated with them.1
Activation and Memory
Perhaps the most important feature of the framework in accounting for language
phenomena, including the two phenomena chosen for consideration in this chapter,
is activation. It is also a notion that is easily relatable to the neural level of descrip-
tion. Activation takes place in the stores. Two basic concepts are especially impor-
tant in this context: working memory and resting levels of activation.
In the psychological literature on memory, one common distinction that is made
is between long term memory (LTM) and working memory (WM) from which
items in LTM are momentarily accessed during online processing. Although
Baddeley’s (2012) WM approach seems to be the default option for many psycho-
linguistic studies, MOGUL adopts Cowan’s view (Cowan, 1993, 2005) that WM is
a state rather than separate component in memory architecture (Anderson, 1983;
D’Esposito & Postle, 2015). In the framework, then, it is those structures in a given
memory store that are currently in an activated state that are ipso facto in the WM
of that module. Hence during visual processing, for example, a limited number of
structures in the visual memory store at one time will be in WM, i.e., in a state of
activation. While in this state, the visual processor can arrange and combine them
following its own unique principles. Also, connecting interfaces will be able to
coactivate any coindexed structures in adjoining stores. In this way, the visual iden-
tification of a presented object with its name and its meaning can be accomplished
as associated structures across the relevant stores are activated in parallel.
To illustrate activation graphically without actually implying that WM is any-
thing more than just that (i.e., a state of activation), consider the following spatio-
temporal vertical metaphor. This metaphor portrays inactive items in memory as if
1
It is an open question whether the interface between the visual system is with the auditory system
and only then with the phonological system or whether there is a direct interface between vision
and phonology.
they were suspended at rest in higher or lower positions but at any rate below an
upper area. WM is portrayed as if it were this upper area, a place where the proces-
sor can combine and order items, and where the interfaces can create associations
between items in other stores or coactivate already ones that have already been
associated. In this way, activated, structures in a store rise from their current resting
level of activation to somewhere in this upper area for potential selection in a cur-
rent processing operation and later, once they cease to be active, fall back again, as
it were descending out of the upper WM area to a lower resting position. The WM
functions as a workbench for the manipulation of structures belonging to given
module as also their association with other structures in other modules.
The use of the vertical metaphor, imagining that activation involves structures
rising up in to working memory and deactivation involves falling back again pro-
vides an easy way of graphically illustrating the accessibility of structures during
processing and the competition between structures that happen to be candidates for
selection in a given processing operation. For example, when some perceived object
is visually identified or one word rather than another is selected for association with
that visual object, certain candidate structures that have been activated in the rele-
vant memory stores can be presented as rising up to a given point: some of them will
reach a position higher than others and arrive there quicker and the highest, those in
the winning position, will then have the best chance of participating in a current
chain of activated representations while the others are left to fall back, unused, to
their resting level of activation.
Growth
Acquisition by processing theory (APT) describes how growth takes place in the
system as a whole. Initially, novel input, a completely unfamiliar sound for exam-
ple, is presented to a given module which then attempts to respond by assigning
some structure to it: it could be novel auditory input resulting from a perceived frag-
ment of speech, processed initially in the form of acoustic input to the auditory
system. Then the resulting auditory structure (AS) that has been activated in
response to the acoustic input in turn becomes input for the phonological processor
to deal. In this way an external speech event in the environment that happens to
involve a novel linguistic structure triggers a chain reaction, the processing ending
in this example with the activation of a phonological structure (PS) and the first time
assignment of an index to all implicated structures. This at least would be what hap-
pens when the matching procedure is successful.
A particular PS will be selected straightforwardly because it outcompetes its
rivals, from an existing number of candidates that have been activated in phonologi-
cal memory. It is likely to be a well-established native PS that gets selected initially
simply because it is well established. Otherwise, if nothing can be made to fit the
input from easily available structures, the phonological processor will have create
on its working memory workbench a new phonological representation. Two pro-
cesses will happen then. (1) The selected structure will, as already mentioned, be
coindexed across the interface with the auditory input representation thus: AS123 ⇔
PS123. (2) Once deactivated, these newly indexed structures will sink down to a rest-
ing level of activation which, typically for new structures, will be relatively low.
However, APT states that every time this structure is activated in the future, it will
sink down to a resting level that is slighter higher than its original one giving it a
slight boost in accessibility. Hence, the more frequently it is activated, the more it
will gain in accessibility, and the greater its chances ultimately, of outperforming
any rival candidates in a given act of processing. The new structure will gain a small
boost if it is activated briefly but then outcompeted by a rival candidate. In this case
the rival is selected for participation in a current chain. Note that selected is a con-
venient but somewhat misleading term since there is no monitoring mechanism that
decides to use a structure or not: the end result just falls out by virtue of the fact that
a given structure happens to win the current competition.
A new structure, which normally starts out life at a very low resting level of
activation can still experience quite a strong activation boost due to interaction
with the affective system. For example, a newly acquired structure may for a short
time at least benefit from a novelty effect. The main thing to keep in mind is that,
as now generally recognized, only a small portion of what is activated during
processing shows up in actual performance. This goes for both monolingual per-
formance but also for bilingual performance as well. This means, for example,
that something native-like might have been acquired in the sense that the correct
associations have been made and the correct structures activated during online
processing but still get regularly outcompeted, at least for the time being.
Gradually, repeated activation allows a new structure to acquire higher and higher
resting levels and thereby greater accessibility to working memory. Its increased
competitiveness in combination with other facts such as the support of the affec-
tive system on a particular occasion will in time give it a better chance of outcom-
peting its more established rivals.
It might seem that the growth of new representations is driven in a simple and
straightforward manner by frequent input. This should not be taken to mean that
frequent input in the environment necessarily triggers growth, APT applies inter-
nally within individual modules. So, for example, syntactic growth cannot take
place just because an individual is frequently exposed to a given syntactic construc-
tion, even where an appropriate auditory representation (AS) of the speech contain-
ing that construction is activated. That AS has first to trigger a response from the
phonological module so that a matching phonological representation (PS) may be
found or constructed in order to build a AS ⇔ PS chain. Only if that takes place can
the syntactic module attempt to respond to the resulting phonological input and
extend the chain to AS ⇔ PS ⇔ SS. All the steps leading from the acoustic input up
to the gates of syntax have to be completed. In addition, a separate computing of the
meaning has to be made in order that the conceptual stage is completed because the
syntactic structure corresponding to the construction in question has be matched not
only with a PS but also with a CS, i.e., a conceptual representation, yielding AS ⇔
PS ⇔ SS ⇔ CS. This failure to respond directly and automatically to external fre-
quency will be familiar to developmental linguistic researchers from the early days
in both first and second language acquisition where highly frequent structures are
not necessarily learned earlier than less frequent structures (Gass & Mackey, 2002;
Hatch & Wagner-Gough, 1976).
Competition and Language Modes
The dominant view today seems to be that multilinguals activate both or all of their
languages any time they are engaged in linguistic activity whatever the situation
(Kroll & Tokowicz, 2005). This means that all the structures associated with the
different languages known to the users—and this includes dialects of the same lan-
guage—enter into a competition with each other. It is, however, not an equal playing
field. Various factors help to give the advantage to one set of structures over another.
Probably the most well-known expression of this idea is the notion of language
mode (Grosjean, 2010). When a bilingual is in a bilingual mode, both languages are
activated more or less equally, so switching from one to another is easy. In a mono-
lingual mode, one language dominates and interference from one language on
another is generally less likely and access to the nondominant language takes rela-
tively longer than accessing the currently dominant one (Olsen, 2016). In MOGUL
terms, structures associated with a currently dominant language will have high rest-
ing levels and speed up access to working memory They will therefore be likely to
win any competition with those structures associated with the less dominant lan-
guage. In this process the value-assigning affective system plays a major role as
discussed in the previous section.
As already mentioned, resting levels can change at any moment. This is because
there are certain situations when a resting level is boosted by the activation of
another structure that is associated with it and which happens to have a high resting
level. This will typically be a conceptual structure that defines something new in the
current situation, such as the arrival of someone who does not speak the language of
the ongoing conversation plus an associated value (affective structure). This raises
the current value of anything associated with that newcomer including the language
and or social status. This then will impact, via spreading activation, on the resting
level of all associated structures in different modules. The result will be changes in
behavior in the participants including, for example, switching to a different social
register or a different language (Truscott & Sharwood Smith, 2016).
Consciousness and Language
Some reference has already made to conscious awareness as a perceptual experi-

ence. As mentioned earlier, the majority of mental operations are carried out beneath
the level of awareness and most of those are inaccessible to consciousness. The
constant competition between structures within modules and the simultaneous
activation of alternative-solution chains of structures across modules is mercifully

kept hidden as far as our conscious mental life is concerned. Nevertheless, we do
have a conscious mental life which needs to be accounted for some way or other
even if the essential nature of consciousness itself remains a mysterious phenome-
non. In any case, there are some basic points to be made about how the framework
deals with conscious awareness. Firstly, its architecture is based on a commitment
to an activation theory of consciousness (Kinsbourne, 1988; Truscott, 2015a). More
specifically, the phenomenon we call conscious awareness takes place when struc-
tures are in a very highly activated state. Activation by itself, however, is not the
whole story. The second basic feature of the framework is the claim that conscious-
ness is, as just mentioned, a perceptual experience involving a simultaneous activa-
tion (and synchronization) of many modular systems. That is to say, the perceptual
group of modules covering vision, audition, somatosensory experiences and so
forth is where the contents of consciousness are projected. In sum, consciousness is
characterized by an intense, synchronized activation of structures across a range of
modular systems ending up what is in fact a perceptual experience.
For linguists of various persuasions, the conscious use of language and reflec-
tions on language, in general, are usually captured by the term metalinguistic knowl-
edge, literally knowledge about language as opposed to knowledge of a language,
the latter type not requiring any formalized knowledge of grammar rules, differ-
ences between languages and so on and so forth. Four-year-olds may already be
competent native speakers but they will as yet have little notion of what language is
all about. They may have some metalinguistic awareness enabling them to rhyme
and make puns but it will not be expressible in any formal way: for this kind of
knowledge they will have to wait until they go to school. The terms explicit knowl-
edge is also used with the precise meaning dependent upon particular theoretical
perspectives but very often it is the same or similar to metalinguistic knowledge.
Speakers of more than one language naturally have their conscious attention fre-
quently drawn to the difference between the languages they possess in whatever
measure and perhaps generally reflect more on language than monolinguals. Still,
everyone is prompted to think consciously about language when, for example, fail-
ing to produce an irritatingly elusive word or expression. Some basic knowledge of
what language is, that it consists of words and combinations of words that mean
something is the possession of most people. Multilinguals however may often find
themselves reflecting on the difficulty of particular languages and the relative dis-
tance between the grammars and vocabularies of the various languages that they
speak. They may also develop ideas about the value to them and their peer group of
words, structures and indeed particular accents, dialects and languages. For the
more curious and for academic students of language, this knowledge can be quite
elaborate. Either way, this consciously accessible knowledge is enshrined in the
conceptual system and not in the phonological and syntactic modules of the core
language system. For this reason, explicit or metalinguistic knowledge of language
can be categorized alongside all other types of knowledge of the world as being
encoded in the conceptual system. Reflections on grammatical structure, for exam-
ple, involves the deployment of conceptual grammatical knowledge rather than the
phonological and morphosyntactic knowledge that drives unreflecting production

and comprehension. Core grammatical knowledge remains permanently inaccessi-
ble and operates efficiently and unhindered at levels of activation that are insuffi-
ciently intense to ignite slow, conscious experience (see Dehaene 2014,
pp. 142–150).
The Framework Summarized
To sum up the main features of the MOGUL framework, the mind, like the brain,
can be divided into a network of systems, each uniquely equipped to cope with a
particular type of task with not one of them dominating all the time. A modular
system consists of a processor, a store, and interface between the store and the store
of one or more other modules enabling myriad types of collaboration to taken place
via spreading activation creating temporary chains or more often whole networks of
associated structures across the working memories involved. Each processor han-
dles items that are currently in WM state within its store. Connecting interfaces
activate associated (coindexed) items in other stores. In this way, millisecond by
millisecond, chains of structures are temporarily created in parallel, rapidly becom-
ing networks and competition continues within and across stores until a best-fit
overall solution is found for the task in hand.
A store contains structures (representations) and can be characterized in the fol-
lowing ways:
(a) Structures are built according to the principles that are unique to that module
(b) Structures include primitive elements (structural properties or features) that are
part of our biological inheritance for that module.
(c) Responding to input, primitive structures may be combined to build new com-
plex structures.
(d) All structure combinations within a store obey the principles of that particular
processor.
(e) We begin with a basic starter set that has been evolved to optimize survival
from the moment of birth onwards, and most probably even in the womb..
(f) This basic set includes certain pre-assigned interface associations across stores
belonging to other modules.
(g) The rest of the contents of any store is created in response to the life experience
of each individual.
Knowledge is manifested in the particular complex structures that have been cre-
ated in various stores as a result of life experience and in the networks linking them
across the interfaces that connect stores with one another system. Associated struc-
tures are coindexed allowing interfaces subsequently to coactivate them. Knowledge
grows via the formation of new structural combinations within modules and new
associations across different modules and also through changes in the resting levels
of the individual structures involved making that knowledge relative accessible and
hence relatively usable. By the same token, lack of activation will cause resting
levels to decay making this knowledge less accessible and less usable until it is, to
all intents and purposes, forgotten. Any knowledge that features in conscious aware-
ness (i.e., in thinking and reflection) is due to the highly active, and interactive
perceptual group of modules. We can become consciously aware of the sounds of
language, the visual images projected by text, signing and the feel of Braille text but
the perceptual systems also provide the means for the contents of the conceptual
store to be projected as perceptual experience. To make this happen, the implicated
conceptual structures involved in thinking have to be intensively activated. This
means, amongst other things, resource limitations: the resource-hungry mental
operations required for thinking cannot be carried out in parallel but must be sequen-
tial with one thought following another. Concepts essentially become percepts. This
includes the thoughts we might have about language, the product of our metalin-
guistic (conceptual) knowledge. Contrast this with linguistic processes emanating
from the core language system which can to operate in parallel, below the level of
awareness with much less processing load, and in fact always do so.
With the framework being sufficiently elaborated for present purposes, the dis-
cussion can now proceed by applying the framework to account to the two areas of
special focus, namely active and passive bilingualism and the interaction of differ-
ent languages in the individual bilingual mind. The advantage of having a detailed
account of how the different systems of the mind both develop and collaborate
makes possible a basic, succinct but still quite precise explanation of the processing
mechanisms underlying these two phenomena.
Two Implementations of the Framework
Overview
The architecture of the MOGUL framework will be applied to (1) the distinction
between active and passive bilingualism and (2) the distance between a bilingual’s
languages and its influence on performance. It allows us to embed accounts of these
two multilingual phenomena within a much wider frame of reference, that is, not
just with regard to language but to cognitive functioning in general. The sociolin-
guistic literature details the various ways multilinguals function in a social context.
This section will be strictly confined to the inner, mental context, that is, the lan-
guage user’s mind.
Active and Passive Bilingualism
With regard to the active- passive distinction as applied to language use, most peo-
ple would concede the term passive is a misnomer since the inner psychological
state of passive bilinguals who are comprehending what is going on in a language is
actually marked by a great deal of mental activity. In other words, passive must be
taken to refer to observable, external activity and so indicate the receptive-only use
of a language or receptive bilingualism (Döpke, 1992; Bialystok, 2001, p. 4).
There are many different situations that can lead people to understand one of
their languages but not to speak it. The question that concerns us here is not so much
to describe the different social situations and age groups in which this phenomenon
can occur but to understand what is happening in the mind of the bilingual that leads
to this decision, be it initially a completely subconscious decision or a very deliber-
ate one. There is the developmental perspective, for instance. Young children acquir-
ing language from a very early age will learn receptively by listening, so their
production will give distinctly conservative impression of their current ability. The
second language acquisition literature makes much of the silent period with learn-
ers, if they are allowed to, will start only by listening to the language and not speak-
ing. The result is that when they eventually do speak, some of the groundwork in
establishing the language system has already taken place. This may also apply to
young bilingual learners (Tabors, 1997). Older learners will also adopt this initial
strategy although, in formal instruction situations and depending on the method
used, they may be forced to be active much earlier than they would choose: they
then have to produce language at a very early stage.
Whatever the underlying causes of passive bilingualism are, it is appropriate to
begin with the particular value that the individual places on speaking the language
concerned. The assignment of value is what the affective system does, It influences
pretty much every choice we make consciously or unconsciously and that includes
the choice not to speak a language that we have, to a greater or lesser extent, already
acquired.
The Vital Role of Affect
As suggested before, affect will play a crucial role in the decision whether or not to
desist from active language use. Even where passive bilingualism is due mainly to
deficient linguistic resources, the decision not to speak and only to comprehend is,
like most other things we do, driven by evaluation or appraisal as it is usually
known (see for example, Frijda & Mesquita, 1998). Viewed from the point of view
of MOGUL, this appraisal, which is the domain of affective processing, happens
most directly in conjunction with the conceptual system. Associations are estab-
lished, via the mutual interface between the two systems, between, on the one hand,
CS in the system where abstract meanings are built, stored and processed and, on
the other hand, given AfS values which determine the negative or positive value of
those meanings thus:
MEANING ( coded as CS) ⇔ AFFECT ( coded as AfS)

The idea in the individuals’ mind of attempting to produce utterances in a given
language rather than just comprehend them, or alternatively just to comprehend
them and not to attempt production, is created and encoded in the conceptual sys-
tem as CS. In the case of passive bilingualism, the appraisal leads to the association
of negative value, encoded as an AfS, with the complex CS that binds together
individual CS that together represent the act of producing utterances and the iden-
tity of the language concerned. This renders active bilingual performance unat-
tractive. Additionally, positive affective structures may become associated with a
comprehension-only approach to exchanges in the relevant language making pas-
sive bilingualism correspondingly attractive. So, whenever the use of the language
is called for, coindexation between given CS and given AfS will ensure that cur-
rently positively valued passive bilingual behavior is the most likely outcome. That
in broad terms is the way the framework explains the basic processing responses of
the receptive bilingual. Some classic examples of passive bilingualism are dis-
cussed next.
Language Attitude
One kind of passive bilingualism is based on a negative attitude towards the given
language itself. This can happen in various circumstances, for instance with children
in a formal classroom environment where foreign languages are not valued positively
by them or in the home environment where the home language is similarly valued
negatively. In both cases exposure to the negatively valued language is imposed on
the child. Without being able to avoid its use entirely, the next best strategy for the
child is whenever possible to comprehend the language without speaking it. In a heri-
tage language situation, a child is raised bilingually by its parent one of which will
speak their current version of their native language, which is not the language of the
community at large (Montrul, 2016). The community language, which is often also
the language of the child’s peer group, is likely to be very positively valued by the
child but the home language may be negatively valued even though it is the language
of its parents. In terms of MOGUL, the core association involved will be the negative
value that is associated (coindexed) with the conceptual representation that represent
the identity of the language concerned, say for the sake of argument Quechua in a
wider Spanish-speaking community. This association across the affective-conceptual
interface, is formalized, using MOGUL notation, as follows:
AfS567! negative! ⇔ CS567 QUECHUA .

This should be a negative value attached to all things related to Quechua. An
alternative possibility would involve the negative association being only on the
Quechua language alongside a positive association with Quechua speakers making
just the language unattractive but not the speakers of that language.
There is no reason in principle why the child should not be perfectly happy
speaking both languages, but by becoming aware of the fact that other, positively
valued individuals such as the child’s non-Quechua-speaking peer group have a
negative view of his or her home language, the child may nevertheless come to
adopt the same negative affective association as an expression of solidarity with

that peer group. This association may be triggered when interacting with both
peers and parents, or just with peers in which case the two situations of use as rep-
resented in the child’s conceptual system is valued differently: the child who val-
ues the home language positively and the parents who speak it, will, other things
being equal, not adopt a passive strategy. The child will express solidarity with
both language groups in the appropriate circumstances depending, for example, on
the location, and the identity of current participants in the exchange. The concep-
tual structure to which the positive or negative AS is associated is invariably com-
plex. The status of the language in question is determined by various other
considerations each of which will be encoded somehow in the individual’s concep-
tual system. Consequently, the appraisal that leads to the assignment of a given
value will involve much more than the conceptual structure that identifies the lan-
guage in isolation. Sociolinguists study the external circumstances which are
understood to affect bilingual behavior but the focus of psycholinguist should be
on the resulting internalized context, the inner world as encoded within and across
the various modular systems that make up the individual mind and where the affec-
tive and conceptual systems play a major role.
It is worth noting here that, once you have acquired a language to a greater or
lesser degree, you cannot decide not to understand something that you hear. This is
simply because linguistic processing is triggered automatically: input processing is
mandatory (in the sense of Fodor, 1983). The only way of stopping it is to cut off the
source of the signal by leaving the scene, covering your ears or shouting to drown
out the sounds of speech. This means that the only decision you can effectively
make in most situations is whether to participate actively or not; one way or another,
the affective system will be involved. Looking at it from another point of view, stay-
ing receptive means that to do so will bring benefits even if that is only the avoid-
ance of something negative and not something more like the assertion of one’s own
individual identity vis-a-vis that of one’s parents. Positive and negative affect can
combine to shift behavior one way or another.
Developmental Factors
Receptive bilingualism may also be driven by the bilingual’s appraisal of their own
current ability in the language, how far and how fast they have developed and what
this current ability does and does not provide them with. Again the affective-
conceptual interface is involved, but this time the object of the appraisal is different.
The conceptual structures that receive negative or positive associations are not asso-
ciated with the social status of the language itself. Rather what is valued negatively
is the expressive power, or lack of it that the language currently makes available to
that individual. This may include the individual’s appraisal of the psychological
impact that performance in that language will have on other people—for example,
will they struggle to understand you?—and, as a consequence of that appraisal, also
and appraisal of the psychological impact on the individual: how will that struggle
to understand what you are saying make you feel?
To take the expressive ability example, this is where a bilingual appraises their
communicative ability in a given language negatively, judging it to be inadequate
for effective and reasonably effortless communication to take place. An avoidance
strategy is the obvious solution so negative value is placed on anything that has to
do with production, i.e., speaking or writing. However, the precise target of negative
appraisal, i.e., what precisely the negative affective value is associated with, may
again involve one or more of a number of more specific conceptual structures, that
is to say some specific aspect of language ability such as the perceived available
lexical repertoire or the range of easily accessible and useful grammatical construc-
tions: this more focused self-evaluation naturally depends on the extent of an indi-
vidual’s metalinguistic knowledge of the language system. Very often the
self-perceived ability may be limited to knowing what words are and their useful-
ness in communication: I don’t know enough words in this language, my vocabulary
is very limited, and so on.
To take another aspect of expressive ability bilinguals may evaluate as inadequate
for efficient communication their mastery of pronunciation. This will be based on the
self-produced acoustic input to the auditory system which represents the sound of
their own speech; this sound can be appraised directly via the auditory-affective
interface although the conceptual system will be brought into play at some point to
include an appraisal of the wider impact of the individual’s perceived poor speech.
Apart from the concept of pronunciation itself, this particular self-evaluation must
also involve the motor system that drives the movements of the organs of speech and
the somatosensory system where the sensations associated with speaking are encoded.
This will allow the individual to appraise any difficulty encountered in producing the
speech sounds of a given language: I can’t get my tongue round these sounds.
This is a simplified view of the various types of association that could potentially
be involved here but it provides a sense of how the framework can make explicit
what is going on psychologically when someone decides to opt out of the active use
of a language or, indeed, once any formerly negative evaluations turn into positive
ones, decides to opt in. Apart from anything else, one can say about the way the
modular systems of the mind collaborate (via their respective interfaces) to produce
specific types of behavior, this discussion illustrates the importance of the much
under-estimated role of the affective system.
rosslinguistic Influence in Bilingual Representation

C
and Performance
Multilingual Storage across Modules
As should be clear by now, language, in the current perspective, is stored and pro-
cessed across the whole system that we refer to as the mind. The specifically pho-
nological and syntactic representations that drive language performance are stored
in the core language system, the meaning representations associated with language
are stored in the conceptual system, the perceptual representations that are involved
in language are stored in one or other of the perceptual modules, typically the audi-
tory and visual modules and the movements associated with comprehension and
especially the production of language are stored in the motor system. In other words,
many of the collaborating systems that make up the mind are involved.
When the question arises about how different language systems are represented
so that they can be differentiated one from another in performance, we have to have
a slightly different answer, especially when answering questions about the struc-
tural relatedness between languages and how a greater or less distance between
them might affect performance. Clearly, the individual mind can accommodate any-
thing from one to, in some cases, a large number of different languages, let alone
different versions of the same language: even monolinguals manage different variet-
ies and registers of their single language. This obviously means that the mind needs
some way of differentiating between them, otherwise chaos would ensue: every-
thing would get mixed up with everything else. At the same time, we need to explain
why bilinguals can indeed keep languages apart in their receptive and productive
performances on some occasions but demonstrate crosslinguistic effects—often
experienced as interference—on other occasions and, on yet other occasions, delib-
erately mix languages and switch back and forth between them.
Any theoretical framework that is worth its salt will need to account for the
mechanisms that allow the variable use of the languages possessed by an individual
and indeed the factors that influence them. The answer cannot, in any straightfor-
ward sense, come from the core language system: whatever happens to the outcome,
the syntactic and phonological modules, at least in the perspective adopted here,
operate blindly, using any input that can potentially be assigned core linguistic rep-
resentations and straightforwardly applying the principles enshrined in their proces-
sors. In other words, there are no internal language identification tags within this
core language system showing that a specific structure (PS or SS) belongs to a
particular language. It might be said that the core language system does not have to
worry about such niceties because that is simply not part of its job: language dif-
ferentiation is handled elsewhere. In point of fact, it is actually based on particular
structures in the conceptual system. The relevant CS are formed as a consequence
of the individual’s past experience in identifying different languages. These identi-
fying representations may be directly associated with other types of structure (also
outside the core language system), namely:
1. Given meaning representations, also coded as CS
2. Given sound representations coded as AS
3. Given visual representations (representing the look of text or the look of signs
and gestures), codes as VS (visual structure).
The identities of specific languages are thus, like the meanings of words them-
selves, represented as structures written in conceptual code. A conceptual struc-
ture representing the concept, Quechua, for example, will be linked to any sound
of Quechua as represented in the auditory system as auditory structures (AS),
thus:
AS567(a sound produced by Quechua speech ) ⇔ CS567 QUECHUA .

The visual appearance of written Quechua will be similarly processed via the
interface between the same Quechua CS shown above except this time with the
visual structure associated with Quechua text. Focusing on speech now, when
Quechua is acquired, the sounds of Quechua that are processed are coindexed with
specific PS and SS as the core language system tries to trigger their own representa-
tions in their attempts to make sense of the auditory input. At the same time, the AS
will be coindexed with that conceptual structure that serves to identify the sound as
Quechua. Note that this can but does not have to be a conscious process. The point
is that language identification takes place in the conceptual store. In this way, a com-
mon index is shared across a range of structures outside the core language system
and with specific PS and SS within the core language system ensuring the activation
of a particular set of associations when the need arises. For example, one might
learn to recognize a particular word, say yaku, as being a Quechua word but still not
know the meaning (water). Then the relevant network of associations would look
something like Fig. 7.3. Here the only meaning (CS) associations that have been
established are those that identify the sound as a word in the Quechua language. The
association with concept water has not yet been made. Its phonological and syntac-
tic structure, however, have been established. The individual concerned knows that
s/he has heard a Quechua word and, whether or not they are consciously aware of it,
that it is a noun.
If and when the meaning of the yaku is acquired, it would look something like
Fig. 7.4. where the displayed conceptual associations now include the meaning of
water: not only is the listener aware of the sound as being a word in Quechua but
now knows the word is Quechua for water:
A bilingual knowing the equivalent word in both Spanish (agua) and Quechua
(yaku) would have the network that is displayed in Fig. 7.5, ignoring here any pos-
sible differences in the detailed phonological, syntactic (nominal) and conceptual
structure. Note, in passing. That the use, for the sake of convenience, of phonetic
script between square brackets to describe the AS in Figs. 7.3–7.5 does not imply
that all AS can be specified this way: the sound of a splash would naturally not be
very amenable to phonetic notation.
Crosslinguistic Effects and Psychotypology
Languages can differ structurally to a greater or lesser extent and describing the
nature and range of possibilities is what language typologists do. The mind of a
bilingual possesses its own typology or psychotypology which does not necessarily
agree with that of the academic linguist and is based on to the way the individual
currently perceives the distance of proximity relationships between the language
systems they are concerned with (Kellerman, 1983). Psychotypology is created in
the conceptual system as the individual develops beliefs about the relatedness of the
Fig. 7.3 Recognizing the

sound and language
identity of yaku
Fig. 7.4 Recognizing the

sound, language identity,
and meaning of yaku
Fig. 7.5 A Spanish–

Quechua bilingual’s
network of associations
based on the meaning of
water
different languages that that individual has acquired or is acquiring (see for exam-
ple, Rast, 2010). This includes the notion that certain kinds of words and expres-
sions may or may not be found in other languages. In other words, they are
nontransferable. Acceptability judgment tests such as those carried out by Jordens
(1977) and Kellerman (1977), showed that, for example, internationalisms, typi-
cally with a Latin or Greek origin like automatic and, indeed, international itself
were treated in acceptability judgement tests as freely transferable. More idiomatic
expressions like to drink someone under the table were, however, treated often erro-
neously as unique to the individual’s own language. Furthermore, results showed
that languages that learners initially treated as very close, and therefore a good
source for attempting new words and constructions at a later stage proved to be not
so close as learners first perceived them to be.
This changing psychotypological knowledge or set of beliefs about languages
is part of the individual’s metalinguistic awareness in general. To the extent the
individual can find a vocabulary to articulate it and reflect upon it, that part of his
or her psychotypological CS will also form part of an explicit (consciously acces-
sible) typological knowledge. Note that knowledge here is a strictly relative term
and is often replaceable by beliefs. In any case, consciously or unconsciously,
an individual’s current psychotypological assumptions will certainly influence

conceptual aspects of their language performance. This therefore includes all

semantic and pragmatic aspects of language, which are coded in CS, this being pre-
cisely where most research into the learner’s beliefs have concentrated. The use of
German drehen and Dutch draaien can both be used to express the meaning ‘to roll
a cigarette’ but only in Dutch can draaien be used to express the meaning ‘to dial a
telephone number’ (Jordens’ example).
As discussed above, tests reported by Jordens (1977) showed how the beliefs of
Dutch students about the transferability of Dutch words, collocations and expres-
sions change as they progressed through 3 years of university education. At first,
they operate on the apparent assumption that Dutch and German are close enough
to sanction the free use of Dutch-like expressions in German. At a later stage, on
further exposure to German, they find out that this assumption can lead them into
error so they respond by switching to an avoidance strategy. Avoidance also results
in errors on those occasions where Dutch and German turn out in fact to be the
same. Only later do they finally acquiring German sufficiently well to be able to
distinguish in a nativelike manner exactly what does and does not work like Dutch
(Jordens, 1977, 1978, Jordens & Kellerman, 1981).
In the MOGUL perspective, however, the situation with syntax, and indeed with
phonology as well is more complicated. As discussed above, neither phonological
or syntactic processing will determine the identification of the language being spo-
ken, or, in this case, the written language. Psychotypology can have no impact on
the core language system. If presented with an ungrammatical sentence in German
that looks like a direct literal translation from Dutch (e.g., with the exact Dutch
word order), the core language system processing of this sentence will proceed
language-neutrally simply assigning PS and SS as described earlier irrespective of
any beliefs the individual may have. This means that the identification that a piece
of text is, say, in German will be made via the visual-conceptual interface as previ-
ously explained. In an acceptability judgement text, an incorrect acceptance of the
sentence by a Dutch learner based purely on the basis of explicitly but still errone-
ously held views on language distance is made on conceptual grounds. This high-
lights the dangers of such judgment tasks where explicit knowledge of grammar is
allowed to intrude on the decisions taken by participants.
In many cases, if there is indeed a degree of conscious metagrammatical influence
operating, an internal conflict may arise between a) the blind processing of SS, and b)
the beliefs, couched in CS, that, for instance, the two languages are syntactically almost
identical. At first incorrect beliefs will influence performance on a regular basis even
though the nativelike associations may already have been formed. However, the native-
like associations may still not be strongly enough established to make any impact. This
will continue until, on sufficient exposure to any German that does not follow the syn-
tax of Dutch, the appropriate structures (SS) are in place and established enough to be
able to compete strongly for selection. In the period that leads up to this stage, all
Dutch-based structures will have been strong enough to outperform the competition
from newly acquired representations with their lower resting levels. The still more
strongly established Dutch-like PS and SS will, in situations where conceptually-based
knowledge can influence performance, receive support from erroneous beliefs (couched
in conceptual structure) giving preference to Dutch-like constructions In German. This

will ensure a period of Dutch dominance before sufficient exposure to German finally
triggers a switch to more nativelike versions of the language, possibly accompanied by
a corresponding change in the learner’s psychotypological beliefs.
We can turn now to languages which are both typologically distant in an objective
sense and also psychotypologically distant as well because a learner/bilingual also
views it this way. One example would be the syntax of Dutch and Japanese. In sequen-
tial second language acquisition where an L2 is acquired after an L1, we can assume
that, in terms of explicit grammatical knowledge, any initial Japanese = Dutch period
will be extremely short if not nonexistent. It should be nonexistent if the two lan-
guages are perceived from the start as totally different. On exposure to either lan-
guage, the core language system would simply carry on blindly doing its job. The
structural dissimilarity between the two languages will initially cause problems irre-
spective of any typological beliefs. The learner would face an initial period when, for
example, on exposure to Japanese, the activation strength of Dutch-associated PS and
SS in a given sentence would promote a misanalysis of speech sounds and also word
order. For example, it will render it difficult to establish any further meaning associa-
tions, i.e., between conceptual structures and individual syntactic elements. It will
be a challenge to identify a meaning for each object and subject nouns and verbs
appearing in an unfamiliar order, for example. However, this structural distance effect
may work differently where different aspect of the language system are involved. In
two languages generally perceived to be highly unrelated, perceived similarity can
still occur at other levels. The lexical level may provide an example. Because the
Japanese loan word feminisuto means a man who pampers women, this might lead a
Japanese speaker to misinterpret the English word feminist.
Language distance may or may not play a major part in determining bilingual
performance depending on the aspects or levels of the language concerned, as just
illustrated: it may also vary according to what language mode the speaker is operat-
ing in. For example, when operating in bilingual mode, (i.e., where both systems are
highly activated at the same time), Japanese–English bilinguals that have a good
command of both languages conversing with one another will not always find it
easy to keep these two typologically distinct languages apart while they, or rather
their minds can exploit the combined resources of the languages they share. They
may sometimes freely mix the two together, even without specifically intending to,
or experience access delays in finding words and phrases.
Conclusion
To briefly sum up, what this chapter has hopefully demonstrated is that a working
model of the mind is crucial to understanding a disparate set of complex processing
phenomena in bilinguals, such as the two phenomena in focus, passive bilingualism
and language distance effects. This wide angle perspective can shed light on pro-
cesses that, looked at from a narrow point of view, can only be explained within,
say, linguistic terms, or structural linguist terms or online processing terms. With
the power provided by a broad-ranging framework such as the one discussed in this
chapter, phenomena that have been analyzed using only one of these narrower per-
spectives may turn out to be the natural consequence of principles that extend not
only to all of language processing but to cognitive processing in general. Without a
grasp, at least in terms of a currently plausible and well specified working model of
the mind, of how language interacts with other mental system, researchers are
thrown back on frameworks than only cover their own particular area of expertise.
Apart from that limitation, they may also be subtly influenced by models of the
mind they do happen to have but often vaguely specified ones and largely implicit
and hence unreliable. That has been the default situation for most people so far.
Looking at data with a more explicit working model of the mind as a whole should
become second nature to a researcher. The discussion has hopefully provided some
examples of how bilingual performance can profit from this broader view. Finally,
another advantage of using frameworks with an interdisciplinary reach is that these
phenomena can be effectively studied in detail by researchers with different research
interests and, to conclude on a more specific note, the framework illustrated in this
chapter has underlined the importance of including in explanations of processing
the crucial role of the affective system. This calls for a rethink. Affect should be
seen not as a side issue and not just about the role of human emotions. It should be
seen as an integral part of all cognitive processing.
Acknowledgement My sincere thanks are due to the editors, reviewers and to John Truscott for
their comments on earlier drafts.
References
Anderson, J. (1983). The architecture of cognition. Cambridge, MA: Harvard University Press.
Baddeley, A. (2012). Working memory: Theories, models, and controversies. Annual Review of
Bialystok, E. (2001). Bilingualism in development: language. Literacy, and cognition. Cambridge:
Cowan, N. (1993). Activation, attention, and short-term memory. Memory and Cognition, 21,
162–167.
Cowan, N. (2005). Working memory capacity. New York: Psychology Press.
Crick, F., & Koch, C. (2003). A framework for consciousness. Nature Neuroscience, 6, 119–126.
Döpke, S. (1992). One parent one language. An interactional approach. Amsterdam: John
Benjamins.
D’Esposito, M., & Postle, B. (2015). The cognitive neuroscience of working memory. Annual
Review of Psychology, 66, 115–142.
Dehaene, S. (2014). Consciousness and the brain: Deciphering how the brain codes our thoughts.
New York: Viking Press.
Fitch, W., Hauser, M., & Noam, C. (2005). The evolution of the language faculty: Clarifications
and implications. Cognition, 97, 179–210.
Fodor, J. (1983). The modularity of mind. Cambridge, MA: MIT Press.
Frijda, N. H., & Mesquita, B. (1998). The analysis of emotions: Dimensions of variation. In
M. F. Mascolo & S. Griffin (Eds.), What develops in emotional development? (pp. 273–295).
New York: Plenum Press.
Gass, S. N., & Mackey, A. (2002). Frequency effects in second language acquisition: A complex
picture? Studies in Second Language Acquisition, 24, 249–260.
Hatch, E., & Wagner-Gough, J. (1976). Explaining sequence and variation in second language
acquisition. Language Learning, 4, 39–47.
Kinsbourne, M. (1988). Integrated cortical field model of consciousness. In A. J. Marcel &
E. Bisiach (Eds.), The concept of consciousness in contemporary science (pp. 239–256).
London: Oxford University Press.
Jackendoff, R. (1987). Consciousness and the computational mind. Cambridge, MA: MIT Press.
Jackendoff, R. (1997). The architecture of the language faculty. Cambridge, MA: MIT Press.
Jordens, P. (1977). Rules, grammatical intuitions and strategies in foreign language learning. lnter-
language Studies Bulletin, 2(2), 5–76.
Jordens, P. (1978). Rules, grammatical intuitions and strategies in foreign language learning. lnter-
language Studies Bulletin, 2, 5–76.
Jordens, P., & Kellerman, E. (1981). Investigations into the “transfer strategy” in second language
learning. In J. G. Savard & L. Laforge (Eds.), Proceedings of the 5th AILA Congress, Montreal
(pp. 195–215). Quebec: Les Presses de I’Université Laval.
Kellerman, E. (1977). Towards a characterization of the strategy of transfer in second language
learning. lnterlanguage Studies Bulletin Utrecht, 2, 1, 58–145.
Kellerman, E. (1983). Now you see it, now you don’t. In S. Gass & L. Selinker (Eds.), Language
transfer in language learning (pp. 112–134). Rowley: Newbury House.
Kroll, J., & Tokowicz, N. (2005). Models of bilingual representation and processing. Looking back
and to the future. In J. Kroll & A. De Groot (Eds.), Handbook of bilingualism (pp. 530–554).
Oxford: Oxford University Press.
Montrul, S., M (2016). The acquisition of heritage languages. Cambridge: Cambridge University
Press, 2016.
Olsen, D. (2016). The gradient effect of context on language switching and lexical access in bilin-
gual production. Applied Psycho Linguistics, 37(3), 725–756.
Rast, R. (2010). The use of prior linguistic knowledge in the early stages of L3 acquisition.
International Review of Applied Linguistics in Language Teaching, 48(2/3), 159–183.
Sharwood Smith, M., & Truscott, J. (2005). Stages or continua in second language acquisition: A
Mogul solution. Applied Linguistics, 22, 219–240.
Sharwood Smith, M., & Truscott, J. (2006). Full transfer full access: A processing-oriented inter-
pretation. In S. Unsworth, T. Parodi, A. Sorace, & M. Young-Scholten (Eds.), Paths of develop-
ment in L1 and L2 acquisition. Amsterdam: John Benjamins.
Sharwood Smith, M. (2017). Introduction to language and cognition: Mapping the mind.
Sharwood Smith, M., & Truscott, J. (2010). Consciousness and language: A processing perspec-
tive. In E. Perry, D. Collerton, F. LeBeau, & H. Ashton (Eds.), New horizons in the neurosci-
ence of consciousness (pp. 129–138). John Benjamins: Amsterdam.
Sharwood Smith, M., & Truscott, J. (2014). Explaining input enhancement. International Review
of Applied Linguistics, 52(3), 253–281.
Tabors, P. (1997). One child, two languages. Baltimore, MD: Paul H. Brookes.
Truscott, J. (2015a). Consciousness and second language learning. Bristol: Multilingual Matters.
Truscott, J. (2015b). Consciousness in SLA: A modular perspective. Second Language Research,
31, 413–434.
Truscott, J., & Sharwood Smith, M. (2004). The multilingual mind: A modular processing perspec-
tive. Cambridge: Cambridge University Press.
Truscott, J., & Sharwood Smith, M. (2011). Input, intake and consciousness: The quest for a theo-
retical foundation. Studies in Second Language Acquisition, 33(4), 297–528.
Truscott, J. & Sharwood Smith, M. (2016). Representation, processing and code-switching.
Bilingualism: language and cognition. Published online: 1 Aug 2016, pp. 1–14. .doi: https://
doi.org/10.1017/S1366728916000742.
Part III
Bilingualism Across the Life-Span
Chapter 8
Aging in Bilinguals: Normal and Abnormal
Aviva Lerman and Loraine K. Obler
Contents
Introduction 189
Language Abilities during the Normal Aging Process 190
Bilingual Language Abilities During the Abnormal Aging Process 198
Conclusion 205
References 206
Introduction
Clearly, bilinguals of every age have an advantage over monolinguals in terms of

communication because they can communicate in two languages, and therefore
with more people and in more diverse situations. However, the questions that we try
to answer in this chapter relate to specific cognitive and linguistic abilities, and
whether advantages or disadvantages are observed in aging bilinguals when com-
pared to both aging monolinguals and young bilinguals. Both normal and abnormal
aging will be discussed: for the purposes of this chapter, normal aging can be defined
as changes in communication ability among those—bilinguals and monolinguals—
who continue to function in daily life (Obler, Albert, & Lozowick, 1986). Abnormal
aging, by contrast, includes individuals suffering a variety of processes and diseases
that affect successful functioning. Of these, we will first focus on dementia in mono-
linguals, for the purpose of comparison to bilinguals, followed by a brief discussion
of other abnormal aging processes, all in relation to normal aging.
A. Lerman (*) • L.K. Obler

The Graduate Center of the City University of New York, New York, NY, USA
e-mail: avivapolus@hotmail.com

190 A. Lerman and L.K. Obler
Language Abilities during the Normal Aging Process
Monolingual Language Abilities
In monolingual adults, language abilities appear to change relatively subtly over

the adult lifespan, with some areas of language well maintained in healthy older
adults, and others diminishing with old age. For example, vocabulary knowledge,
word recognition ability, syntax, and comprehension skills of noncomplex material
have all been observed to be preserved in old age (Burke, 1997; Goral, 2004; Goral,
Libben, Obler, Jarema, & Ohayon, 2008; Kavé & Nussbaum, 2012; Nicholas,
Obler, Albert, & Goodglass, 1985), with vocabulary definition increasing over the
lifespan, at least up to the age of 50 years (Nicholas et al., 1985), which is thought
to be a result of continued exposure to new words throughout one’s lifetime
(Hartshorne & Germine, 2015). Similarities have also been found between older
(70–85 years) and younger (20–40 years) adults during picture description tasks
(in which adults are asked to describe in detail a picture of a simple or complex
scene) in measures of total word output, percentage of noun types (nouns men-
tioned at least once) and percentage of noun tokens (total nouns mentioned) (Kavé,
Samuel-Enoch, & Adiv, 2009).
On the other hand, lexical retrieval in production has consistently been shown to
be problematic for older adults when compared with younger adults, especially for
nouns but also for verbs to a lesser degree, with a sharp group decline over the age
of 70 years (Au et al., 1995; Conner, Spiro, Obler, & Albert, 2004; Goral, 2004;
Kavé et al., 2009; Nicholas et al., 1985). Word-finding difficulties in older adults
appear to be the result of a deterioration in the ability to locate the phonological
shape—the word-form—of the words rather than in lexical knowledge per se. We
infer this for a number of reasons. First, older adults (here, those above age 70)
respond well to phonemic cues (i.e., providing the first sound or syllable of a target
word) in order to aid retrieval (Kavé et al., 2009; Nicholas et al., 1985). Likewise,
when tested on category- and letter-fluency tasks in which participants are asked to
list as many items as they can in a semantic category (e.g., animals, clothing items)
or starting with a specific letter (often F, A and S) in 1 min, some studies have sug-
gested that older adults perform worse on category-fluency when compared to
younger adults, but not on letter-fluency (Crossley, D’arcy, & Rawson, 1997; Goral,
2004; Mathuranath et al., 2003). Letter-fluency tasks are a type of phonemic cue,
and therefore these studies are consistent with the notion that semantics is better
preserved with advancing age than is retrieval from the phonological lexicon.
Second, older adults experience more tip of the tongue (TOT) states than
younger adults, suggesting difficulty specifically with accessing the phonological
forms of words (Burke, MacKay, Worthley, & Wade, 1991). Finally, during naming
tasks older adults produce more circumlocution errors (i.e., instances where they
define a word rather than speaking it, e.g., for radiator, one of those things near the
wall that gives heat) when compared to younger adults, indicating that their seman-
tic knowledge is relatively preserved (Au et al., 1995; Goral, 2004).
8 Aging in Bilinguals: Normal and Abnormal 191
With regard to narrative production, in which participants are asked to tell the
story of what is going on in a picture, or to tell what happened during a specific
event such as a vacation, differences in the patterns of language use have also been
documented in older adults when compared to younger adults, with greater lexical
diversity often being noted in older adults (Kavé & Nussbaum, 2012; Obler et al.,
2014). This diversity does not appear to result from an outdated lexicon when com-
pared to younger adults, but rather from a certain deviation from the pictures being
described. This deviation does not in itself imply weak linguistic skills; rather older
adults have a different idea as to what makes a good story and they focus on differ-
ent topics from those younger adults focus on (Burke, 1997; Kavé & Nussbaum,
2012; Kavé et al., 2009). Alternatively, Heller and Dobbs (1993) noted that during
narrative production, older adults (60–76 years) labeled fewer objects correctly and
explained or qualified their choice of labels more than younger adults (28–
59 years)—these most commonly took the form of self-hedges, referring to them-
selves while labeling (e.g., I guess it is a sunken ship) or qualified hedges (e.g., some
kind of a … or some type of a ….)
The Connection Between Cognition and Language
It would be impossible to continue to describe language abilities without also look-

ing at the broader picture of cognition which interlinks with and/or underlies the
abilities necessary for processing language (Kohnert, 2013). General cognitive abil-
ities, as well as those more specifically required for language, change throughout
the lifespan, and there is no one age at which people perform at peak on all tasks
measuring cognition (Hartshorne & Germine, 2015). Similar to the language abili-
ties described previously, some cognitive abilities remain stable over the post-
childhood lifespan and some decline in aging (Hartshorne & Germine, 2015). For
example, performance on tasks involving knowledge that is learned incrementally
over the lifespan, such as standard intelligence quotient (IQ) tasks like vocabulary
definition, arithmetic and explaining how two items are similar, tends to peak later
than tasks which require explicit strategies in order to perform them, such as let-
ter–number sequencing, discrimination of pre-exposed faces from novel faces, and
memorizing sets of word pairs (Hartshorne & Germine, 2015).
Furthermore, declines in working memory have been suggested to be the under-
lying cause of reduced linguistic abilities in adults aged 65–80 years old, affecting
comprehension of grammatical complexity and verbal processing (Williams,
Holmes, Kemper, & Marquis, 2003). Similarly, Burke (1997) suggested that tasks
requiring working memory are performed worse with age, whereas those which do
not require retention of information are performed evenly across the lifespan, and
therefore the deficit in aging is not in understanding language but rather in remem-
bering language long enough to understand it. Taken together, the link between
memory aspects of cognition and language across the lifespan is clear, although the
link between other aspects of cognition (e.g., executive functions such as planning
and organizing) and language is only in the early stage of discovery (e.g., Gollan,
Sandoval, & Salmon, 2011).
Still within the boundaries of healthy aging, a variety of health burdens have
been suggested as one reason why cognitive processing declines in older adults
(Conner et al., 2004). While still successfully functioning within healthy limits,
groups of older adults with risk factors for cerebrovascular disease have evidenced
increased deterioration of cognitive functions such as verbal fluency, working mem-
ory, and memory retrieval compared to those without such risk factors (e.g., Brady,
Spiro, & Gaziano, 2005; Conner et al., 2004). This pattern of reduced ability with
increased risk factors has been explained by the cognitive reserve hypothesis, which
is based on looking at changes in the neural substrate in brain areas relating to cog-
nition. Whereas brain reserve is a physical measurement of brain size and neuronal
count, cognitive reserve refers to the flexibility and effectiveness of using this brain
reserve (Tucker & Stern, 2011). When a cognitively active lifestyle is embraced,
greater neural efficiency and capacity develop (or are retained), including the poten-
tial for recruiting compensatory pathways and regions, and therefore some protec-
tion to the neural substrate is offered (Tucker & Stern, 2011). Conversely, the more
risk factors, the greater the damage to brain reserve, and thus to cognitive reserve. A
cognitively active lifestyle can include, among other factors, high levels of intelli-
gence, education (Conner et al., 2004) and, arguably, bilingualism (Bialystok,
Martin, & Viswanathan, 2005), which either provide high levels of cognitive reserve
thereby directly reducing negative effects of aging, or enhance the ability to com-
pensate for them throughout the lifespan (Conner et al., 2004).
Bilinguals, Cognition, and Language
While it is true that there is a strong connection between cognition and language in
monolinguals, as described above, in bilinguals there is an additional connection in
the form of language control. Choosing which language to use at any given time
may seem automatic to many bilinguals, but for the aging brain these processes are
often more of a challenge than is first appreciated. The processes involved are
among those of executive functions: inhibition (i.e., suppression) and monitoring
the communicative process.
Language control arguably requires using areas of the brain less typically used in
language processing, including areas of the pre-frontal cortex also used for general
cognitive processing, such as nonlinguistic interference suppression and online
monitoring (e.g., Abutalebi & Green, 2007; Bialystok, Craik, & Luk, 2008;
Bialystok, Craik, & Luk, 2012; Bialystok, Craik, & Ryan, 2006). Older adults’ life-
time of practice monitoring two languages appears to simultaneously result in
increased cognitive abilities in nonlinguistic cognitive tasks that are known to rely
on pre-frontal cortex: selective attention, inhibitory control, and monitoring two
streams of information, with a proportionally larger advantage being observed in
older adults than in younger adults when compared to older and younger
onolinguals (Bialystok, Craik, Klein, & Viswanathan, 2004; Bialystok, Craik, &
m
Ruocco, 2006; Salvatierra & Rosselli, 2011). In addition, Bialystok, Craik, and
Ryan (2006) found enhanced functioning of the anterior language area (Broca’s
area), as measured by functional magnetoencephalography (MEG) in bilinguals
when compared to monolinguals while carrying out a nonlinguistic cognitive task.
Taken together, the association between language and improved cognition in bilin-
guals appears to be based on improved neural connections in bilinguals in some of
the classic brain areas of both language and cognition.
However, not all studies have observed such a bilingual cognitive advantage
across the lifespan (e.g., de Bruin, Bak, & Della Sala, 2015; Zahodne, Schofield,
Farrell, Stern, & Manly, 2014) and there is still some discussion in the literature
regarding which age groups and which tasks show the greatest advantage, if any. On
the one hand, several studies support the view that a bilingual advantage exists
across the lifespan in cognitive tasks measuring selective attention, inhibitory con-
trol, or monitoring two streams of information. For example, Bialystok, Craik, and
Ryan (2006) found that 68-year-old adults were slower than 20-year-old adults in
performing a task designed to test response suppression, inhibitory control and task-
switching (in this case, the ability to switch between languages in response to a
cue), but the change across age was less extreme in bilinguals than it was in mono-
linguals. More specifically, Salvatierra and Rosselli (2011) found that 60-year-old
bilinguals were better than 60-year-old monolinguals at inhibitory control but only
under a simple task condition (e.g., the Simon task, which involves two different
colored squares on either side of the screen, and participants hit a button on the left
or right side of the keyboard depending on the color shown, rather than on the
squares’ location), and not under a complex condition (e.g., the Simon task as
above, but with four different colored squares). However, they suggested that in a
much older bilingual group a difference may be found even under the complex con-
dition when compared to monolinguals, since bilingual cognitive advantages are
subtle across the lifespan, and nominally older adults may not show the same advan-
tages as considerably older adults (Salvatierra & Rosselli, 2011).
Another example of a bilingual advantage found only in specific age groups
comes from Bialystok et al. (2005) who found that bilinguals performed better than
monolinguals in the Simon task (in the simple, 2-color condition) across the lifes-
pan in early childhood, adulthood and later adulthood, but not in early adulthood
(20–30 years old). They attributed this effect to the cognitive reserve hypothesis,
whereby the positive influence of bilingualism as a boost to development or as pro-
tection against decline is most obvious at the age where children are still acquiring
these skills, or when adults are losing these skills, but not when cognitive control is
at a stable peak in early adulthood (Bialystok et al., 2005). Following this, Bialystok
et al. (2012) suggested that a bilingual advantage for young adults may be more
noticeable in complex cognitive tasks rather than simple ones.
By contrast, in a longitudinal study, Zahodne et al. (2014) found that cognitive
function deteriorated at a similar rate over time for both older adult monolinguals
and bilinguals, and no single cognitive domain showed a rate of change over time
that was associated with bilingualism. Similarly, de Bruin et al. (2015) found no
differences between active bilinguals, inactive bilinguals, and monolinguals, all

over the age of 60 years old, in their executive control abilities after carefully match-
ing for lifestyle, socioeconomic status, education, IQ, age, and gender. Bialystok
et al. (2012) acknowledge that at all ages across the lifespan there are some studies
that have reported similar performance between monolinguals and bilinguals on
complex cognitive tasks (e.g., conflict tasks, in which participants must ignore con-
flicting stimuli in order to carry out the task, such as ignoring visual distractors
when completing nonverbal problems). There are also some tasks that show a trend
towards a bilingual advantage but it is unclear whether this advantage is replicable.
For example, with regard to working memory, a clear deterioration has been seen in
older adults when compared to younger adults, but only minor differences were
observed between monolinguals and bilinguals (Bialystok et al., 2012).
Another aspect of cognitive function that has been reported in the literature is
the deterioration of the ability to resist interference from competing languages as
bilinguals get older (Ardila & Ramos, 2008; Kohnert, 2013; Mendez, Perryman,
Pontón, & Cummings, 1999). Although resisting interference from a competing
language is a skill that improves during childhood and remains stable in adulthood,
after the age of 65 years it starts to deteriorate (Kohnert, 2013). Ardila and Ramos
(2008) explain that older bilinguals mix their languages more than younger bilin-
guals because older bilinguals are less able to correctly select the appropriate lan-
guage at a given time or in a certain situation, and to switch when required. This
weakened control of the dual language system in older bilinguals may also contrib-
ute to the finding that cued switching (i.e., switching the language during a naming
task as a response to a specific cue) between two languages is harder for older
bilinguals than for younger bilinguals (Kavé, Eyal, Shorek, & Cohen-Mansfield,
2008; Kohnert, 2013), and older adults find it much harder to keep track of the
language they should be responding in when the switches are cued (Gollan &
Ferreira, 2009). However, Gollan et al. (2011) found that non-cued switching (i.e.,
the bilingual chooses when to switch rather than being told when to) did not show
differences between the age groups, and they attributed this to the fact that non-
cued switching is common in bilinguals, since switching costs are relatively small
compared to the potential benefits of pragmatically switching languages in order to
aid communication when necessary.
It should be noted that, as discussed above, research has pointed to both better
inhibitory control over the lifespan in bilinguals due to practice and increasing
interference of the two languages as bilinguals get older—not in daily life, but
rather in cued switching tasks. This apparent conflict may be due to inhibitory
control often being tested nonverbally, and interference being tested during lin-
guistic tasks. Alternately, the inhibition advantages may be real and the apparent
decline on cued switching tasks may result from efforts to keep up with the task
under the general cognitive slowing that people like Salthouse have documented
(e.g., Salthouse, 1996).
Finally, in relation to cognition and languages in normal aging, Kavé et al. (2008)
revealed a strong link between cognitive performance and number of languages spo-
ken in 75–95-year-olds. They conducted a longitudinal study over 12 years and con-
cluded that knowing and using multiple languages predicts cognitive p erformance,
whereby the more languages spoken (1–4 or more) the better the performance on
cognitive tasks (Kavé et al., 2008). Their research determined that the number of
languages that participants spoke predicted their performance on two cognitive
screening tests (the Mini Mental State Examination—MMSE, and the Katzman cog-
nitive screening test) even when other variables (testing age, age at immigration, or
years of education) were accounted for. They also found that the strength of predic-
tion of cognitive scores based on number of languages spoken was even stronger in
the noneducated group than the educated group. From this they suggested that mul-
tilingualism may have protected both educated and noneducated older adults from
cognitive decline, but the prediction factor was weaker in the educated group because
of the positive influence that education itself has on cognition over the lifespan.
Kavé et al. (2008) further suggested that those whose second (L2) or third lan-
guage (L3) was their most proficient, as opposed to their first language (L1), are
likely to be individuals who had compelling reasons for learning the L2 (e.g., immi-
gration, that required them to invest more cognitive effort into learning the new
languages throughout their lifespans than those learning another language out of
choice). They argued that this extra effort to learn another language may have
increased their cognitive reserve in old age more than those who chose to learn
another language (Kavé et al., 2008). Similarly, Zahodne et al. (2014) suggest that
there is not enough evidence to date regarding whether bilingualism itself causes
superior cognitive skills or whether superior cognitive skills aid the acquisition of a
second language. They proposed that for the late bilinguals (i.e., those that learned
the L2 late in life) that they studied, both a higher level of education and good pre-
morbid cognitive abilities may have influenced their cognitive advantage more than
their bilingualism did (Zahodne et al., 2014).
Language Changes in the Normal Aging Process of Bilinguals
The two languages of a bilingual are not stable over the lifespan, both due to aging
effects of language and cognition, as well as changes in lifestyle. In some older
bilingual populations, a tendency for older people to withdraw into single language
use, even if they were bilingual for all or most of their lives, has been noted. For
example, Ardila and Ramos (2008) propose that this is because an L2 is more asso-
ciated with work and schooling and during retirement there is a shift back to one’s
home life and family, where the L1 is more commonly used. In addition, due to the
difficulties faced by aging bilinguals with regard to control of the dual language
system, as discussed above, by reverting back to one language these control chal-
lenges can be avoided (Ardila & Ramos, 2008; Mendez et al., 1999). Ardila and
Ramos (2008) argue that bilinguals typically prefer to use their L1 in old age, and
L2 usually declines at a faster rate than L1. However, they also clarify that age of
acquisition of the L2, reason for acquisition (e.g., migration, for work), proficiency
and daily usage will all affect language changes across the lifespan.
Therefore it comes as no surprise that other researchers have found different

p atterns of language in aging bilinguals. For example, Rosselli et al. (2000) studied
Spanish-speakers in the US who learned English as an L2 during childhood or early
adulthood, and found that those who used both their languages daily did not per-
form worse in either language at age 50–70 years old, when compared to monolin-
guals. Likewise Goral et al. (2008) found that for older bilinguals who lived in their
L2 environment, decisions regarding whether a string of letters is a word (lexical
decision) in L1 were much slower than in L2. They concluded, however, that since
only one language was affected, L1 attrition was the cause of a slower response,
rather than a more general decline of language due to aging (Goral et al., 2008).
This could also be true for bilinguals who retreat into their L1 as they age and then
experience attrition of L2 due to reduced use. The pattern of language change
described by Goral et al. (2008) follows the opposite pattern to that described by
Ardila and Ramos (2008), but in both studies the bilinguals are using one language
less than the other (e.g., after retirement, migration, less opportunity to socialize
etc.) which might appear to lead to attrition in the other language (either L2 or L1,
depending on which language is used less). It should be noted that attrition is a
dynamic process of a slow decline of language abilities over time and could mani-
fest itself as difficulty in retrieving a word, or changes in the representation or orga-
nization of the lexicon (Goral, 2004). Language changes in aging also follow a
dynamic process (e.g., De Bot, Lowie, & Verspoor, 2007) but, as described previ-
ously, the difficulties appear to be more phonologically based than founded on
semantic weakness. Aging and reduced language use, however, are not exclusive of
one another, since both age and proficiency level interact in such a way that they
jointly affect the progression and the rate of language attrition (Goral, 2004).
In order to understand specific language changes in older bilinguals, we may
compare them to changes in older monolinguals. As discussed above, a number of
areas of language are preserved in old age in monolinguals, including comprehen-
sion skills, vocabulary knowledge, and word recognition ability, whereas lexical
retrieval abilities are repeatedly seen to decline in old age. Older, proficient bilin-
guals generally perform to a similar level as older monolinguals on a variety of
language tests, including free spontaneous fluency tasks (in a picture description
task), letter-fluency and sentence-repetition tasks (Rosselli et al., 2000), and there
are similarities in performance between the two languages of the bilingual (Obler
et al., 1986). As with monolinguals, bilinguals also deteriorate in their lexical
retrieval skills as they get older, but the pattern of change is complex.
First, when tested in one of their languages, bilinguals have lower lexical retrieval
skills than monolinguals generally, usually attributed to bilinguals having less expe-
rience with each language, as well as a necessity to inhibit the language not being
used in order to produce words in the target language (Bialystok et al., 2008).
Gollan, Montoya, Cera, and Sandoval (2008) developed their weaker links hypoth-
esis to explain their finding that both older and younger bilinguals show slower
naming than older and younger monolinguals, and that this difference was larger for
low-frequency words where reduced language use would be more likely to affect
words’ retrieval than for high-frequency words. Similarly, they found that d ifferences
in naming times between the dominant and the nondominant language of younger
and older bilinguals were more evident during naming of low-frequency words (that
are rarely encountered in daily life) than those that are high-frequency (relatively
often encountered). In older bilinguals, when using their nondominant language,
age-related slowing of naming occurred only for high-frequency words. They
argued that because bilinguals speak and hear any given word less than the respec-
tive monolinguals, the links between the phonological shape of a given word and its
meaning are weaker for the bilingual than for the monolingual as posed by the
weaker links hypothesis. Moreover, since frequency effects of naming were more
pronounced in the nondominant language than the dominant one, for their partici-
pants, this is further consistent with the hypothesis (Gollan et al., 2008).
Second, the specific tests and stimuli used to examine word retrieval abilities in
bilinguals may be problematic, since many items from word naming tests, such as
the Boston Naming Test (BNT; in which participants are asked to name a set of
black-and-white line drawings of more familiar items (e.g., bed, tree) and later less-
familiar items (e.g., protractor, trellis)) are more likely to be learned in a school
environment than at home. If tested on these items in their home language, bilin-
guals are likely to be at a relative disadvantage compared to monolinguals who did
all their schooling in one language, even when tested in old age (Acevedo &
Loewenstein, 2007). Similarly, bilinguals named pictures better if the words were
cognates (related in the two languages both in sound/spelling and in meaning, e.g.,
English camel, Spanish camello) in their two languages than if they were not
(Gollan, Fennema-Notestine, Montoya, & Jernigan, 2007), which would therefore
differently affect naming scores in bilinguals with different pairs of languages. Type
of lexical retrieval task (e.g., list-generation vs. picture-naming) could also affect
naming ability, for example category-fluency has been suggested to be more nega-
tively affected in bilinguals than letter-fluency (Gollan, Montoya, & Werner, 2002;
Portocarrero, Burright, & Donovick, 2007). Bilinguals have been seen to retrieve
fewer words in semantic, letter and proper name categories than monolinguals, but
the semantic category-fluency showed the largest differences between groups, and
it has been suggested that this effect carries over into older adults as well (Rosselli
et al., 2000). However, Salvatierra, Rosselli, Acevedo, and Duara (2007) found the
opposite, whereby healthy aging bilinguals performed better on category-fluency
than letter-fluency.
Third, language proficiency may also affect the naming skills of older bilinguals,
whereby a high proficiency in spoken abilities in L2 can reduce age-related decline
in lexical retrieval in L1 (Ashaie & Obler, 2014). Although this reduction of age-
related decline in older adults is closely related to education level in the Ashaie and
Obler study, in uneducated bilinguals this advantage still remains strong.
As can be seen from the discussion above, advantages and disadvantages exist
for bilinguals (relative to monolinguals) in the realms of cognition and language,
but they are not altogether consistent over the general population of bilinguals.
Small differences or changes in language background, proficiency, age of acquisition,
and language use will affect the abilities of a bilingual in specific tasks, and there-
fore when comparing abnormal aging populations to normal, healthy bilinguals,
these small changes must be taken into account when using norms for cognitive and
linguistic tests. If they are not, it becomes highly detrimental to try to diagnose
abnormal aging based on unreliable norms for the normal bilingual population.
ilingual Language Abilities During the Abnormal Aging

B
Process
Abnormal aging can manifest itself in a variety of ways, both sudden, such as in a
stroke or sudden illness, and progressive, such as in dementia and Parkinson’s dis-
ease, where decline is more gradual over time. In this section we will begin by dis-
cussing language changes in dementia, and dementia onset, followed by Parkinson’s
disease (the only other progressive disease discussed in the literature to date in
terms of bilingual language abilities). Finally we will briefly discuss bilingualism
and stroke, since an in-depth discussion of bilingualism and aphasia is presented in
chapter Psychopathology and bilingualism.
Dementia
Dementia is a neurodegenerative disease that has several subtypes. The most com-
mon type of dementia in the elderly population is Alzheimer’s disease (Roman,
2003; Stevens et al., 2002). Alzheimer’s disease in older adults can be characterized
by a progressive decline of episodic and working memory, followed by language
deficits (Manchon et al., 2015, p. 91). Other subtypes of dementia which directly
involve or are closely related to language loss include (1) vascular dementia, defined
as loss of cognitive function resulting from ischemic, hypoperfusive, or hemorrhagic
brain lesions due to cerebrovascular disease or cardiovascular pathology … caused
by … multiple strokes … or by a single stroke (Roman, 2003, p. S296); and (2) pri-
mary progressive aphasia (PPA)—semantic dementia subtype, which causes grad-
ual damage to the semantic system over time, resulting in the loss of semantic
memory for both words and real objects (Mendez, Saghafi, & Clark, 2004). Other
subtypes of dementia, such as frontotemporal dementia and dementia with Lewy
bodies, are less associated with language impairment.
The question has been asked whether language deterioration in dementia is due
to an inaccessibility of an intact language system, the loss of that system as brain
atrophy progresses or the combination of both (Hyltenstam & Stroud, 1994) and to
date there is no definitive answer. Clinically, language in dementia usually mani-
fests itself initially with word retrieval difficulties, progressing to deficits in oral
production and a reduction in discourse quality and quantity, and at the later stages
affecting language comprehension and written language (Manchon et al., 2015;
Mendez et al., 1999; Obler & Albert, 1984). People with dementia show increasing
difficulty with language over time—both semantic and lexical information, as well
as pragmatic information—during communication (Hyltenstam & Stroud, 1994).
On the other hand, more automatic forms of language, such as counting and sen-
tence repetition, are preserved for longer (Manchon et al., 2015).
Bilingual Dementia
Bilinguals who suffer from dementia follow similar patterns to normal aging bilin-
guals but much more pronounced, such that (1) many researchers, although not all,
have documented a shift towards L1 use over L2 use, with a faster decline in L2 than
L1, (2) word finding difficulties are one of the first symptoms of bilingual dementia,
and (3) language control is compromised, so that cross-language influence is more
pronounced in dementia, producing more interference and codeswitching from the
nontarget language than in normal aging (Ardila & Ramos, 2008; Gollan, Salmon,
Montoya, & da Pena, 2010; Mendez et al., 1999). We shall now discuss these find-
ings in more detail.
It has been argued that language deteriorates more rapidly in L2 than in L1, in
bilinguals with dementia (Ardila & Ramos, 2008), and much research supports this.
For example, Mendez et al. (1999) described how caregivers of people with demen-
tia reported that their patients all preferred using their L1 over their L2, and when
they did use their L2, there was considerably more codeswitching from their L1 into
L2 than from L2 into L1. They explained this finding using the last in, first out
theory in dementia, whereby more recent linguistic information, assumed to be
based more on declarative knowledge (i.e., facts or grammatical rules one can artic-
ulate) is less retained than linguistic information learned in childhood which is
assumed to be based more on procedural (automatic) knowledge (Mendez et al.,
1999); declarative knowledge is supposed to be more affected by dementia than
procedural knowledge (Mendez et al., 1999). In addition, Mendez et al. (1999) sug-
gested that a retreat to L1 use could also be due to the exacerbation of cross-language
difficulties seen in normal aging of bilinguals.
In another study which found better lexical naming in L1 than in L2 for four
bilingual patients with Alzheimer’s disease, Meguro et al. (2003) supported the
theory that Alzheimer’s disease affects declarative knowledge more than procedural
knowledge based on the patterns of language deterioration that they observed: in
both languages tested (Japanese and Portuguese), irregular items, which are learned
rather than derived from a set of rules, (i.e., Kanji in Japanese, irregular words in
Portuguese) were seen to be more impaired than regular items (i.e., Kana in
Japanese, regular words in Portuguese) (Meguro et al., 2003).
Studies of patients with PPA-semantic dementia subtype have shown a similar
pattern, with one study showing better linguistic skills in L1, even though it was
used less during adulthood than L2 (Larner, 2012), and another study concluding
that lexical naming was progressively more impaired in L2 and L3 when compared
to L1 (Mendez et al., 2004).
Ardila and Ramos (2008) also support the first in, last out theory, suggesting that
a retreat to L1 further reduces L2 abilities due to lack of use. In addition, they point
out that when assessing bilinguals with dementia for cognitive impairment, it is
recommended to test in L1, if not both L1 and L2, since people with dementia
appear to function better cognitively in an L1 environment (see also Terrazas-
Carrillo, this volume); this also extends to choosing a caregiver based on the lan-
guage or languages they speak (Ardila & Ramos, 2008).
However, other studies have contradicted these findings, showing that there are
few or no differences between the two languages of a bilingual with dementia. For
example, Manchon et al. (2015) found that in a group of patients with Alzheimer’s
disease, L1 and L2 were similarly impaired at the levels of semantic, lexical, and
syntactic processing, when compared to a control group with similar language
backgrounds. Compared with this control group, patients with Alzheimer’s dis-
ease performed worse on all language tests except those that tested more auto-
matic skills, such as counting and sentence repetition—similar to previous
findings in monolinguals with dementia (Manchon et al., 2015). They explained
that a deterioration in both L1 and L2 supports the view that cortical representa-
tions of the two languages of a bilingual overlap, and therefore brain atrophy
caused by dementia affects both languages in a similar way. A study by Gómez-
Ruiz, Aguilar-Alonso, and Espasa (2012) supports this hypothesis, since they
found that in a group of Catalan–Spanish bilinguals with Alzheimer’s disease, a
parallel impairment in L1 and L2 was seen in lexical retrieval, vocabulary rich-
ness (as measured for spontaneous speech with a type/token ratio), and abilities in
comprehension of complex grammatical structures. Again, as with monolinguals
in the initial stages of Alzheimer’s disease, automatic linguistic skills were pre-
served in both languages, along with comprehension of words and simple syntac-
tic structures (Gómez-Ruiz et al., 2012).
Similarly, Costa et al. (2012) found that Alzheimer’s disease appears to affect
both languages of early, highly proficient bilinguals to a similar extent. They also
noted the parallels to healthy, aging bilinguals, whereby cognate and frequency
status affected word retrieval in the following ways in bilinguals with and without
Alzheimer’s disease: (1) cognate and high-frequency words were retrieved more
than noncognates and low-frequency words, (2) these effects were more pro-
nounced in the nondominant language than in the dominant one, and (3) as the
cognitive decline increased, the cognate and word-frequency effects also increased
(Costa et al., 2012).
Other researchers have suggested that order of acquisition is not the driving fac-
tor behind language loss in one language over the other in bilingual dementia; rather
language dominance or recency of language use is. For example, Machado,
Rodrigues, Simoes, and Santana (2010) described a case of a 56-year-old bilingual
with PPA-semantic dementia subtype who lived in Portugal until age four, then
moved to France and lived there or in other French-speaking countries until age 42,
and then moved back to Portugal. Although he was proficient in both languages, his
French deteriorated at a much faster rate than his Portuguese as his semantic demen-
tia progressed. The researchers argued that the recency of use determined the
d eterioration, more than order of acquisition: although the language less used at the
onset of dementia symptoms was not the first language, both languages were
acquired in childhood to high/native-like proficiency, yet the less recently used lan-
guage was considerably more affected by the dementia (Machado et al., 2010).
Meguro et al. (2003) also suggested that the language of the environment might
deteriorate less in bilingual patients with dementia since consistent, daily use of that
language may help prevent its deterioration early on in the disease.
In the same way, language dominance has also been suggested as a factor in the
loss of language in bilingual dementia. For example, Gollan et al. (2010) tested
Spanish–English bilinguals with Alzheimer’s disease and found that the nondomi-
nant language deteriorated faster than the dominant language. They argued that this
is consistent with the hypothesis that the dominant language has stronger semantic
representations than the nondominant language has, since semantic representations
are one of the predominantly affected domains in Alzheimer’s disease.
Another aspect of language deterioration prominent in bilingual dementia is that
of language control and the pragmatic use of each language, which can greatly
affect the communicative interactions and the social integration of people with
Alzheimer’s disease (Hyltenstam & Stroud, 1994). Choosing the correct language
for the interlocutor and maintaining that choice become challenging for many bilin-
guals with dementia (De Santi, Obler, Sabo-Abramson, & Goldberger, 1990;
Hyltenstam & Stroud, 1994), and the patients are often not aware that the interlocu-
tor does not understand the language they have chosen to use (Hyltenstam & Stroud,
1994). De Santi et al. (1990) emphasize that, although dementia severity is corre-
lated with language choice abilities and codeswitching problems, those bilinguals
with dementia who acquired their L2 simultaneously with or shortly after their L1
appear to have fewer problems with language-choice abilities than those who
acquired their L2 later in life. In addition, Hyltenstam and Stroud (1994) found that
patients who were highly proficient in their L2 were better able to control their lan-
guage choices through into the later stages of dementia than those whose L2 was not
highly proficient before the onset of dementia. The authors suggested that those
with higher L2 proficiency before the onset of the dementia required fewer resources
to activate the L2 and inhibit their L1.
In terms of codeswitching, Hyltenstam and Stroud (1994) noted that the extent
of language deterioration was not an indication of the amount of inappropriate
codeswitching. Friedland and Miller (1999) found that some, but not all, bilinguals
with dementia show inappropriate codeswitching, and that codeswitching is most
pronounced from L1 into L2 when proficiency of L2 is low.
Overall it can be seen that in bilinguals with dementia there is a deterioration
of language skills that follows a similar pattern to that of healthy aging bilinguals,
but is more pronounced, especially in areas such as choosing which language one
will speak and maintaining use of that language, and regression into L1 or into the
language of the environment to the detriment of the other language. Similarly,
although healthy bilinguals can often benefit from the knowledge of two lan-
guages by using one to fill in for the other when necessary, bilinguals with demen-
tia are unable to use their two languages to their advantage during communication
(Ardila & Ramos, 2008). One other area concerning bilingual dementia is the
relationship between bilingualism and dementia onset. We will now address that
question in detail and discuss the relevant literature to date.
Advantage as a Delay to the Onset of Dementia
The idea of cognitive reserve, whereby a greater brain reserve exists as well as an
efficient use of this reserve occurs when a cognitively active lifestyle is embraced
(Conner et al., 2004; Tucker & Stern, 2011) fueled the hypothesis that bilingualism,
as one of the possible definitions of a cognitively active lifestyle, may delay the emer-
gence of dementia (Fischer & Schweizer, 2014; Perani & Abutalebi, 2015). Since the
neural basis for bilingualism is accepted to be more extensive than once speculated,
based on a broad neural network, this neural basis should be able to resist neurodegen-
eration or compensate for it (Fischer & Schweizer, 2014). As mentioned above, Kavé
and her colleagues have demonstrated, moreover, that knowledge of multiple lan-
guages should delay cognitive decline in accordance with the number of languages.
Bialystok, Craik, and Freedman (2007) were the first to test out the hypothesis
that bilingualism delays the onset of dementia, and they found that in bilinguals who
acquired the L2 before early adulthood, dementia onset was on average 4.1 years
later than monolinguals. Since then their findings have been supported by a series of
studies. Most recently, they took an obverse approach, studying the amount of brain
atrophy in monolinguals and bilinguals with Alzheimer’s disease. In monolingual
and bilingual patients matched on age, cognitive level, and other factors, the bilin-
guals had more Alzheimer’s pathology on computerized tomography (CT) scans than
monolinguals (Schweizer, Ware, Fischer, Craik, & Bialystok, 2012), suggesting that
bilinguals symptomatically continue to cope and to function better than monolin-
guals, even after the disease has already started to atrophy the brain.
Since Bialystok et al.’s (2007) pioneering study, a growing body of research has
been added to this field, identifying a number of factors that confound the result of
a delay of dementia onset in bilinguals. For example, education level has been sug-
gested to be related to a cognitively healthy lifestyle (Ashaie & Obler, 2014; Conner
et al., 2004; Kavé et al., 2008), and if an upper limit of cognitive reserve is reached
due to education (or other factors), bilingualism may not have any further effects
(Ashaie & Obler, 2014). Correspondingly, Gollan, Salmon, Montoya, and Galasko
(2011) found that the relative benefit of bilingualism to a later dementia onset held
true only for Spanish-speaking immigrants to the US who had a low level of educa-
tion, whereas those with an average or high level of education were not protected by
their bilingualism against the development of dementia, suggesting that maximum
cognitive reserve had already been reached with high levels of education.
Immigration has also been suggested as another influencing factor on bilinguals’
apparent delay in the onset of dementia. For example, Chertkow et al. (2010) studied
nonimmigrant bilinguals, immigrant bilinguals, and immigrant multilinguals and
found no bilingual advantage to later onset of Alzheimer’s disease in the nonimmi-
grant group when compared to monolinguals, whereas for immigrants a delay of

dementia onset of up to 5 years was observed. They found a correlation between the
number of languages spoken and dementia onset, whereby those who spoke more
languages had a later age of onset of symptoms and diagnosis, accordingly. They
concluded that multilingualism delays the onset of dementia, along with bilingual-
ism in immigrants (Chertkow et al., 2010). Therefore bilingualism itself might not
be considered enough of an active cognitive lifestyle without an added factor of
effort (e.g., from immigration or a third or fourth language; Chertkow et al., 2010).
As described above, Kavé et al. (2008) suggest that those who have been forced by
circumstances to acquire another language, such as after immigration, rather than
for interest, may have invested more cognitive effort while learning, or continually
using, another language than those who chose to learn one, and this may have been
the reason that immigrants’ cognitive reserve was larger than nonimmigrants’, and
therefore more protective against deterioration as they aged.
On the other hand, immigrants do not constitute a random sample of any given
population, and they cannot be compared to bilinguals living in an L1 environment,
since immigrants in an L2 environment bring with them potentially confounding
factors to bilingualism (Perani & Abutalebi, 2015; Woumans et al., 2015). For
example, first- and second- generation immigrants may have differences in many
environmental factors, such as lifestyle, education, attitudes to health and access to
health services etc. when compared with the native population, as Perani and
Abutalebi (2015) point out. Also, immigrants in the studies discussed so far are
mostly adult learners of the L2, as opposed to the simultaneous or early bilinguals
described in Bialystok et al.’s (2007) first study. Early and late bilinguals likely
require different cognitive efforts when acquiring their L2, and this could poten-
tially affect cognitive reserve levels (Perani & Abutalebi, 2015).
Therefore, two more recent studies were conducted in places where nonimmi-
grant bilingual populations were available, in order to ascertain whether a bilingual
advantage in dementia onset delay exists in nonimmigrant bilingual populations.
The first study, by Alladi et al. (2013), was conducted in Hyderabad, India, where
all the participants were drawn from the same environment. They found a 4.5-year
delay not only in the onset of Alzheimer’s disease, frontotemporal dementia and
vascular dementia, but also trends towards a delay of onset in dementia with Lewy
bodies and mixed dementia. The researchers argued that in this part of India the use
of two or more languages was daily, with many participants being illiterate, so con-
founding effects of increased education correlating with increased proficiency of
bilinguals was not a concern (Alladi et al., 2013). They concluded that, due to the
nature of this population and the bilinguals’ high proficiency over the lifespan, no
additional benefit was necessary for them to reach maximum cognitive reserve
(Alladi et al., 2013). A second study by Woumans et al. (2015) conducted in Belgium
found similar results, even when controlling for confounding variables such as sex,
education, occupation level, and initial Mini Mental State Examination (MMSE)
scores, whereby bilingualism delayed the onset of symptoms of Alzheimer’s disease
by 4.6 years, and the age of diagnosis by 4.8 years. This delay was evidenced for
both early and late bilinguals (L2 age of acquisition 0–25 years). The researchers
concluded that bilingualism contributes to cognitive reserve when bilinguals are

nonimmigrants and living in an L1-dominant environment, in addition to those
immigrants and nonimmigrants living in an L2 environment, as other authors have
previously reported (Woumans et al., 2015).
To summarize, it appears that a bilingual advantage of 4–5 years in dementia
onset delay does exist for both immigrants and nonimmigrants; those living in an L1
environment and those in an L2 one; those with high and low education levels
(although not in a parallel manner); for early and late bilinguals; and not only for
Alzheimer’s disease but for other types of dementia as well.
Bilingualism and Parkinson’s Disease
A related phenomenon to the dementias described above is Parkinson’s disease,

which itself results in dementia in at least one-third of cases (Aarsland, Zaccai, &
Brayne, 2005). A small number of studies have been conducted on language abili-
ties of bilinguals with Parkinson’s disease, a degenerative disease known to cause
mild to moderate language deficits, including speech dysarthria and morpho-
syntactic impairments (Zanini et al., 2004). In two studies of Friulian–Italian bilin-
guals with Parkinson’s disease, morpho-syntactic deficits were found to be
significantly more pronounced in L1 than in L2, when compared to healthy con-
trols, both in comprehension and production, after linguistic errors in L2 were taken
into account for both groups (their L2 errors were similar across groups) (Zanini,
Tavano, & Fabbro, 2010; Zanini et al., 2004). This finding has been replicated in a
study of Azari–Farsi bilinguals with Parkinson’s disease, where those with
Parkinson’s disease had a more pronounced syntactic deficit in L1 than L2—a find-
ing which was not observed in the healthy control group (Johari et al., 2013). These
findings have been explained based on neural evidence, which suggests that the
procedural learning process is recruited more when acquiring morpho-syntax in L1,
and which involves the basal ganglia, among other regions, to do so. On the other
hand, declarative learning is recruited in order to acquire the grammatical rules of
L2 which, along with lexical-semantic processing of both L1 and L2, is mainly
subserved by the temporal cortex and temporoparietal regions (Johari et al., 2013;
Zanini et al., 2004, 2010). Since Parkinson’s disease is known to primarily affect
subcortical structures, including the basal ganglia, this would explain why morpho-
syntax is more affected by Parkinson’s disease in L1 than in L2 (Zanini et al., 2004).
Stroke and Bilingualism
Having a stroke can affect cognitive and linguistic abilities in a variety of ways. One
study, by Alladi et al. (2016) found that bilingualism affects the chance of having a
cognitive impairment because of the stroke, in that 77% of monolinguals had some
cognitive impairment after a stroke (note that they included participants with
vascular dementia and vascular mild cognitive impairment), compared with 49% of
bilinguals. They attributed this result to bilinguals having better cognitive reserve
than monolinguals, which helped their post-stroke recovery.
Incidence of aphasia after stroke is similar between bilinguals and monolinguals
(10.5% and 11.8% respectively) (Alladi et al., 2016). However, recovering language
patterns may be parallel in L1 and L2 (the more common pattern), or they may
recover differentially in each language (Albert & Obler, 1978; Mendez et al., 1999;
Paradis, 1993). When Obler and Albert (1977) considered a set of 106 cases of
bilingual aphasia in the literature to see whether age at aphasia onset had an effect
on the language patterns of aphasia, they found that, in differential aphasia recovery,
the language of the environment recovered better than chance for individuals up to
age 65, but after that age neither the most-used language nor the first-learned one
recovered predictably. How the two or more languages of a multilingual recover
depends on a number of factors which interact with each other, including order of
acquisition, proficiency levels, language use and the language of the environment
(Mendez et al., 1999). This complex relationship is discussed in detail in chapter
Psychopathology and bilingualism—Dissociated language disorders in bilinguals:
aphasia, alexias, dyslexia, dysphasia, dementia.
Conclusion
When aging is imposed on a bilingual or multilingual brain, the normal language

changes seen in older monolinguals are affected in various ways, and additional
phenomena specific to bilingualism have been reported. Bilingualism researchers
have focused more on the lexical retrieval problems commonly reported with
advancing age in monolinguals than they have on the difficulties comprehending
complex syntax. Lexical retrieval is slower and/or less accurate in older bilinguals—
and in younger ones—compared to monolinguals, because the bilinguals have used
each individual words less than their monolingual counterparts.
Among bilingualism-specific advantages, the language-switching that is used
in bilingual populations to greater or lesser extent has been linked to more general
cognitive advantages that extend beyond language, such as inhibition, or to those
that underlie more language-specific behaviors, such as codeswitching. Those that
are more natural seem not to change with age, whereas those that are less natural
(e.g., cued switching) decline with advancing age, perhaps due, we have argued,
to general cognitive slowing. The biggest cognitive advantage that has been
reported for bilinguals, of course, is the delay of 4 or 5 years in onset of dementia
behaviors, and, by the report of Alladi and her colleagues (Alladi et al., 2016),
cognitive performance in individuals with stroke and related brain lesions.
Advantages on cognitive testing can be seen even in late bilinguals who learned
their L2 in school and remained in the L1 environment (Bak, Nissan, Allerhand,
& Deary, 2014). Substantial cognitive advantages pertain to bilingualism in
healthy older adults as well, especially for inhibition, according to many studies.
The advantages of bilingualism in older adults, we conclude, extend well beyond
the ability to communicate with more people, as we mentioned originally in our
introduction to this chapter.
Acknowledgement We would like to thank Dr. Mira Goral for her helpful suggestions on an earlier
draft of this chapter.
References
Aarsland, D., Zaccai, J., & Brayne, C. (2005). A systematic review of prevalence studies of demen-
tia in Parkinson’s disease. Movement Disorders, 20(10), 1255–1263. http://doi.org/10.1002/
mds.20527
Abutalebi, J., & Green, D. (2007). Bilingual language production: The neurocognition of lan-
guage representation and control. Journal of Neurolinguistics, 20(3), 242–275. http://doi.
org/10.1016/j.jneuroling.2006.10.003
Acevedo, A., & Loewenstein, D. A. (2007). Performance on the Boston Naming Test in
English-Spanish Bilingual Older Adults: Some considerations. Journal of the International
Neuropsychological Society, 13(2), 212–214.
Albert, M. L., & Obler, L. K. (1978). The bilingual brain: Neuropsychological and neurolinguistic
aspects of bilingualism. NY: Academic Press.
Alladi, S., Bak, T. H., Duggirala, V., Surampudi, B., Shailaja, M., Shukla, A. K., et al. (2013).
Bilingualism delays age at onset of dementia, independent of education and immigration sta-
tus. American Academy of Neurology, 81, 1938–1944.
Alladi, S., Bak, T. H., Mekala, S., Rajan, A., Chaudhuri, J. R., Mioshi, E., et al. (2016). Impact of
bilingualism on cognitive outcome after stroke. Stroke, 47(1), 258–261. http://doi.org/10.1161/
STROKEAHA.115.010418
Ardila, A., & Ramos, E. (2008). Normal and abnormal aging in bilinguals. Dementia and
Ashaie, S., & Obler, L. (2014). Effect of age, education, and bilingualism on confrontation naming
in older illiterate and low-educated populations. Behavioural Neurology, 2014, 1–10. http://doi.
org/10.1155/2014/970520
Au, R., Joung, P., Nicholas, M., Obler, L. K., Kass, R., & Albert, M. L. (1995). Naming ability
across the adult life span. Aging and Cognition, 2(4), 300–311.
Bak, T. H., Nissan, J. J., Allerhand, M. M., & Deary, I. J. (2014). Does bilingualism influence
cognitive aging?: Bilingualism and aging. Annals of Neurology, 75(6), 959–963. http://doi.
org/10.1002/ana.24158
Bialystok, E., Craik, F., & Luk, G. (2008). Cognitive control and lexical access in younger and
older bilinguals. Journal of Experimental Psychology: Learning, Memory, and Cognition,
34(4), 859–873. http://doi.org/10.1037/0278-7393.34.4.859
Bialystok, E., Craik, F. I. M., & Freedman, M. (2007). Bilingualism as a protection against the
onset of symptoms of dementia. Neuropsychologia, 45(2), 459–464. http://doi.org/10.1016/j.
neuropsychologia.2006.10.009
Bialystok, E., Craik, F. I. M., Klein, R., & Viswanathan, M. (2004). Bilingualism, aging, and cog-
nitive control: Evidence from the Simon task. Psychology and Aging, 19(2), 290–303. http://
doi.org/10.1037/0882-7974.19.2.290
Bialystok, E., Craik, F. I. M., & Luk, G. (2012). Bilingualism: Consequences for mind and brain.
Trends in Cognitive Sciences, 16(4), 240–250. http://doi.org/10.1016/j.tics.2012.03.001
Bialystok, E., Craik, F. I. M., & Ruocco, A. C. (2006). Dual-modality monitoring in a classifi-
cation task: The effects of bilingualism and ageing. The Quarterly Journal of Experimental
Psychology, 59(11), 1968–1983. http://doi.org/10.1080/17470210500482955
Bialystok, E., Craik, F. I. M., & Ryan, J. (2006). Executive control in a modified antisaccade task:
Effects of aging and bilingualism. Journal of Experimental Psychology: Learning, Memory,
and Cognition, 32(6), 1341–1354. http://doi.org/10.1037/0278-7393.32.6.1341
Bialystok, E., Martin, M. M., & Viswanathan, M. (2005). Bilingualism across the lifespan: The rise
and fall of inhibitory control. International Journal of Bilingualism, 9(1), 103–119.
Brady, C. B., Spiro, A., III, & Gaziano, J. M. (2005). Effects of age and hypertension status on cog-
nition: The veterans affairs normative aging study. Neuropsychology, 19(6), 770–777. http://
doi.org/10.1037/0894-4105.19.6.770
Burke, D. M. (1997). Language, aging, and inhibitory deficits: Evaluation of a theory. Journal of
Gerontology: Psychological Sciences, 52B(6), 254–264.
Burke, D. M., MacKay, D. G., Worthley, J. S., & Wade, E. (1991). On the tip of the tongue: What
causes word finding failure in young and old adults? Journal of Memory and Language, 30,
542–579.
Chertkow, H., Whitehead, V., Phillips, N., Wolfson, C., Atherton, J., & Bergman, H. (2010).
Multilingualism (but not always bilingualism) delays the onset of Alzheimer disease: Evidence
from a bilingual community. Alzheimer Disease & Associated Disorders, 24(2), 118–125.
http://doi.org/10.1097/WAD.0b013e3181ca1221
Conner, L. T., Spiro, A. I., Obler, L. K., & Albert, M. L. (2004). Change in object naming ability
during adulthood. Journal of Gerontology: Psychological Sciences, 59, 203–209.
Costa, A., Calabria, M., Marne, P., Hernández, M., Juncadella, M., Gascón-Bayarri, J., et al.
(2012). On the parallel deterioration of lexico-semantic processes in the bilinguals’ two lan-
guages: Evidence from Alzheimer’s disease. Neuropsychologia, 50(5), 740–753. http://doi.
org/10.1016/j.neuropsychologia.2012.01.008
Crossley, M., D’arcy, C., & Rawson, N. S. B. (1997). Letter and category fluency in community-
dwelling canadian seniors: A comparison of normal participants to those with dementia of the
Alzheimer or vascular type. Journal of Clinical and Experimental Neuropsychology, 19(1),
52–62. http://doi.org/10.1080/01688639708403836
De Bot, K., Lowie, W., & Verspoor, M. (2007). A dynamic systems theory approach to second
language acquisition. Bilingualism: Language and Cognition, 10(1), 7. http://doi.org/10.1017/
S1366728906002732
de Bruin, A., Bak, T. H., & Della Sala, S. (2015). Examining the effects of active versus inactive
bilingualism on executive control in a carefully matched non-immigrant sample. Journal of
Memory and Language, 85, 15–26. http://doi.org/10.1016/j.jml.2015.07.001
De Santi, S., Obler, L. K., Sabo-Abramson, H., & Goldberger, J. (1990). Discourse abilities in
multilingual dementia. In Y. Joanette & H. Brownell H. (Eds.), Discourse ability and brain
damage: Theoretical and empirical perspectives (pp. 224–235). New York, NY: Springer.
Fischer, C. E., & Schweizer, T. A. (2014). How does speaking another language reduce the risk of
dementia? Expert Review of Neurotherapeutics, 14(5), 469–471. http://doi.org/10.1586/14737
175.2014.892831
Friedland, D., & Miller, N. (1999). Language mixing in bilingual speakers with Alzheimer’s
dementia: A conversation analysis approach. Aphasiology, 13(4–5), 427–444. http://doi.
org/10.1080/026870399402163
Gollan, T. H., Fennema-Notestine, C., Montoya, R. I., & Jernigan, T. L. (2007). The bilingual
effect on Boston Naming Test performance. Journal of the International Neuropsychological
Society, 13, 197–208.
Gollan, T. H., & Ferreira, V. S. (2009). Should I stay or should I switch? A cost–benefit analy-
sis of voluntary language switching in young and aging bilinguals. Journal of Experimental
Psychology: Learning, Memory, and Cognition, 35(3), 640–665. http://doi.org/10.1037/
a0014981
Gollan, T. H., Montoya, R. I., Cera, C., & Sandoval, T. (2008). More use almost always a means
a smaller frequency effect: Aging, bilingualism, and the weaker links hypothesis. Journal of
Memory and Language, 58(3), 787–814.
Gollan, T. H., Montoya, R. I., & Werner, G. A. (2002). Semantic and letter fluency in Spanish-English
bilinguals. Neuropsychology, 16(4), 562–576. http://doi.org/10.1037/0894-4105.16.4.562
Gollan, T. H., Salmon, D. P., Montoya, R. I., & da Pena, E. (2010). Accessibility of the non-
dominant language in picture naming: A counterintuitive effect of dementia on bilin-
gual language production. Neuropsychologia, 48(5), 1356–1366. http://doi.org/10.1016/j.
neuropsychologia.2009.12.038
Gollan, T. H., Salmon, D. P., Montoya, R. I., & Galasko, D. R. (2011). Degree of bilingualism
predicts age of diagnosis of Alzheimer’s disease in low-education but not in highly-educated
Hispanics. Neuropsychologia, 49(14), 3826–3830.
Gollan, T. H., Sandoval, T., & Salmon, D. P. (2011). Cross-language intrusion errors in aging bilin-
guals reveal the link between executive control and language selection. Psychological Science,
22(9), 1155–1164. http://doi.org/10.1177/0956797611417002
Gómez-Ruiz, I., Aguilar-Alonso, Á., & Espasa, M. A. (2012). Language impairment in Catalan-
Spanish bilinguals with Alzheimer’s disease. Journal of Neurolinguistics, 25(6), 552–566.
http://doi.org/10.1016/j.jneuroling.2011.06.003
Goral, M. (2004). First-language decline in healthy aging: Implications for attrition in bilingual-
ism. Journal of Neurolinguistics, 17(1), 31–52. http://doi.org/10.1016/S0911-6044(03)00052-6
Goral, M., Libben, G., Obler, L. K., Jarema, G., & Ohayon, K. (2008). Lexical attrition in younger
and older bilingual adults. Clinical Linguistics & Phonetics, 22(7), 509–522. http://doi.
org/10.1080/02699200801912237
Hartshorne, J. K., & Germine, L. T. (2015). When does cognitive functioning peak? The asynchro-
nous rise and fall of different cognitive abilities across the life span. Psychological Science,
26(4), 433–443. http://doi.org/10.1177/0956797614567339
Heller, R. B., & Dobbs, A. R. (1993). Age differences in word finding in discourse and nondis-
course situations. Psychology and Aging, 8, 443–450.
Hyltenstam, K., & Stroud, C. (1994). Second language regression in Alzheimer’s disease. In
K. Hyltenstam & A. Viberg (Eds.), Progression and regression in language: Sociocultural,
neuropsychological and linguistic perspectives (pp. 222–242). Cambridge, UK: Cambridge
University Press.
Johari, K., Ashrafi, F., Zali, A., Ashayeri, H., Fabbro, F., & Zanini, S. (2013). Grammatical deficits
in bilingual Azari—Farsi patients with Parkinson’s disease. Journal of Neurolinguistics, 26(1),
22–30. http://doi.org/10.1016/j.jneuroling.2012.02.004
Kavé, G., Eyal, N., Shorek, A., & Cohen-Mansfield, J. (2008). Multilingualism and cognitive state in
the oldest old. Psychology and Aging, 23(1), 70–78. http://doi.org/10.1037/0882-7974.23.1.70
Kavé, G., & Nussbaum, S. (2012). Characteristics of noun retrieval in picture descriptions across
the adult lifespan. Aphasiology, 26(10), 1238–1249. http://doi.org/10.1080/02687038.2012.6
81767
Kavé, G., Samuel-Enoch, K., & Adiv, S. (2009). The association between age and the frequency
of nouns selected for production. Psychology and Aging, 24(1), 17–27. http://doi.org/10.1037/
a0014579
Kohnert, K. (2013). Language disorders in bilingual children and adults 2. Melbourne: Plural
Publishing.
Larner, A. J. (2012). Progressive non-fluent aphasia in a bilingual subject: Relative preservation
of ‘mother tongue’. Journal of Neuropsychiatry and Clinical Neuroscience, 24(1), E9–E10.
Machado, A., Rodrigues, M., Simoes, S., & Santana, I. (2010). The Portuguese who could no lon-
ger speak French: Primary progressive aphasia in a bilingual man. Journal of Neuropsychiatry
and Clinical Neuroscience, 22(1), E31–E32.
Manchon, M., Buetler, K., Colombo, F., Spierer, L., Assal, F., & Annoni, J.-M. (2015). Impairment
of both languages in late bilinguals with dementia of the Alzheimer type. Bilingualism:
Language and Cognition, 18(1), 90–100. http://doi.org/10.1017/S1366728914000194
Mathuranath, P. S., George, A., Cherian, P. J., Alexander, A., Sarma, S. G., & Sarma, P. S. (2003).
Effects of age, education and gender on verbal fluency. Journal of Clinical and Experimental
Neuropsychology, 25(8), 1057–1064.
Meguro, K., Senaha, M., Caramelli, P., Ishizaki, J., Chubacci, R., Meguro, M., et al. (2003).
Language deterioration in four Japanese–Portuguese bilingual patients with Alzheimer’s dis-
ease: A trans-cultural study of Japanese elderly immigrants in Brazil. Psychogeriatrics, 3,
63–68.
Mendez, M. F., Perryman, K. M., Pontón, M. O., & Cummings, J. L. (1999). Bilingualism and
dementia. Journal of Neuropsychiatry and Clinical Neuroscience, 11, 411–412.
Mendez, M. F., Saghafi, S., & Clark, D. (2004). Semantic dementia in multilingual patients.
Journal of Neuropsychiatry, 16(3), 381–381. http://doi.org/10.1176/appi.neuropsych.16.3.381
Nicholas, M., Obler, L. K., Albert, M. L., & Goodglass, H. (1985). Lexical retrieval in healthy
aging. Cortex, 21, 595–606.
Obler, L. K., & Albert, M. L. (1977). Influence of aging on recovery from aphasia in polyglots.
Brain and Language, 4, 460–463.
Obler, L. K., & Albert, M. L. (1984). Language in aging. In M. L. Albert (Ed.), Clinical neurology
of aging (pp. 245–253). New York: Oxford University Press.
Obler, L. K., Albert, M. L., & Lozowick, S. (1986). The aging bilingual. In J. Vaid (Ed.), Language
processing in bilinguals: Psycholinguistic and neuropsychological perspectives (pp. 221–231).
Hillsdale, NJ: Erlbaum.
Obler, L. K., Hyun, J. M., Paplikar, A., Higby, E., Goral, M., & Cahana-Amitay, D. (2014).
Measuring lexical retrieval in older adults’ discourse. Presented at the INS, Jerusalem.
Paradis, M. (1993). Bilingual aphasia rehabilitation. In M. Paradis (Ed.), Foundations of aphasia
rehabilitation (pp. 413–419). Oxford: Pergamon Press.
Perani, D., & Abutalebi, J. (2015). Bilingualism, dementia, cognitive and neural reserve. Current
Opinion in Neurology, 28(6), 618–625. http://doi.org/10.1097/WCO.0000000000000267
Portocarrero, J., Burright, R., & Donovick, P. (2007). Vocabulary and verbal fluency of bilingual
and monolingual college students. Archives of Clinical Neuropsychology, 22(3), 415–422.
http://doi.org/10.1016/j.acn.2007.01.015
Roman, G. C. (2003). Vascular dementia: Distinguishing characteristics, treatment, and pre-
vention. Journal of the American Geriatrics Society, 51(5s2), S296–S304. http://doi.
org/10.1046/j.1532-5415.5155.x
Rosselli, M., Ardila, A., Araujo, K., Weekes, V. A., Caracciolo, V., Padilla, M., et al. (2000).
Verbal fluency and repetition skills in healthy older Spanish-English bilinguals. Applied
Neuropsychology, 7(1), 17–24. http://doi.org/10.1207/S15324826AN0701_3
Salthouse, T. A. (1996). The processing-speed theory of adult age differences in cognition.
Psychological Review, 103(3), 403–428. http://doi.org/10.1037//0033-295X.103.3.403
Salvatierra, J., & Rosselli, M. (2011). The effect of bilingualism and age on inhibitory control.
International Journal of Bilingualism, 15(1), 26–37. http://doi.org/10.1177/1367006910371021
Salvatierra, J., Rosselli, M., Acevedo, A., & Duara, R. (2007). Verbal fluency in bilingual Spanish/
English Alzheimer’s disease patients. American Journal of Alzheimer’s Disease and Other
Dementias, 22(3), 190–201. http://doi.org/10.1177/1533317507301792
Schweizer, T. A., Ware, J., Fischer, C. E., Craik, F. I. M., & Bialystok, E. (2012). Bilingualism as
a contributor to cognitive reserve: Evidence from brain atrophy in Alzheimer’s disease. Cortex,
48, 991–996.
Stevens, T., Livingston, G., Kitchen, G., Manela, M., Walker, Z., & Katona, C. (2002). Islington
study of dementia subtypes in the community. The British Journal of Psychiatry, 180(3), 270–
276. http://doi.org/10.1192/bjp.180.3.270
Tucker, A. M., & Stern, Y. (2011). Cognitive reserve in aging. Current Alzheimer Research, 8(4),
354–360.
Williams, K., Holmes, F., Kemper, S., & Marquis, J. (2003). Written language clues to cognitive
changes of aging: An analysis of the letters of King James VI/I. The Journals of Gerontology
Series B: Psychological Sciences and Social Sciences, 58(1), P42–P44. http://doi.org/10.1093/
geronb/58.1.P42
Woumans, E., Santens, P., Sieben, A., Versijpt, J., Stevens, M., & Duyck, W. (2015). Bilingualism
delays clinical manifestation of Alzheimer’s disease. Bilingualism: Language and Cognition,
18(3), 568–574. http://doi.org/10.1017/S136672891400087X
Zahodne, L. B., Schofield, P. W., Farrell, M. T., Stern, Y., & Manly, J. J. (2014). Bilingualism
does not alter cognitive decline or dementia risk among Spanish-speaking immigrants.
Neuropsychology, 28(2), 238–246. http://doi.org/10.1037/neu0000014
Zanini, S., Tavano, A., & Fabbro, F. (2010). Spontaneous language production in bilingual
Parkinson’s disease patients: Evidence of greater phonological, morphological and syntactic
impairments in native language. Brain and Language, 113(2), 84–89. http://doi.org/10.1016/j.
bandl.2010.01.005
Zanini, S., Tavano, A., Vorano, L., Schiavo, F., Gigli, G. L., Aglioti, S. M., et al. (2004). Greater syn-
tactic impairments in native language in bilingual Parkinsonian patients. Journal of Neurology,
Neurosurgery & Psychiatry, 75(12), 1678–1681. http://doi.org/10.1136/jnnp.2003.018507
Chapter 9
Dissociated Language Disorders in Bilinguals
Alfredo Ardila
Contents
I ntroduction 211
Aphasia 212
Alexia 214
Developmental Dysphasia 215
Developmental Dyslexia 216
Dementia 218
Conclusions 221
References 221
Introduction
Bilingual individuals can sometimes present language disorders that are not com-
pletely equivalent in their first (L1) and second (L2) languages. This situation can
be observed both in acquired (as a result of a brain pathology) and developmental
(as a delay in the normal language development) language disorders. Furthermore,
dissociated language disorders can be found not only in oral language, such as in
acquired aphasia and developmental dysphasia, but also in written language, such as
acquired alexia and developmental dyslexia.
Language dissociations of cognitive abilities could also be observed in bilinguals
who suffer from dementia syndromes; in these cases, the cognitive decline that
characterizes the dementia is not equally severe and parallel when using the L1 and
L2. The ability to use the L2—especially in late bilinguals—tends to decrease
during normal but very particular abnormal aging. Weak language mastery will nec-
A. Ardila (*)
e-mail: ardilaa@fiu.edu

212 A. Ardila
essarily impact scores in intellectual testing; this is valid not only during normal and
abnormal aging, but also during L2 acquisition: children with a weak knowledge of
the L2 will have a poor academic performance when testing in the L2.
In this chapter, five neuropsychological conditions that can potentially be lin-
guistically dissociated in bilinguals will be examined: aphasia, alexia, dysphasia,
dyslexia, and dementia (see also Learman & Obler, this volume).
Aphasia
Aphasia can be defined as a language disturbance associated with brain pathology

(Ardila, 2014; Benson & Ardila, 1996). In bilingual individuals, language distur-
bance is usually observed in both languages, although the specific aphasia clinical
manifestations can be different. The aphasia profile (i.e., type of aphasia) and sever-
ity, as well as the pattern of recovery are not necessarily coincidental in L1 and L2.
Aphasia Profile
Clinical observations have demonstrated that bilingual individuals with aphasia do

not necessarily manifest the same language disorders to the same degree of severity
in both languages (Albert & Obler, 1978). Aphasia can be parallel (where both
languages are impaired in a similar way) or dissociated (where there is a different
aphasia profile for each one of the languages). Fabbro (2001) observed in a sample
of 20 bilingual aphasics, parallel aphasia in 65% of the subjects; in the rest (35%)
aphasia was dissociated: 20% showed a greater impairment of L2, while 15% of the
patients showed a greater impairment of L1. These percentages can be considered
relatively typical.
It is assumed that parallel aphasia is usually found in early bilinguals because it
is associated with a similar brain representation of L1 and L2; while dissociated
aphasia is characteristic of late bilinguals. In this subgroup of bilinguals, the brain
representation of L1 is not completely coincidental with the brain representation of
the L2 (Goral, Levy, & Obler, 2002; Green, 2008). On the one hand, the L2 seems
to be acquired through the same neural structures responsible for L1 acquisition; on
the other, neural differences may be observed in terms of more extended activity of
the neural system mediating L2 processing (Abutalebi, 2008). Indeed many studies
have reported that later acquired languages may involve broader activation locations
than the first acquired language, largely overlapping. However, sometimes distinct
cortical areas are involved in the comprehension and production of L1 and L2
(Obler, Hyun, Conner, O’Connor, & Anema, 2007).
Kim, Rilkin, Lee, and Kirsch (1997) ran an fMRI study where participants used
widely different L1 and L2 pairs of languages (English–French, Korean–English,
Spanish–English). While silently describing to themselves what they did during the
9 Dissociated Language Disorders in Bilinguals 213
morning, afternoon, or evening of the previous day, it was found that within the
frontal lobe language area (Broca’s area), L2 s acquired in adulthood (i.e., in late
bilinguals) were spatially separated from the native language. However, when
acquired during the early language acquisition stage of development (i.e., in early
bilinguals), the L1 and L2 tended to be represented in common frontal cortical
areas. In both late and early bilingual subjects, the temporal lobe language-sensitive
regions (Wernicke’s area) also showed effectively little or no separation of activity
based on the age of language acquisition.
In cases of dissociated aphasia, the L2 is usually, but not always, the most
impaired language. For instance, Ardila (2008) reported the case of a 63-year-old
right-handed native Spanish speaker female who had been living in the US for
38 years. She never formally studied English, but after years of being exposed,
she learned basic English. Suddenly, she presented an extensive left temporal
intracerebral hemorrhage. A significant language understanding defect was
found, associated with severe impairments in verbal memory (i.e., Wernicke’s
aphasia), difficulties in language repetition, severe anomia with phonological and
semantic paraphasias and neologisms, alexia, and aphasic agraphia. The naming
defect was more severe in Spanish than in English; furthermore, there was also a
clear tendency to answer in English, to switch to English, and to mix English and
Spanish. The patient—a late bilingual with a relatively weak knowledge of L2—
consequently presented a dissociated aphasia with a better conservation of L2
(English) than L1 (Spanish).
Occasionally, bilinguals can present a different pattern of aphasia in L1 and L2.
For example, Silverberg and Gordon (1979) reported two cases of dissociated apha-
sia; following a left parietotemporal lesion, moderate nonfluent aphasia was found
in the native language of the first patient, in contrast to less severe, fluent aphasia in
the patient’s L2. Conversely, mild anomia was found in L1 of a second patient,
while global aphasia was found in L2. The lesion was located in the left posterior
frontal area.
Patterns of Recovery
Two opposite points of view were proposed during the nineteenth century to explain
the language recovery in bilingual aphasics.
• Ribot’s law or Ribot’s rule (Ribot, 1882) states that the language best recovered
by polyglot aphasics is the mother language.
• Pitres’ law or Pitres’ rule (Pitres, 1895). Pitres described seven cases of bilingual
aphasics presenting differential recovery of the two languages. He suggested that
patients tended to better recover the language that was most familiar to them
prior to the aphasia onset, regardless of whether it was or was not the mother
tongue.
214 A. Ardila
Both patterns of language recovery have been shown to be partially valid.

Supporting one or the other proposed laws, Paradis (1977) refers to six different
patterns of aphasia recovery in bilinguals.
1. Differential. Each language is impaired separately and recovered at the same or
different rate
2. Parallel. Different languages are similarly impaired and restored at the same
rate.
3. Antagonistic. Recovery of one language progresses, while the other regresses.
4. Successive. One language does not show any recovery until the other has been
restored.
5. Selective. One language is not recovered at all.
6. Mixed. Both languages are used in some combinations.
However, most patients present the first (differential) or second (parallel) recov-
ery pattern. The other patterns are indeed unusual. Fabbro (1999) reports a parallel
recovery in about 40% of the cases; a better recovery of L1 in 32% of the patients,
and a better recovery of L2 in about 28% of the cases.
Alexia
It has been proposed that characteristics of alexia correlate with the idiosyncrasies
of writing systems (Ardila & Cuetos, 2016; Coltheart, 1982). The lexical organiza-
tion and processing strategies, characteristics of skilled readers, in different orthog-
raphies are affected by different developmental constraints in different writing
systems (Ziegler & Goswami, 2005). In bilingual speakers, alexia can often be
restricted to only one language (Kremin, Chomel-Guillaume, Ferrand, & Bakchine,
2000).
Alexias, however, have been studied mostly in Indo-European language writing
systems, and cross-linguistic analyses are scarce. Psycholinguistic models of alex-
ias have been developed in English and French, two languages with rather irregular
writing systems. In English, with a significant number of irregular words (words
that cannot be read using grapheme-to-phoneme correspondence rules and can only
be recognized as a whole), the existence of two different reading strategies or read-
ing routes (indirect and direct) is understandable. Developmental dyslexia has been
found to be more frequent in irregular writing systems, such as English or French,
than in regular orthographic systems, such as Italian (Paulesu et al., 2001). The
applicability of the double route reading models to regular (phonologic) writing
systems has been challenged (Karanth, 2003; Lukatela & Turvey, 1998).
Alexia can be dissociated across different writing systems (phonological vs.
logographic; e.g., Japanese Kana and Kanji), across two phonological systems (e.g.,
Latin and Cyrillic); and even within the same Latin writing system (e.g., English
and Spanish; Ardila, 2012; 2014). For Instance, Luria (1956) reported the case of a
French journalist who, after a brain condition, had difficulties reading and writing
in different languages (French, Russian, German, and Polish), but the defect was
most severe in French (irregular writing system) and least in Russian (regular writ-
ing system).
Reports about alexia in logographic writing systems (e.g., Chinese) are relatively
scarce. With the exception of some studies on the Japanese Kana and Kanji reading
systems, comparative research on alexias and agraphias in non-Indo-European lan-
guages has been limited (Sakurai, 2004; Yamada, Imai, & Ikebe, 1990). Pure alexia,
selectively impairing Kana (but not Kanji) reading, has been reported in cases of left
posterior occipital lobe damage (Sakurai, Ichikawa, & Mannen, 2001), similar to
the anatomy of pure alexia in other phonographic systems. Conversely, alexia with
agraphia in Korean Hanja (logographic), while preserving Hangul (phonographic)
reading and writing have been reported after a left posterior inferior temporal lobe
infarction (Kwon et al., 2002). Sakurai and colleagues (Sakurai, 2004) distinguished
two different types of pure alexia: pure alexia for Kanji (and Kana; fusiform type:
pure alexia for words) characterized by impairments of both whole-word reading, as
represented in Kanji reading, and letter identification; and pure alexia for Kana
(posterior occipital type: pure alexia for letters) in which letter identification is pri-
marily impaired. Thus, individuals using two different writing systems (e.g., ideo-
grams and phonograms as found in Japanese and Korean) may present a dissociated
alexia. Yamawakiet al. (2005) observed in a specific form of alexia that oral reading
of Kanji words significantly correlates with naming pictures in relation to their cor-
responding word. This suggests that naming objects and reading their logographic
Kanji words share common underlying mechanism.
These studies, as a whole, indicate that reading strategies and alexia characteris-
tics are under the influence of the idiosyncrasies of the individual reading systems
(Karanth, 2003) and, ultimately, alexia can have partially different manifestations in
the L1 and the L2 of a bilingual individual.
Developmental Dysphasia
The diagnosis of developmental language impairment or specific language impair-

ment (developmental dysphasia) in bilingual children represents a clinical chal-
lenge. As Bedore and Pena (2008) point out, children from bilingual backgrounds
are sometimes over-identified with developmental language impairment because of
the apparent slowness in developing language. At other times, bilingual children are
under-identified because it is assumed that learning a second language imposes a
normal delay in language acquisition.
Some authors have speculated that children with specific language impairment
present a disorder that affects some basic cognitive and perceptual processing
mechanisms, resulting not only in difficulties learning language, but also in more
extended cognitive difficulties (e.g., Kohnert & Windsor, 2004; Paradis, 2007,
2010). Alternatively, it has been suggested that these children have selective and
specific linguistic deficits affecting the normal acquisition of language (Rice, 2003;
216 A. Ardila
Wexler, 2003); consequently, the deficit is limited to the language domain. In any
case, the impairment involves language acquisition, and it can be assumed that the
defect will affect any language the child is exposed to. However, bilingual children
are rarely equivalently exposed to each language. Therefore, it is not surprising to
find that children present an uneven language performance in each one of their lan-
guages. Sometimes language performance is context-dependent (i.e., they have a
better academic L2 at school and a better everyday L1 at home).
Another question raised with children with specific language impairment refers
to the ability to handle two different languages. It has been assumed that it was
inconvenient for a child with difficulties in learning a language to be exposed to two
different languages, and frequently parents were advised to avoid bilingualism.
Recent research has supported the idea that children with specific language impair-
ment can handle two languages, even though in both the language impairment will
be manifested (Genesee, Paradis, & Crago, 2004; Paradis, 2007, 2010). Furthermore,
bilingualism not only has a diversity of personal and social advantages, but also
there is a social bilingualism in which the child is inevitably exposed to two differ-
ent languages (e.g., when living in a bilingual environment).
Developmental Dyslexia
Developmental dyslexia as a delay in the normal process of learning to read can

have specific manifestations in each of the languages in bilingual children (Durkin,
2000). These differences are related to the characteristics of the orthographic sys-
tem, particularly, its transparency or opacity. Learning to read is easier in a transpar-
ent orthography, such as Spanish, and harder in an opaque orthography, such as
English (Dulude, 2012).
Paulesu et al. (2001) found that Italian dyslexics, using a shallow (transparent)
orthography which facilitates reading, performed better on reading tasks than did
English and French dyslexics; noteworthy, English and French represent two lan-
guages with a quite opaque orthography. However, in Paulesu et al.’s (2001) study,
all dyslexics, regardless of the native language, were equally impaired depending on
their controls on reading and phonological tasks. The authors conclude that there is
a universal neurocognitive basis for dyslexia. Therefore, differences in reading per-
formance among dyslexics of different countries exist due to the differences in the
orthographic systems.
Significant differences in learning to read have been reported between languages
with transparent and opaque orthographies: learning to read in transparent lan-
guages is notoriously easier and quicker than learning to read in an opaque lan-
guage. In an interesting cross-linguistic study, Seymour, Aro, and Erskine (2003)
confirmed the strong effect orthography has on learning to read. Children in the first
year of reading from 14 European countries had to read aloud a list of words and a
list of nonwords. Children speaking transparent languages had a much better perfor-
mance than children speaking languages with opaque orthographies. For instance,
Finnish, Greek, Italian, and Spanish children properly read 98%, 92%, 95%, and
89% of the words, respectively. However, Danes read 71% of the words and 54% of
the nonwords and Scots were able to read only 34% of words and 29% of nonwords.
The results confirmed that children from a majority of European countries become
accurate and fluent in foundation level reading before the end of the first school
year; however, that is not true for languages with an opaque orthography such as
French, Portuguese, Danish, and, particularly, English.
In addition, children speaking languages with transparent orthographies are not
only more accurate in reading, but also faster. Consequently, when comparing
Spanish-speaking children to children of irregular writing systems (such as English
or French) during the initial years, Spanish-speaking children read sooner
(Caravolas, Lervåg, Defior, Seidlová Málková, & Hulme, 2013; Caravolas et al.,
2012; Serrano et al., 2011). In general, it is assumed that 1 year training is sufficient
to learn the basic reading rules in Spanish (Seymour et al., 2003), whereas in an
irregular orthography such as English, the time required to acquire a basic reading
(e.g., to read the newspaper) is notoriously longer. As Cuetos and Suárez-Coalla
(2009) have emphasized, children whose languages have irregular orthographies
must learn to pronounce larger units (morphemes, or whole words), to achieve the
correct pronunciation of some words. However, children whose languages have
transparent orthographies, such as Spanish, only need to learn to pronounce differ-
ent graphemes to read successfully.
The rate of reading development in English is more than twice as slow as in regu-
lar orthographies. In other words, to obtain the reading level that a Spanish-speaking
child achieves in less than a year, an English-speaking child would require about 3
years. In consequence, the reading threshold (i.e., when an individual can be con-
sidered as literate) is lower in Spanish than in English, simply because there are no
reading irregularities. People with just one to 3 years of education could be consid-
ered readers of Spanish. These people would be regarded as illiterate or semi-
illiterate in English because one to 3 years may be insufficient to learn to read in
English.
Although reports of differing types and difficulties in the acquisition of two or
more scripts among developmental dyslexics (dissociated dyslexia) are relatively
rare, Karanth (1992) describes two such cases of developmental dyslexia, in whom
learning to read English as compared to Kannada and Hindi (two Indian scripts)
were differentially affected. Kannada and Hindi are regular (transparent) writing
systems, whereas English represents a quite irregular (opaque) writing system; it is
not surprising that the difficulties in learning to read were considerably more sig-
nificant in English than in the two Indian languages. A similar situation is frequently
found with Spanish–English bilingual children in the US: abnormal difficulties in
learning to read (dyslexia) are notoriously more severe in English than in Spanish.
Furthermore, Valdois et al. (2014) reported the case study of a French–Spanish
bilingual dyslexic girl who exhibited a severe visual attention span deficit but pre-
served phonological skills. Behavioral investigation showed a severe reduction of
reading speed for both single items (words and pseudo-words) and texts in the two
218 A. Ardila
languages. However, her performance was more affected in French (opaque orthog-
raphy) than in Spanish (transparent orthography).
Differences in dyslexia manifestations (dissociated dyslexia) have also been
reported when comparing different writing systems. Wydell and Butterworth (1999)
reported the case of a 16-year-old English–Japanese bilingual boy whose reading/
writing difficulties were confined to English only. The patient was born in Japan to
a highly literate Australian father and English mother and attended a Japanese senior
high school in Japan. His reading in logographic Japanese Kanji and syllabic Kana
was equivalent to that of Japanese undergraduates or graduates. In contrast, his per-
formance in various reading and writing tests in English, as well as tasks involving
phonological processing was very poor, even when compared to his Japanese con-
temporaries. Chung and Ho (2010) examined the relations between reading-related
cognitive skills and word reading development of Chinese dyslexic children in their
Chinese language (L1) and in English (L2). A total of 84 bilingual children, where
28 were dyslexic, 28 were chronological age (CA) controls and 28 were reading
level (RL) controls participated. They were administered measures of word reading,
rapid naming, visual-orthographic skills, phonological and morphological aware-
ness in both L1 and L2. Children with dyslexia showed weaker performance than
CA controls in both languages, and had/demonstrated difficulties in phonological
awareness in English but not in Chinese.
Dementia
Communication abilities in bilingual demented patients and language decline for L1

and L2 in dementia are questions not frequently mentioned in the dementia litera-
ture. It is well known, however, that the ability to maintain fluency in more than one
language decreases with aging (Hyltenstam & Obler, 1989; see also Terrazas-
Carrillo, this volume). With aging, people may tend to retreat to a single language,
regardless of a life long history of bilingualism. Additionally, the L2 is frequently
associated with active working life, and retirement is often associated with moving
to a more limited familiar environment. Moreover, older bilinguals may have
increased difficulties handling two different languages due to the effects of cross-
language interference. These effects in aging bilinguals could be further accentu-
ated in those who develop dementia.
In general, it is assumed that in bilingual individuals, regression to a primary
language may be associated with the development of cognitive impairment and
increased risk for the development of dementia (McMurtray, Saito, & Nakamoto,
2009). Frequently, difficulties in using the L2 represent an early sign of cognitive
decline. Using a sample of elders (ages 69–91 years), Salvatierra, Rosselli, Acevedo,
and Duara (2007) observed that cognitively normal participants retrieved signifi-
cantly more items under a semantic condition compared to the phonemic one,
whereas the performance of patients with Alzheimer’s disease was similar under
both conditions; this observation suggests a greater decline in semantic verbal
uency tests. Interestingly, this pattern was found in both languages, implying a
fl
related semantic decline in L1 and L2.
Mendez, Perryman, and Pontón (1999) studied 51 patients who consistently used
another language and were fluent in English. All patients were regularly exposed to
English as an L2 after the age of 13. Despite patients’ differences in educational
level, their age of acquisition of English, frequency of use, and baseline fluency in
English, all caregivers reported a greater preference of the patients for their original
language and decreased conversation in English. Patients exhibited a tendency for
words and phrases from their native language to intrude into English conversational
speech. The authors found that bilingual dementia patients tended to present asym-
metrical language impairment with preferential preservation and use of the L1.
They suggested that in dementia, recently learned information is retained the least;
and older, more remote information is often relatively preserved, consistent with a
regression toward the predominant use of the patient’s earliest language. According
to Mendez and colleagues, in dementia, a retreat to the original language could
result from an exacerbation of the cross-language difficulties that typically increase
with age. People who are bilingual never totally deactivate either of their two lan-
guages, and this can result in interference or intrusions, particularly from the domi-
nant language into the other one. Dementia patients tend to mix languages, and they
have special problems with language selection.
De Santi, Obler, Sabo-Abrahamson, and Goldberger (1990) observed in a group
of patients with Alzheimer’s disease that the language problems may be exhibited
differentially in each of the demented patient’s languages. Problems with language
choice and code-switching did not necessarily occur in both languages for all
patients. Only one of the analyzed patients, the most demented one, exhibited prob-
lems with language choice and code-switching. In this particular sample, a strong
correlation between severity of dementia, problems of language choice and code-
switching was observed.
General cognitive functioning has been found to be higher in demented patients
if communication is carried out mostly in L1 than in L2. Ekman, Wahlin, Norberg,
and Winblad (1993); Ekman, Wahlin, Viitanen, Norberg, and Winblad (1994) for
instance, studied demented persons who were born in Finland and had immigrated
to Sweden (Finnish–Swedish bilinguals). They found that many of these Finnish
immigrants had difficulties communicating with their Swedish-speaking caregivers,
while their communication with a Finnish-speaking caregiver was adequate. The
demented Finnish immigrants functioned on a level of manifest competence that
seemed far below their level of latent competence. The authors concluded that the
presence of Finnish-speaking caregivers is an environmental change which would
markedly enhance the demented Finnish immigrants’ performance and quality of
life as well as reduce the costs for their care.
Mendez, Saghafi, and Clark (2004) studied two polyglot patients with semantic
dementia. The first case was a 71-year-old man who experienced a slow, progressive
loss of his ability to use and understand Spanish and German. The patient was a
language teacher who was fluent in Spanish and used it daily for work. Confrontational
naming in English was decreased. The patient had great difficulty understanding,
220 A. Ardila
including common nouns in Spanish, and was no longer able to understand any
German. On an aphasia test battery, word comprehension was moderately impaired
in English, but severely impaired in Spanish and German. Words that were compre-
hended in Spanish or German were not consistently comprehended in English. The
magnetic resonance imaging (MRI) studies showed anterior temporal atrophy, left
greater than right.
The second case was a 66-year-old man who had a 2-year history of progressive
loss of the meaning of words and inability to retrieve words. Although Spanish was
his L1, he spoke English at work and knew some Polish as well. His examination
was intact except for naming and recognizing famous faces. His confrontational
naming in Spanish was impaired. He could not name pictures of items and made
some semantic errors (e.g., zero for circle). His performance was worse in English
than in Spanish, and his Polish was lost. If he comprehended a word in one lan-
guage, he did not necessarily comprehend it in the other. His MRI results showed
left anterior temporal atrophy. The authors concluded that in multilingual patients
with semantic dementia, semantic anomia was progressively more impaired in their
second and third languages compared to their primary language. Words named and
comprehended in one language were not consistently named and comprehended in
other languages they knew. These findings were interpreted as compatible with sep-
arate lexical semantic systems for each language.
The difficulty in selecting the appropriate language is observed in aging bilin-
guals, but it becomes more severe in cases of dementia. It has been suggested that
bilingual speakers with dementia, even in the early stages of deterioration, make
errors in selecting the appropriate language and maintaining the correct language
during conversational speech (De Santi et al., 1990; Hyltenstam & Stroud, 1989).
There is, however, large variability in the extent of inappropriate language use,
where some individuals mix languages more than others (Friedland & Miller,
1999). Hyltenstam and Stroud (1989) described two cases of Alzheimer‘s disease
in bilinguals. In one of them, the major problem laid mainly in the area of language
choice, whereas in the other the major difficulties were observed in the ability to
separate the languages. De Santi et al. (1990) concluded that the ability to make the
correct language choice and keep languages separated was correlated with the over-
all stage of dementia. The mixture of languages might become so predominant that
it may be challenging to recognize what language the patient is attempting to speak.
Finally, it is interesting to note that normal bilinguals can use the knowledge of
two languages to increase verbal production, whereas dementia patients are unable
to profit from the knowledge of two different languages. De Picciotto and Friedland
(2001) studied verbal fluency abilities in 30 normal aging English–Afrikaans bilin-
gual speakers and six bilingual subjects with Alzheimer’s disease. A semantic ver-
bal fluency task (animals) was administered in the bilingual mode, Afrikaans and
English. There was no significant difference between monolingual and bilingual
performance. It was observed that some normal bilingual subjects used code-
switching as a strategy; there was not, however, a relationship between age of acqui-
sition, pattern of use and verbal fluency scores. In comparison, subjects with
Alzheimer’s disease did not make use of code-switching strategies, but there was
some relationship between age of acquisition, pattern of use, and verbal fluency
scores. Ultimately, it was concluded that normal bilinguals can recur to both lan-
guages in attempt to improve performance, while demented patients were unable to
use this strategy.
Conclusions
In bilingual individuals, neuropsychological syndromes involving language can

have different clinical manifestations and different degree of severity in each one of
their languages. In aphasia, the major variable affecting the aphasia profile and
severity of L1 and L2 is the age of acquisition of the L2 and, consequently, how
much distance there is in the brain organization of each of the languages. Specific
language impairment in children (i.e., developmental dysphasia) tends to similarly
affect both languages, but in diverse conditions—such as the specific use of each
one of the languages—can impact the manifestation of the language difficulty.
Although there are some general principles in the neurological organization of
written language (i.e., certain specific brain areas support written language, regard-
less of the particular idiosyncrasies of the system), alexia does not seem to be com-
pletely coincidental across different orthographies (e.g., English and Spanish) and
writing systems (i.e., sonographic systems such as English or Spanish that represent
sounds; or logographic systems such as Chinese or Japanese Kanji that represent
meanings). In some cases, a dissociated alexia can be found. The frequency and
severity of developmental dyslexia is significantly associated with the transpar-
ency—opacity of the orthographic system; dyslexia is much more frequent and
severe in opaque (irregular) orthographies. Moreover, dyslexia could eventually be
dissociated between L1 and L2 in bilingual subjects.
In these individuals, aging is associated with a decreased language ability
affecting very specially L2. Furthermore, regression to a primary language may be
associated with development of cognitive impairment and increased risk for devel-
opment of dementia. Bilingual patients with dementia usually have difficulties
making the correct language choice and keep languages separated. It has been
observed that, including caregivers, speaking patient’s L1 language can be signifi-
cantly beneficial because it can enhance the general cognitive performance and
raise the quality of life.
References
Albert, M. L., & Obler, L. K. (1978). The bilingual brain: Neuropsychological and neurolin-
guisitic aspects of bilingualism (pp. 95–130). New York: Academic Press.
222 A. Ardila
Ardila, A. (2008). Anomia disociada en un bilingüe tardío con mejor conservación de L2 [dis-
sociated anomia in a late bilingual with a better conservation of L2]. Neuropsicología,
Neuropsiquiatría y Neurociencias, 8(2), 91–95.
Ardila, A. (2012). Neuropsychology of writing. In E. Grigorenko, E. Mambrino, & D. Preiss
(Eds.), Handbook of writing: A mosaic of perspectives and views (pp. 309–321). London:
Psychology Press.
Ardila, A. (2014). Aphasia handbook. Miami, FL: Florida International University.
Ardila, A., & Cuetos, F. (2016). Applicability of dual-route reading models to Spanish. Psicothema,
28(1), 71–75.
Bedore, L. M., & Pena, E. D. (2008). Assessment of bilingual children for identification of lan-
guage impairment: Current findings and implications for practice. International Journal of
Bilingual Education and Bilingualism, 11(1), 1–29.
Benson, D. F., & Ardila, A. (1996). Aphasia: A clinical perspective. Oxford: New York.
Caravolas, M., Lervåg, A., Defior, S., Seidlová Málková, G., & Hulme, C. (2013). Different pat-
terns, but equivalent predictors, of growth in reading in consistent and inconsistent orthogra-
phies. Psychological Science, 24, 1398e1407.
Caravolas, M., Lervåg, A., Mousikou, P., Efrim, C., Litavský, M., Onochie-Quintanilla, E., et al.
(2012). Common patterns of prediction of literacy development in different alphabetic orthog-
raphies. Psychological Science, 23(6), 678–686.
Chung, K. K. H., & Ho, C. S. H. (2010). Second language learning difficulties in Chinese chil-
dren with dyslexia: What are the reading-related cognitive skills that contribute to English and
Chinese word reading? Journal of Learning Disabilities. 43(3):195-211.
Coltheart, M. (1982). The psycholinguistic analysis of acquired dyslexia: Some illustrations.
Philosophical Transactions of the Royal Society B: Biological Sciences., 298(1089), 151–164.
Cuetos, F., & Suárez-Coalla, P. (2009). From grapheme to word in reading acquisition in Spanish.
Applied PsychoLinguistics, 30(4), 583–601.
De Picciotto, J., & Friedland, D. (2001). Verbal fluency in elderly bilingual speakers: Normative
data and preliminary application to Alzheimer’s disease. Folia Phoniatrica et Logopaedica,
53, 145–152.
De Santi, S., Obler, L. K., Sabo-Abrahamson, H., & Goldberger, J. (1990). Discourse abilities and
deficits in multilingual dementia. In Y. Joanette & H. H. Brownell (Eds.), Discourse ability and
brain damage: Theoretical and empirical perspectives (pp. 224–235). New York: Springer.
Dulude, L. (2012). Writing system, phonemic awareness, and bilingualism: Cross-linguistic issues
in dyslexia. Indiana University Undergraduate Journal of Cognitive Science, 7, 22–30.
Durkin, C. (2000). Dyslexia and bilingual children—Does recent research assist identification?
Dyslexia, 6(4), 248–267.
Ekman, S. L., Wahlin, T. B., Norberg, A., & Winblad, B. (1993). Relationship between bilingual
demented immigrants and bilingual/monolingual caregivers. International Journal of Aging
and Human Development, 37, 37–54.
Ekman, S. L., Wahlin, T. B., Viitanen, M., Norberg, A., & Winblad, B. (1994). Preconditions for
communication in the care of bilingual demented persons. International Psychogeriatrics, 6,
105–120.
Fabbro, F. (1999). The neurolinguistics of bilingualism: An introduction. New York: Psychology
Press.
Fabbro, F. (2001). The bilingual brain: Bilingual aphasia. Brain and Language, 79, 201–210.
Friedland, D., & Miller, N. (1999). Language mixing in bilingual speakers with Alzheimer‘s
dementia: A conversation analysis approach. Aphasiology, 13(4–5), 427–444.
Genesee, F., Paradis, J., & Crago, M. (2004). Dual language development and disorders: A hand-
book on bilingualism and second language learning. Baltimore, MD: Brookes.
Goral, M., Levy, E. S., & Obler, L. K. (2002). Neurolinguistic aspects of bilingualism. International
Journal of Bilingualism, 6(4), 411–440.
Green, D. W. (2008). Bilingual aphasia: Adapted language networks and their control. Annual
Review of Applied Linguistics, 28, 25–48.
Hyltenstam, K., & Obler, L. K. (1989). Bilingualism across the lifespan: Aspects of acquisition,
maturity and loss. Cambridge University Press.
Hyltenstam, K., & Stroud, C. (1989). Bilingualism in Alzheimer‘s dementia : Two case studies. In
K. Hyltenstam & L. K. Obler (Eds.), Bilingualism across the lifespan: Aspects of acquisition,
maturity and loss (pp. 202–226). Cambridge: Cambridge University Press.
Karanth, P. (1992). Developmental dyslexia in bilingual-biliterates. Reading and Writing, 4(3),
297–306.
Karanth, P. (2003). Cross-linguistic study of acquired reading disorders: Implications for reading
models, disorders, acquisition, and teaching (Vol. 24). New York: Springer Science & Business
Media.
Kim, K., Rilkin, R., Lee, K., & Kirsch, J. (1997). Distinct cortical areas associated with native and
second language. Nature, 388, 171–174.
Kohnert, K., & Windsor, J. (2004). The search for common ground: Part II. Nonlinguistic perfor-
mance by linguistically diverse learners. Journal of Speech Language and Hearing Research,
47, 891–903.
Kremin, H., Chomel-Guillaume, S., Ferrand, I., & Bakchine, S. (2000). Dissociation of reading
strategies: Letter-by-letter reading in the native language and normal reading in the learned
language. A case study. Brain and Cognition, 43, 282–286.
Kwon, J. C., Lee, H. J., Chin, J., Lee, Y. M., Kim, H., & Na, D. L. (2002). Hanja alexia with
agraphia after left posterior inferior temporal lobe infarction: A case study. Journal of Korean
Medical Sciences, 17, 91–95.
Lukatela, G., & Turvey, M. T. (1998). Reading in two alphabets. American Psychologist, 53,
1057–1072.
Luria, A. R. (1956). On the problem of reading and writing disturbance in polyglots. Fiziologicheskyj
zhurnal, 2:173–177.
McMurtray, A., Saito, E., & Nakamoto, B. (2009). Language preference and development of
dementia among bilingual individuals. Hawaii Medical Journal, 68(9), 223.
Mendez, M.F., Perryman, K.M., Pontón, M.O. & Cummings, J.L (1999). Bilingualism and demen-
tia. Journal of Neuropsychiatry and Clinical Neurosciences, 11, 411–412.
Mendez, M. F., Saghafi, S., & Clark, D. G. (2004). Semantic dementia in multilingual patients.
Journal of Neuropsychiatry and Clinical Neurosciences, 16, 381.
Obler, L. K., Hyun, J., Conner, P., O’Connor, B., & Anema, I. (2007). Brain organization of lan-
guage in bilinguals. In A. Ardila & E. Ramos (Eds.), Speech and language disorders in bilin-
guals (pp. 21–46). New York: Nova Science International.
Paradis, J. (2007). Bilingual children with specific language impairment: Theoretical and applied
issues. Applied PsychoLinguistics, 28, 551–564.
Paradis, J. (2010). The interface between bilingual development and specific language impairment.
Applied PsychoLinguistics, 31, 227–252.
Paradis, M. (1977). Bilingualism and aphasia. Studies in neurolinguistics, 3, 65–121.
Paulesu, E., Demonet, J. F., Fazio, F., McCrory, E., Chanoine, V., Brunswick, N., et al. (2001).
Dyslexia: Cultural diversity and biological unity. Science, 291, 2064–2065.
Pitres, A. (1895). Etude sur l’aphasie chez les polyglottes. Revue de médecine, 15, 873–899.
Ribot, T. (1882). Les maladies de la mémoire. G. Bailliere.
Rice, M. (2003). A unified model of specific and general language delay: Grammatical tense as
a clinical marker of unexpected variation. In Y. Levy & J. Schaeffer (Eds.), Language compe-
tence across populations: Towards a definition of specific language impairment (pp. 63–94).
Mahwah, NJ: Erlbaum.
Sakurai, Y. (2004). Varieties of alexia from fusiform, posterior inferior temporal and posterior
occipital gyrus lesions. Behavioural Neurology, 15, 35–50.
Sakurai, Y., Asami, M., & Mannen, T. (2010). Alexia and agraphia with lesions of the angular
and supramarginal gyri: Evidence for the disruption of sequential processing. Journal of
Neurological Sciences, 288(1–2), 25–33.
224 A. Ardila
Sakurai, Y., Ichikawa, Y., & Mannen, T. (2001). Pure alexia form a posterior occipital lesion.
Neurology, 56, 778–781.
Salvatierra, J., Rosselli, M., Acevedo, A., & Duara, R. (2007). Verbal fluency in bilingual Spanish/
English Alzheimer’s disease patients. American Journal of Alzheimer’s Disease and Other
Dementias, 22(3), 190–201.
Serrano, F., Genard, N., Sucena, A., Defior, S., Alegria, J., Mousty, P., et al. (2011). Variations in
reading and spelling acquisition in Portuguese, French and Spanish: A cross-linguistic com-
parison. Journal of Portuguese Linguistics, 10(1), 183–204.
Seymour, P. H. K., Aro, M., & Erskine, J. M. (2003). Foundation literacy acquisition in European
orthographies. British Journal of Psychology, 94(2), 143–174.
Silverberg, R., & Gordon, H. W. (1979). Differential aphasia in two bilingual individuals.
Neurology, 29(1), 51–55.
Valdois, S., Peyrin, C., Lassus-Sangosse, D., Lallier, M., Démonet, J. F., & Kandel, S. (2014).
Dyslexia in a French–Spanish bilingual girl: Behavioural and neural modulations following a
visual attention span intervention. Cortex, 53, 120–145.
Wexler, K. (2003). Lennenberg’s dream: Learning, normal language development, and specific lan-
guage impairment. In Y. Levy & J. Schaeffer (Eds.), Language competence across populations.
Towards a definition of specific language impairment (pp. 11–62). Mahwah, NJ: Erlbaum.
Wydell, T. N., & Butterworth, B. (1999). A case study of an English-Japanese bilingual with
monolingual dyslexia. Cognition, 70(3), 273–305.
Yamada, J., Imai, H., & Ikebe, Y. (1990). The use of the orthographic lexicon in reading kana
words. Journal of General Psychology, 117, 311–323.
Yamawaki, R., Suzuki, K., Tanji, K., Fujii, T., Endo, K., Meguro, K., et al. (2005). Anomic alexia
of kanji in a patient with anomic aphasia. Cortex, 41, 555–559.
Ziegler, J. C., & Goswami, U. (2005). Reading acquisition, developmental dyslexia, and skilled
reading across languages: A psycholinguistic grain size theory. Psychological Bulletin, 131,
3–29.
Part IV
Bilingual Cognitive and Personality
Dimensions
Chapter 10
Psychopathology and Bilingualism
Elizabeth Terrazas-Carrillo
Contents
Introduction�� 227
Language and Bilingualism�� 228
Dynamics of Language and Psychopathology�� 229
Cultural Idioms and Psychopathology�� 229
Personality Disorders and Bilingualism�� 232
Externalizing Psychopathology and Bilingualism�� 236
Developmental Disorders in Bilingual Children�� 238
Schizophrenia�� 240
Theories Explaining Differential Symptomology in Bilingual Schizophrenics�� 244
Bilingualism as a Barrier to Diagnosis and Treatment of Psychopathology�� 245
Clinician Bias�� 245
Emotional Expression Across Languages�� 246
Personality Assessment Issues�� 247
Conclusion�� 247
Clinical Recommendations�� 248
References�� 250
Introduction
It has been estimated that about half of the world’s population speaks a second lan-
guage (L2; Grosjean, 2010). In the United States alone, bilinguals make up about
20% of the population, with even higher percentages in urban areas such as Los
Angeles and Toronto (Ryan, 2013; Statistics Canada, 2011). Higher percentages of
bilinguals are estimated in Europe, where at least 50% of respondents to a survey
E. Terrazas-Carrillo, Ph.D. (*)

e-mail: elizabeth.terrazas@tamiu.edu

228 E. Terrazas-Carrillo
from the European Commission identified as bilingual, and over 90% identified as
bilingual in countries such as Luxemburg and Latvia (European Commission, 2006;
Nardelli, 2014). Bilingualism presents a fascinating opportunity to understand the
impact of language on psychopathology. Psychopathology encompasses the study
of psychological dysfunction within an individual that is associated with distress or
impairment in functioning and a response that is not typical or culturally expected
(Barlow & Durand, 2015, p. 2). The term psychological dysfunction indicates an
interruption or deterioration in the domains of cognitive (i.e., thought processes,
memory), emotional, or behavioral functioning. These domains are impacted by
acquisition of an L2. Therefore, it can be asserted that psychopathology as a form of
psychological dysfunction bears a relationship to the same domains impacted by L2
learning (Javier, 2007; Paradis, 2008). Language is an integral part of assessing and
diagnosing psychopathology; thus, bilingualism can alter the way in which psycho-
pathology-related symptoms are expressed across languages. This potential change
in symptom expression can have a significant impact on the accurate diagnosis of
psychopathological conditions, leading to faulty and unintentionally biased psychi-
atric diagnosis. Faulty psychiatric diagnosis, in turn, is a barrier to accessing timely
and appropriate psychological treatment. Therefore, acquiring knowledge about the
potential ways in which speaking two languages may impact the presentation of
various forms of psychopathology is essential for mental health practitioners work-
ing with bilingual populations.
It has been argued that monolinguals and bilinguals function differently in
regards to thinking and emotional processes, memory functions and accessibility of
information (Centeno & Obler, 2001; Chen, Benet-Martinez, & Ng, 2014; Hull,
2003). Although there has been a renewed interest in the context-dependent nature
of cognitive and emotional processes activated by dual language fluency, the
research exploring the impact of bilingualism on the development and expression of
psychopathology has been limited and mostly reliant on case studies and reports.
This chapter summarizes the literature available to date exploring psychopathology
and bilingualism, and points to future research directions.
Language and Bilingualism
Language acquisition is a process that starts in childhood and continues through-

out life. Language serves as a vehicle to developing the capacity to organize per-
ception and experiences into abstract concepts (Mahler, Pine, & Bergman, 1975;
Sullivan, 1953). These experiences become internalized, creating patterns of
organization that serve as the linguistic foundation of an individual’s cognitive
understanding of reality throughout life (Piaget, 1980; Vygotsky, 1962). For
example, Bucci (1997) proposed that language can determine, mediate, and pro-
duce specific thoughts. However, for bilinguals there are two languages that could
potentially organize perceptions of reality. Although there has been controversy
regarding the idea of the bilingual mind possessing two independent mental
10 Psychopathology and Bilingualism 229
representations, there is the possibility that linguistic processing among bilinguals

might be different from that of monolinguals (see for example Javier, 2007).
These findings are relevant to understanding the expression of psychopathology
among bilinguals, as integrative psychopathological models acknowledge that
domains of cognitive and emotional processes do not operate in isolation, but
rather, influence each other in complex ways to create psychological dysfunction
(Barlow & Durand, 2015). However, the fields of psychopathology and bilingual-
ism intersect in (1) the exploration of the dynamics of language as it impacts
expression of psychopathology and (2) the effect of language as a barrier to accu-
rate diagnosis and treatment of psychopathology.
Dynamics of Language and Psychopathology
Psychological dysfunction emphasizes a breakdown in functioning in cognitive,

behavioral, and emotional domains (Barlow & Durand, 2015). The direct implica-
tion of the bilingual literature for psychopathology is the potentially different set of
cognitive, behavioral, and emotional manifestation of symptoms in the first lan-
guage (L1) and L2. Westernmeyer and Janca (1997) indicate that the traditional
clinical interview is the main way of gathering information about an individual’s
presenting symptoms and consists of asking the client a standardized set of ques-
tions about current mental state and presenting concern, as well as social, emo-
tional, developmental, and medical history. Paradis (2008) suggests the clinical
interview is subjective and limited; however, he indicates there are no better alterna-
tives available to accurately and objectively diagnose psychopathology. Thus, in a
traditional clinical evaluation, language is the main vehicle for communicating
symptomology, along with nonverbal and interpersonal behaviors (Westernmeyer &
Janca, 1997). However, languages themselves differ in the kinds of words and idi-
oms used to express psychological experiences, which can impact crucial variables
in psychiatric diagnosis among bilinguals such as emotion, affect, thought pro-
cesses, self-concept, and behavioral tendencies (Leff, 1981; Levy, 1983; Paradis,
2008). Although it is possible to translate common words and idioms across lan-
guages, it is also important to understand that culture has a direct impact on the
manifestation of emotional experiences; therefore, culture directly impacts the lan-
guage used to describe them (Steiner, 1992).
Cultural Idioms and Psychopathology
Cultural idioms of distress are encoded in an individual’s culture and language and
they serve to provide explanations about the nature of their illness and suffering
(Kleinman, 1988). While concrete words can be easily translated, more subjective
and abstract words used to describe emotional states such as sadness, anger, and
anxiety are not (Santiago-Rivera & Altarriba, 2002). For example, being sad in the
context of Western English usually involves disappointment and affliction that
denotes isolation, helplessness, hopelessness, and frustration that affect a person at
the individual level. Mesquita and Ellsworth (2001) report that for English speakers
in the United States, sadness implies a violation of deeply entrenched Western val-
ues such as having control, autonomy, and the right of individuals to pursue happi-
ness. However, Postert et al. (2012) argue that for the Hmong peoples of China and
Laos, sadness connotes interactional passivity and resignation leading to withdrawal
from social life. When a Hmong is feeling sad it directly impacts family and social
connections and positively motivates family members to mobilize resources to rein-
corporate the afflicted individual to group life (Postert et al., 2012). Because of
cultural differences between individualistic and collectivistic values, the same word
conveys slightly different affective states and responses from those around the sad
individual (Mesquita & Ellsworth, 2001).
Another example conveying the importance of understanding cultural idioms of dis-
tress comes from studies exploring prevalence of depression among South Asians in the
United Kingdom, which concluded that as a group they exhibit lower rates of depression
compared to their English counterparts (Bhui, 1999). These studies used measures of
depression in English, which rely on Western idioms of psychopathology (e.g., feeling
sad, hopeless, or helpless) because most South Asians living in the U.K. are fluent in
English (Rait, 1999; Williams and Hunt, 1997). However, Bhui, Bhugra, Goldberg,
Sauer, and Tylee (2004) found significantly higher rates of depression among South
Asians when they used a measure of depression written in Punjabi and containing com-
mon linguistic idioms used to describe symptoms of depression within this ethnic group.
The researchers noted that in Punjabi, South Asians expressed social and emotional
distress using cultural idioms such as having a sinking heart, burning sensations in
hands and feet, lack of energy, and lack of affection towards friends (Bhui et al., 2004).
Similarly, researchers found that Spanish-speaking Latinos in the United States are
less likely to be diagnosed with depression by their English-speaking physicians, even
though rates of depression among Latinos can be as high as 33% in primary care set-
tings (Gross et al., 2005). Although there is a direct translation of the word depression
from English to Spanish, Cabassa et al. (2008) found that Latinos are more likely to
allude to depression symptomology by using the cultural idiom of nervios (or nerves),
which is used to denote insomnia, restlessness, headaches, stress, and feeling dys-
phoric, and tired (Cabassa et al., 2008; Pincay & Guarnaccia, 2007). Similar cross-
cultural and linguistic differences have been observed among other cultural and ethnic
groups. For example, rather than verbally expressing depression as feeling sad,
research shows that depressed Chinese individuals are more likely to report feeling
uncomfortable inside my heart, as well as indecisiveness, guilt, and anhedonia accom-
panied by fear and actively thinking about death (Yanping, Leyi, and Qijie, 1986,
p. 240). Researchers noted that only 3 (e.g., indecisiveness, guilt, hopelessness) out of
16 key words (e.g., depressed, agitated, fearful, anxious, self-pity, suicidal interest,
being punished, loss of interest, failure, loss of weight, poor appetite, loss of sexual
drive, and tired) used in the Western-developed scales of depression such as the
Chinese translation of the Beck Depression Inventory, the Carroll Depression Rating
Scale, the Zung Depression and Anxiety Rating Scales, and the Hamilton Depression
and Anxiety Rating scales, were commonly used by the Chinese clients in the study
(Yanping et al., 1986). Cultural idioms about mental illness go beyond language dif-
ferences; they reflect cultural constructs of illness embedded in the language. This
popular nosology is informed by cultural knowledge and is used in the native lan-
guage to articulate and express illness (Guarnaccia, Lewis-Fernandez, & Marano,
2003). Therefore, cross-cultural and cross-linguistic scales to measure depression and
other psychiatric disorders should draw from indigenous words and phrases com-
monly used as idioms of distress (Kleinman & Good, 1985). For example, Chinese
researchers recommend using Chinese cultural idioms such as feeling foolish,
unhappy, or feeling uncomfortable inside the heart (Yanping et al., 1986), and Pincay
and Guarnaccia (2007) recommended the assessment of social dimensions of depres-
sion among Latinos, especially isolation (soledad) as Latinos tend to value social
relationships.
Studying patterns of language use has been presented as a novel methodology
that can provide insights into an individual’s psychological processes (Simsek &
Cerci, 2013). According to Pyszcynski and Greenberg (1987), an intense self-focus
is indicative of lack of social integration, which may in turn be a vulnerability to
depression because individuals focus on their roles on situations, as well as what
those events and situations say about them, thus internalizing their thoughts and
feelings. An early study testing this idea was developed by Bucci and Freedman
(1981), whereby they compared data obtained from English-speaking monolingual
inpatients diagnosed with severe depression, healthy adolescent, undergraduate,
and elderly individuals on measures including a name-retrieval score from the
Stroop Color-Word test, spontaneous speech measured by a 5-mi monologue, and a
quantitative depression score from the Zung Depression Status Inventory and found
that individuals diagnosed with depression tend to overwhelmingly use first person
singular pronouns (e.g., I) compared to nondepressed individuals (Bucci &
Freedman, 1981). These findings were replicated by Rude, Gortner, and Pennebaker
(2004), who studied the narratives of depressed and nondepressed students who
sought out treatment at a university counseling center. Rude et al. (2004) asked
participants to complete the Beck Depression Inventory and then asked them to
write an essay describing their deepest thoughts and feelings about being in college
(p. 1126). This essay was then entered into a computer-based text analysis program,
the Linguistic Inquiry and Word Count (LIWC; Pennebaker, Francis, & Booth,
2001), which provides information on linguistic dimensions such as number of arti-
cles used, number of words per sentence, as well as psychological processes such as
frequency of emotional and cognitive words, and personal concerns addressed in the
text in the form of a percentage of words found for each category (Rude et al., 2004;
Pennebaker et al., 2001). Rude et al. (2004) found that depressed students used
almost exclusively the first person singular pronoun I. Using the same computer-
based text analysis program, LIWIC, Stirman and Pennebaker (2001) entered
approximately 300 poems and literary writings of suicidal and nonsuicidal poets
and found a significantly higher use of first person singular pronouns (e.g., I, me).
This body of research argues that the lack of second and third pronouns in these
individuals’ narratives is evidence of their inability to connect with other people in

their lives, which is a hallmark of the isolation characteristic of depression
(Pennebaker, Mehl, & Niederhoffer, 2003).
Interestingly, Kashima and Kashima (1998) concluded that a language’s number
of personal pronouns used reflected different conceptions of the self and self–other
relationships. Although language use methodology has not been implemented with
bilingual populations to explore patterns of depressive symptomology, a study by
Marian and Kaushanskaya (2004) found that Russian–English bilinguals switched
pronoun use when narrating personal stories in English and Russian. Specifically,
Marian and Kaushanskaya found significant differences in the use of first person
singular and plural pronouns in their Russian and English narratives; English narra-
tives used more first person pronouns and Russian narratives used more group-
oriented pronouns. They further argued that these differences represented different
self-construals among bilinguals; the American culture values the individual above
the group, while Russian culture tends to place the group above the individual
(Marcus & Kitayama, 1991; Marian & Kaushanskaya, 2004). Considering the lit-
erature on linguistic patterns predictive of depression, it could be argued that a bilin-
gual would be able to shift between two self-construals and may exhibit symptoms
of depression in one language but not the other. However, this is only a hypothesis
as no research to date has explored this potential aspect of bilingualism and expres-
sion of depressive symptomology. Future research on the intersection of bilingual-
ism and depression may be advanced by using computer-based text analysis
programs like LIWC pioneered by Pennebaker et al. (2001).
Personality Disorders and Bilingualism
According to the Diagnostic and Statistical Manual of Mental Disorders 5th Edition
(DSM-5), a personality disorder is an enduring pattern of inner experience and behav-
ior that deviates markedly from the expectations of the individual’s culture, is pervasive
and inflexible, has an onset in adolescence or early adulthood, is stable over time, and
leads to distress or impairment (DSM-5, 2013, p. 647). This definition is deeply rooted
in individualistic values of Western cultures, as mental illness is perceived as an internal
state of the self, inherently independent from any social factors (Lewis-Fernandez &
Kleinman, 1994). Although the definition recognizes the role of culture, symptom
thresholds for each personality disorder are outlined and required to justify a diagnosis
(DSM-5, 2013). Researchers argue that personality disorders represent arbitrary thresh-
olds at which society decides a particular way of relating to the world is problematic
(Barlow and Durand, 2015; de Bernier, Kim, & Sen, 2014). In fact, some authors refer
to personality disorders as extreme, caricaturized versions of personality traits valued
by the Western culture (Alarcón & Foulks, 1995b; Fabrega, 1994). Alarcón and Foulks
(1995a) assert that defining or labeling deviances from normal personality is clearly a
culturally relative exercise, and its boundaries are reflective of society’s specific values,
ideals, worldview, resources, and structures (p. 7).
Support for a dimensional and cultural conception of personality has been found
in studies showing that both pathological and nonpathological personality traits exist
on a continuum (Bastiaansen, Rossi, Schotte, & De Fruyt, 2011; Endler & Kocovski,
2002; Livesley, Schroeder, Jackson, & Jang, 1994; Widiger and Simonsen, 2005).
While there is no general consensus about how many dimensions of personality
exist, the most widely accepted structure is the Five-Factor model or the Big Five
(Skodol et al., 2005; Hofstede & McCrae, 2004). Skodol et al. (2005) found that all
personality disorders were characterized by extreme neuroticism, impulse and emo-
tional dysregulation, and negative emotionality. In fact, high scores on all facets of
neuroticism predict global personality dysfunction associated with borderline, avoid-
ant, and dependent personality disorders, and to a lesser extent with schizotypal,
paranoid, and antisocial personality disorders among a variety of psychiatric samples
and populations (Bastiaansen, Rossi, Schotte, & De Fruyt, 2011; Fonseca-Pedrero,
Paino, Lemos-Giraldez, & Muñiz, 2013; Yang et al., 1999; 2000; 2002). However, it
is widely acknowledged that culture and language impact the way individuals per-
ceive and express a problem, therefore impacting the clinical understanding of per-
sonality disorders (Alarcón and Foulks, 1995a; Chavira et al., 2003; Marcos, Alpert,
Urcuyo, & Kesselman, 1973; Rogler, 1993).
Clinical assessment of personality disorders usually entails use of instruments
normed with clinical populations, such as the Minnesota Multiphasic Personality
Inventory-2 (MMPI-2), the Thematic Apperception Test (TAT), and the Personality
Assessment Inventory (PAI;Velásquez, Garrido, Castellanos, & Burton, 2004). The
MMPI-2 is one of the most widely used instruments to assess personality dysfunc-
tion and mental illness and contains ten clinical subscales and eight validity sub-
scales (Butcher et al., 2001). The clinical scales include: (1) hypochondriasis,
tapping complaints about bodily function, (2) depression, measuring clinical depres-
sion traits such as general dissatisfaction with life, (3) hysteria, measuring poor
physical health, cynicism and neuroticism, (4) psychopathic deviate, which mea-
sures general social maladjustment, (5) masculinity/femininity, measuring how
much a person conforms to stereotypical gender roles, (6) paranoia, tapping on
suspiciousness and interpersonal sensitivity, (7) psychasthenia, which measures a
person’s ability to resist thoughts or actions, (8) schizophrenia, which measures
bizarre thoughts, peculiar perceptions, and social alienation, (9) hypomania, mea-
suring excitement and elated mood and psychomotor agitation, and (10) social
introversion, which taps into a person’s level of comfort interacting in social situa-
tions (Velásquez et al., 2004). Validity scales provide information about how the
respondent approached the test and aid in qualifying clinical findings or interpreta-
tions from the test (Butcher et al., 2001). Validity scales include: (1) cannot say
scale, indicating number of unanswered items, (2) variable response inconsistency,
which measures the consistency of the respondent’s answers, (3) true response
inconsistency, which reports whether the respondent answered randomly or care-
lessly, (4) infrequency reports unusual or bizarre experiences, (5) lie, which identi-
fies respondents who are not answering honestly to make themselves look better, (6)
correction, which help identify defensive responding and poor insight (Butcher
et al., 2001). In spite of its widespread use in clinical settings, research indicates
bilinguals may view and describe their problems differently depending on the lan-
guage in which the MMPI-2 is administered (Sandoval & Duran, 1998; Velásquez,
Chavira, Karle, Callahan, Garcia, and Castellanos, 1997).
There is evidence indicating that Hispanics tend to display patterns of underre-
porting and defensiveness when assessed with the MMPI-2 compared to Whites,
specifically on the L scale used to assess defensiveness (Garrido, Velasquez, Parsons,
Reimann & Salazar, 2006). Velasquez et al. (2002) speculate this difference might
be the result of cultural influences such as the importance of keeping negative issues
from discussion outside the family, as well as a guarded attitude resulting from
socialization emphasizing caution towards majority culture institutions (Lucio,
Reyes-Lagunes, and Scott, 1994; Velásquez et al., 2004; Whitworth, 1988). Other
studies comparing MMPI-2 scores across Latinos from different countries of origin
found that Puerto Ricans generally scored higher on the subscale of schizophrenia
(Fontani-Salvador & Rogers, 1997). However, the schizophrenia scale of the
MMPI-2 contains items asking about strange life experiences, ghosts, and dreams,
which might be answered affirmatively by individuals who believe in folk healing
and other native Latino superstitions (Velasquez et al., 2002).
Research indicates bilinguals may view and describe their problems differently
depending on the language in which the MMPI-2 is administered (Sandoval &
Duran, 1998; Velásquez, Chavira, Karle, Callahan, Garcia, and Castellanos, 1997).
Velásquez et al. (2004) provided data on a case study where linguistic factors
impacted the interpretation of the MMPI-2 profiles for Spanish-speaking individu-
als. The first case discussed by Velásquez et al. (2004) is the case of Sandra, a female
college student born and raised in Mexico who immigrated to the U.S. at the age of
13. Sandra reported being comfortable speaking both English and Spanish and
agreed to complete the MMPI-2 in English. Results from this first assessment sug-
gested Sandra presented severe psychopathology and perhaps even a psychotic dis-
order that might have required hospitalization. The clinician decided to administer
the MMPI-2 in Spanish and this time her profile appeared more congruent with the
types of interpersonal problems that had brought her to counseling in the first place.
Velásquez et al. (2004) noted that content scales in the Spanish version of the
MMPI-2 dropped by at least one standard deviation compared to her scores on the
English MMPI-2 profile. The decision to administer the MMPI-2 in Spanish pre-
vented potentially unnecessary hospitalization and psychiatric treatment (Velásquez
et al., 1997).
Others studies have found no meaningful differences between the English and
the Spanish MMPI-2 protocols. For example, Velásquez et al. (1997) administered
the MMPI-2 in English and in 4 weeks administered the Inventario Multifasico de
la Personalidad Minnesota-2, Version Hispana (IMPM-2-VH), which is the
U.S. Spanish translation, to a large group of university students in California and
found moderately high reliability coefficients that suggested comparability of
results in English and Spanish. However, in a second study also drawing from a
university sample, Velásquez et al. (1997) administered the IMPM-2-VH and then
4 weeks later administered another Spanish version of the protocol created for use
in Mexico, the Inventario Multifasico de la Personalidad de Minnesota-2 Español
(IMPM-2-E), and found no statistically significant differences between the subscale

scores yet results had a much lower reliability score than in their first study.
Velásquez et al. (1997) believe that specific items on both Spanish protocols may
have led to the lower correlations. For example, in the Version Hispana, the word
excitación is used in one item; however, the word excitación has an explicit sexual
connotation in Mexico and may not accurately represent its original intent, which
was to measure levels of emotional excitement. Thus, the IMPM-2-E uses the
expression el alboroto instead of excitación Velásquez et al. (1997). Overall, while
there is no major indication of gross linguistic differences across the MMPI-2
English and Spanish protocols, the literature assessing the linguistic equivalence of
the MMPI-2 is inconclusive.
Another commonly used personality assessment tool is the TAT, a projective test
in which individuals are presented with an ambiguous stimulus and are asked to
interpret it (Tomkins & Tomkins, 1947). The goal of the TAT is having the individual
reveal his/her own personality in the process of providing a narrative (Tomkins &
Tomkins, 1947). In an early study of personality using the TAT, Ervin (1964) admin-
istered the TAT in English in one session and then 6 weeks later administered the TAT
in French to French–English bilinguals living in the U.S. and found that they pro-
vided different narratives for each of the stimulus cards presented depending on the
language spoken. For example, when speaking French, a woman displayed higher
levels of aggression and striving for autonomy while speaking in English the same
woman presented as a dutiful and supporting housewife (Ervin, 1964). While themes
and patterns are coded and used to identify psychological traits, the TAT is a projec-
tive test without standardization because each response is unique, thereby raising
questions about its reliability and validity in a research context.
The PAI in Spanish was available since early 1990s, and it is a personality test
commonly used in clinical and forensic settings to aid in diagnosing personality
disorders, mental illness, competency to stand trial, and parental fitness evaluations
(Gardner, Boccaccini, Bitting, & Edens, 2015; Reidy, Sorensen & Davidson, 2016).
The 344-item personality test includes 22 nonoverlapping scales including validity
scales, clinical scales, treatment consideration scales, and interpersonal scales
(Morey, 2007). Validity scales assess the person’s approach to answering questions,
including faking, defensiveness, and inconsistency. The clinical scales measure per-
sonality psychopathology and include (1) somatic concerns, which measures physi-
cal complaints, (2) anxiety, measuring tension and worry, (3) anxiety-related
disorders, which assess traits related to various anxiety disorders, (4) depression,
tapping into feelings of worthlessness and sadness, (5) mania, which measures level
of energy and excitability, (6) paranoia, tapping into suspiciousness, (7) schizophre-
nia, measuring unusual sensory experiences and bizarre thoughts, (8) borderline
features, which measures identity and emotional instability, (9) alcohol problems,
measuring problem drinking, (10) drug problems, measuring excessive use of drugs
(Morey, 2007). Additional treatment consideration scales provide information on
aggression, suicidal ideation, social isolation, present stressors, and inclination to
treatment adherence including motivation, responsibility, and openness to change
(Morey, 2007). Interpersonal scales provide information on dominance, or the
degree to which a respondent displays assertiveness and control, and warmth, which
taps into the respondent’s ability to display empathy in social situations (Charnas,
Hilsenroth, Zodan, & Blais, 2010).
In a study designed to find out whether the Spanish and English versions of the
PAI are equivalent, Fernandez, Boccaccini, and Noland (2008) recruited 80 English–
Spanish bilingual college students reading at the 4th-grade level in both languages
and then randomly assigned them to three groups: respond honestly, overreport psy-
chopathology, and underreport psychopathology. Participants completed English-
language measures in one session, and then Spanish-language measures in a session
about 3 weeks apart. Results showed the PAI in English and in Spanish had high
correlations with very low mean differences across the clinical subscales. In addi-
tion, the PAI successfully detected under and overreporting in the groups where
respondents were instructed to respond the PAI correspondingly. However, as with
the MMPI-2, respondents showed an increased pattern of overreporting when com-
pleting the PAI in Spanish and instructed to respond honestly; this pattern did not
emerge when respondents completed the English PAI (Fernandez, Boccaccini, &
Noland, 2008). Thus, the authors suggest proceeding with caution when interpret-
ing the validity scales of the Spanish PAI, as they may indicate overreporting when
in reality this tendency might be the result of faulty translation or lack of consis-
tency with cultural idioms of distress.
Although it is not clear whether different results obtained with these personality
assessment instruments are the result of faulty translations, some researchers have
found evidence suggesting that immigration status and acculturation have a signifi-
cant impact on the results of clinical personality assessment. For instance, Chavira
et al. (2003) found that Hispanic immigrants in the U.S. are more likely to be diag-
nosed with a personality disorder. Some authors hypothesize that oppression, pow-
erlessness, and marginality coupled with acculturative stress result in feelings of
alienation, abandonment and loss of control, which may, in turn, be a source of
mood and interpersonal instability leading to higher rates of personality disorders in
immigrant communities (Chavira et al., 2003; Castañeda & Franco, 1985;
McGilloway, Hall, Lee, & Bhui, 2010; Montgomery & Orozco, 1985). Nonetheless,
the literature available to date does not provide sufficient evidence to disentangle
the impact of language, culture, and immigration experiences on the diagnosis of a
personality disorder.
Externalizing Psychopathology and Bilingualism
Externalizing psychopathology refers to a spectrum of cognitions, emotions, and

behaviors characterized by disinhibition, such as conduct disorder (CD) or attention
deficit hyperactivity disorder (ADHD; Krueger & Markon, 2006). These disorders
are directly affected by a lack of social cognitive skills. Specifically, social cognition
can be defined as an information process that contributes to the correct perception of
disposition and intentions of other individuals (Brune, 2001, p. 85). Social cognition
is especially important in childhood, as the way children think about other people’s
intentions shapes social interactions and, more specifically, social problem solving
(Dodge, Laird, Lochman, & Zeli, 2002; Yeates, Schultz, and Selman, 1991). Social
problem solving skills provide important ways to approach social conflict, which
requires an understanding of the other person’s beliefs, feelings, and intentions
(Marton, Abramoff, & Rosenzweig, 2005). Language is the medium used to repre-
sent these intangible concepts through verbal labels and interpretations based on
nonverbal cues (Astington & Jenkins, 1999; Cutting & Dunn, 1999). Conversely,
lack of social problem solving skills and language impairment have been consistent
predictors of ineffective interactions with peers, externalizing psychopathology,
aggression, delinquency, maladjustment, and oppositional defiant disorder (ODD),
which is a childhood disorder characterized by showing defiant and disobedient
behavior towards figures of authority (Asarnow & Callan, 1985; Brion-Meisels,
Selman, & Hoffman, 1996; Coy, Speltz, Deklyen, & Jones, 2001; Dodge et al., 1995,
2003). In fact, Zadeh, Im-Bolter, and Cohen (2007) found that language served as a
mediator between social cognition and externalizing psychopathology in a sample of
children in clinical settings. In other words, a child’s ability to use language to appro-
priately express inner thoughts and engage in meaningful communication with peers
may lead to a successful understanding of other people’s intentions and reasoning,
thereby staving off disorders such as ADHD, CD, and ODD. On the other hand, a
child who does not successfully develop the ability to express inner thoughts to com-
municate with peers and family through language will encounter difficulty under-
standing other people’s intentions and reasoning, which in turn might lead to the
development of childhood externalizing disorders.
The process of mastering social cognitive processes using language involves
increasing complexity among bilingual children. Most children acquire the L1 first
and the L2 later (i.e., sequential bilinguals; Hakuta, 1986; Oller, Pearson, & Cobo-
Lewis, 2007), as opposed to both languages simultaneously (Gutiérrez, Zepeda &
Castro, 2010; Paradis, Genesee & Crago, 2004). Children in the U.S. are particularly
encouraged to learn English, the L2, quickly; thus, many children begin to lose their
L1 when they start school (Portes & Hao, 1998; Toppelberg & Collins, 2010).
Cummins (1980) has argued that some immigrant children in the U.S. do not do
well academically because they are placed in English-learning settings without hav-
ing a sound conceptual and proficient base in L1 that can then be transferred to L2.
In fact, children who enter formal schooling contexts emphasizing L2 acquisition
are likely to experience subtractive bilingualism (i.e., deficits in both languages)
because acquisition of L2 comes at the expense of L1 (Fillmore, 1991; Hakuta &
D’Andrea, 1992; Pease-Alvarez & Vasquez, 1994; Tse, 2001). The more desirable
circumstance is additive bilingualism, where support and exposure for both lan-
guages is present and has positive outcomes, such as increased cognitive flexibility
and metacognitive processing (Ben-Zeev, 1977; Bialystok et al., 2012; Collier,
1995; Cummins, 1979; also see de Bruin, Treccani, & Della Sala, 2015).
Research has shown that subtractive bilingualism — acquiring L2 at the expense
of L1 — is related to externalizing psychopathology. Toppelberg, Munir, and Nieto-
Castañón (2006) recruited 50 Latino bilingual children referred to a child psychiatry
clinic due to externalizing psychopathology (e.g., ADHD) and obtained data on

emotional and behavioral problems, language proficiency, nonverbal intelligence,
acculturation, and socioeconomic status (SES). The researchers defined low lan-
guage ability as a child whose scores were 1.25 SD below the mean on the five oral
language tests of the Wodcock Language Proficiency Battery-Revised (WLPB) in
English and Spanish. The authors found that a total of 51% of the children in the
sample had low language ability in both English and Spanish. In addition,
Toppelberg, Munir, et al. (2006) found that having children with L2 (English) domi-
nance but low expressive (i.e., picture vocabulary) abilities in L1 was predictive of
developmental and cognitive delays. Moreover, Toppelberg, Munir, et al. (2006)
found that many of the bilingual children referred for psychological treatment to a
community clinic had significant language deficits in both L1 and L2. Children
presenting deficits in L1 and L2 expressive (i.e., picture vocabulary) and receptive
(i.e., listening comprehension, verbal analogies) areas were likely to have high lev-
els of externalizing psychopathology and low academic achievement (Toppelberg,
Munir, et al. (2006)).
In another study, Toppelberg, Nieto-Castañón, and Hauser (2006) recruited 29
Latino bilingual children referred to a child psychiatry clinic due to externalizing
disorders such as ADHD and conduct issues, and administered the WLPB in
English and Spanish to assess language proficiency, as well as the Test of
Nonverbal Intelligence—2 (TONI-2) to assess their intellectual quotient (IQ)
scores in a culturally appropriate way. The researchers found that low proficiency
in L1 and L2 were generally related to higher psychiatric symptom severity (e.g.,
social problems, thought problems, attention problems, aggression, delinquency)
among these bilingual children (Toppelberg, Nieto-Castañón, et al., 2006). These
findings raise concerns about whether bilingual children are placed at risk of
developing externalizing psychopathology due to the pressures from school sys-
tems to learn English quickly. It is likely these pressures coupled with other
known risk factors for developing externalizing psychopathology, such as pov-
erty and acculturative stress set bilingual children on a pathway towards poor
mental health outcomes (Toppelberg, Nieto-Castañón, et al., 2006). It is clear
that language development in childhood is an important factor influencing
healthy psychological development in children. However, it is also clear that sig-
nificant repercussions on mental health are likely to ensue when children acquire
L2 at the expense of L1.
Developmental Disorders in Bilingual Children
Healthy development is influenced significantly by language acquisition. Children

are able to acquire a native language quickly, figuring out its structure, rules, con-
tent, and form with relatively little input from parents and peers (Bloom & Lahey,
1978). Yet language is a significant medium of communication in social contexts
and mastery of language allows for healthy socioemotional development in children
(Toppelberg, Snow & Tager-Flusberg, 1999). However, for children with develop-
mental disorders (DD) such as intellectual disabilities, autism spectrum disorders
(ASDs), childhood schizophrenia, and language disorders, acquiring a language is
a task that imposes heavy cognitive and emotional demands (Ratner, 1997). Most
children with DD exhibit delays in language acquisition, as it is tremendously dif-
ficult for them to figure out the rules, content, and form of language (Ratner, 1997).
These language delays are significant and can make acquiring L2 an insurmount-
able task (Toppelberg et al., 1999). Specifically, the impairments associated with
social cognition in ASD and childhood schizophrenia impair strategies for L2
acquisition such as frequent and ongoing linguistic and social interactions with
native speakers (Caplan, Guthrie, & Komo, 1996; Fillmore, 1991; Tabors, 1997;
Tager-Flusberg, 1992).
As a result of difficulties in L1 acquisition, many parents and families of children
with DD are advised not to expose children to L2 (Kay-Raining Bird et al., 2012).
For example, a study of bilingual parents of children born with Down syndrome
found that these parents are often advised to only expose children to one language,
but many still chose to raise their children speaking two languages with varying
levels of success because they believed it would facilitate communication with other
members of their community, thus increasing social support (de Houwer, 2009;
Kay-Raining Bird et al., 2012; Kohnert, Yim, Nett, Kan, and Duran, 2005).
ASD is characterized by language impairments in L1, including language delays,
echolalia (i.e., repeating words without intent to communicate), difficulty starting a
conversation and communicating needs to others. Because of the inherent language
difficulties, many bilingual parents are told by professionals to keep communication
in one language only (Paradis, Genesee, and Crago, 2011). In fact, the children in
Kay-Raining Bird et al. (2012)’s study displayed language impairments in both L1
and L2. However, bilingual children with ASD were just as likely as monolingual
children with ASD to exhibit such deficits. In other words, bilingual children diag-
nosed with ASD were not worse off than monolingual children diagnosed with ASD
(Kay-Raining Bird et al., 2012). In another study, Feltmate and Kay-Raining Bird
(2008) studied French–English bilingual children born with Down syndrome and
found that they did not differ significantly on performance measures across lan-
guages. Thus, bilingual children with Down syndrome were able to acquire L2 with-
out any direct impact to their L1 proficiency. In fact, Kay-Raining Bird et al. (2005)
found that higher levels of chronological and mental age among children born with
Down syndrome were predictive of their L2 proficiency. Muller-Vahl (2012) reported
on the cases of two bilingual patients diagnosed with Tourette syndrome, which is
characterized by tics and coprolalia, which is the utterance of short, explosive words.
The patients had acquired L2 at 12 and at 6 years old, and both exhibited coprolalia
(i.e., explosive word utterances) only in L1. Interestingly, their explosive word utter-
ances were articulated in L1, even when they were conversing fluently in L2. These
studies, however, have limited sample sizes and most focus on clinical populations.
Thus, more research is needed to understand the unique and complex interactions of
bilingualism and DDs.
Schizophrenia
Schizophrenia is a disorder impairing an individual’s perception, thought, speech,

and movement (Barlow & Durand, 2015). This mental disorder is considered one
of the most debilitating and stigmatized conditions in the field of psychopathol-
ogy. Schizophrenia is characterized by positive symptoms (e.g., symptoms that
add something to a healthy individual), which include hallucinations and delu-
sions (Lindenmayer & Khan, 2006). Delusions are misrepresentation of reality
(Jaspers, 1963), while hallucinations are sensory events experienced without any
input from the environment (Palaniyappan & Liddle, 2012). Schizophrenia is also
characterized by negative symptoms (e.g., symptoms that subtract something
from a healthy individual), including avolition, alogia, anhedonia, and affective
flattening (Cohen, Natarajan, Araujo & Solanki, 2013). Briefly, avolition refers to
the individual’s inability to engage in any activity, while alogia refers to the
absence of speech (Barlow & Durand, 2015). Anhedonia is the lack of pleasure in
daily activities, whereas affective flattening refers to showing little to no facial
expressions even when the person is experiencing emotions (Berenbaum and
Oltmanns, 1992). A third domain of schizophrenia symptomology is known as
disorganized symptoms and it refers to the characteristically erratic behaviors
impacting speech, movement, and emotional reactions (Barlow & Durand, 2015).
For example, disorganized speech is characterized by lack of insight, tangential
thinking and loose association derailment (Palaniyappan & Liddle, 2012). In
addition, schizophrenic patients often display inappropriate affect such as laugh-
ing at inappropriate times or in inappropriate situations and may also display
bizarre behaviors in public. Among some individuals, schizophrenia is accompa-
nied by catatonia, which ranges from psychomotor agitation to immobility
(Barlow & Durand, 2015).
There are many factors influencing the onset of schizophrenia, and much about
this disorder is still unknown. Sommer et al. (2008) used fMRI to scan the brains of
24 individuals diagnosed with schizophrenia, schizoaffective disorder, and psy-
chotic disorder not otherwise specified (NOS) while they experienced auditory ver-
bal hallucinations and showed that the area of the brain most prominently active
among schizophrenic patients having a hallucination was Broca’s area, which is
related to speech production. Research shows that reproducibility in fMRI studies is
possible with samples of over 20 subjects (Desmond & Glover, 2002), which sug-
gests this study’s sample size is appropriate. Importantly, the Sommer et al. (2008)
study lends support to the metacognition theory of schizophrenia, which suggests
that hallucinations are not a recall of other people’s voices, but rather the individu-
al’s own voice and thoughts, except the schizophrenic individual does not recognize
the difference (Allen & Modinos, 2012).
Metacognition generally refers to the ability of human beings to think about
thinking, and implies both implicit and explicit self-awareness and reflectiveness
(Lysaker et al., 2011, p. 18). Individuals with schizophrenia exhibit a range of
impairments in metacognitive capacity, such as difficulty understanding what
others are thinking, recognizing irony, as well as difficulties with self-reflection,

such as being unable to recognize themselves as the source of their thoughts
(Lysaker et al., 2011). Research with a sample of 33 patients seeking treatment
for schizophrenia found that metacognition appeared to be separate from social
cognition, or the ability to recognize other people’s thoughts and intentions
(Lysaker et al., 2011). In addition, metacognition levels remained stable among
these schizophrenic patients even after their symptoms subsided, indicating sig-
nificant deficits in this cognitive ability continued even in the absence of positive,
negative, and disorganized symptoms (Lysaker et al., 2011).
Research on metacognition with bilinguals is limited, and mostly focuses on the
impact of L2 acquisition on executive functioning, or whether individuals are able
to inhibit irrelevant information (Bialystok et al., 2004). A recent study by Folke,
Ouzia, Bright, De Martino, and Filippi (2016) tested 31 English monolinguals and
31 bilinguals from various linguistic backgrounds by asking them to complete a
two-alternative-forced-choice task, whereby participants were shown two white
circles in a black background and were asked to choose which circle contained the
most dots. During every trial, one circle was randomly assigned to have 50 dots,
while the other circle had a variable number of dots. The task increased in difficulty
(i.e., one dot difference) if participants responded correctly to two successive trials
and vice versa, the task becoming less difficult with two successive failures. At the
end of the study, participants were asked how confident they were of their choice.
An M-ratio was used to compute a score of metacognitive efficiency by using a
hierarchical Bayesian estimation method, which showed that bilinguals exhibit a
disadvantage relative to monolinguals in terms of metacognitive efficiency, defined
as how good a person’s metacognitive accuracy is given their first order accuracy
(p. 120). In other words, metacognitive efficiency refers to the correlation of accu-
racy in choice selection and confidence for each participant. The Folke et al. (2016)
findings suggest that monolinguals and bilinguals’ cognitive systems operate differ-
ently when performing higher-order decision-making. These findings, taken
together, suggest a link between deficits in metacognitive processes resulting from
schizophrenia and metacognitive efficiency disadvantages in bilingual decision-
making. This link, however, has not been explored in the literature.
Research has confirmed that individuals who experience auditory hallucinations
report poor emotional prosody comprehension, which is an aspect of language com-
municating meaning and emotion through pitch, amplitude, and pauses rather than
the actual words spoken (Alba-Ferrara, Fernyhough, Weis, Mitchell, & Hausmann,
2012). These findings are relevant, since they intimately connect some of the defi-
cits and impairments characteristics of schizophrenia to important aspects of lan-
guage and cognition. Most of the research exploring the impact of bilingualism on
schizophrenia symptoms is based on case studies and studies with very small sam-
ples, and it would be difficult to establish causality. However, the research points to
these differential symptomology scenarios that suggest differential levels of impair-
ment across languages, thereby suggesting there are meaningfully complex bilin-
gual dynamics in the expression of psychopathology: (1) more symptoms are
experienced in L1 vs. L2; (2) more symptoms are experienced in L2 vs. L1; and (3)
no difference in symptomology experienced in L1 and L2 (Bersudsky et al., 2005;

Hughes, 1981; Javier, 1989; Paradis, 2008). Examining these scenarios is important
because it relays the different ways in which speaking more than one language
impact psychopathological symptom expression.
Increased Symptomology in L1
In the first study exploring the impact of bilingualism on schizophrenic hallucina-

tions, Hemphill (1971) reported his clinical observations of 30 bilingual South
African patients diagnosed with schizophrenia. Hemphill found that depending on
the language spoken, some of his bilingual patients were able to carry on with daily
activities at home, school, and work environments. In this study, the researcher
reported some of the patients diagnosed with a psychotic disorder in one language
but not on another (Hemphill, 1971). Del Castillo (1970) also reported on his clini-
cal experiences with bilingual schizophrenic patients, finding that they displayed
more psychotic symptoms when interviewed in L1 than when interviewed in L2.
Del Castillo (1970) reported that one of his patients, J.S., had been evaluated by an
English-speaking (L2) psychiatrist and appeared coherent and relevant, yet when
Del Castillo spoke to the patient in Spanish (L1), he showed disorganized thinking,
delusions, and rambling speech. De Zulueta, Gene-Cos, and Grachev (2001)
reported observations on three bilingual patients who exhibited more psychotic
symptoms in L1 but did not appear to show impairments in their thinking and speech
when using L2. For example, De Zulueta et al. (2001) reported on the case of Mr. Z,
who had a history of delusions and hallucinations when interviewed in English, his
native language. However, Mr. Z. had learned Spanish after puberty and one day
addressed his Spanish-speaking physician in Spanish and was able to state that in
Spanish, […] he felt he was ‘sane,’ but when he spoke in English, he went ‘mad’ (De
Zulueta, et al., 2001, p. 278). Another case study report found evidence of more
schizophrenic symptomology in L1 than L2 after conducting structured clinical
interviews and neuropsychological assessment in both languages (Schoeman,
Chiliza, Emsley, & Southwood, 2007). Similar case study reports suggest patients
expressed more and more intense schizophrenic symptoms in L1 than on L2
(Malgady and Costantino, 1998; Oquendo, 1996; De Zulueta, 1984).
An interesting approach adopted by Mattulis (1977) tried to capitalize on the find-
ings of attenuated or absent schizophrenia symptoms in L2 by having a total of 18
patients learn an L2 while interned in a psychiatric ward. While quantitative data was
not used as evidence, Mattulis (1977) reported the majority of the patients appeared to
improve significantly as a result of L2 instruction, with some patients even released
early. These findings, however, cannot be causally connected because Mattulis (1977)
did not use a rigorous experimental design. While it is difficult to ascertain whether
learning an L2 helped these patients, there is evidence that bilinguals diagnosed with
schizophrenia exhibit differential symptomatology (Dugan, 2014).
Increased Symptomology in L2
Using a case study approach, Malo Ocejo, Medrano Albeniz, and Uriarte Uriarte
(1991) reported clinical observations on four bilingual schizophrenic patients from
Spain whose hallucinations only occurred in L2 (Spanish), which was learned in a
school context, as opposed to L1 (Basque), which had been learned since childhood.
Malo Ocejo et al. (1991) speculated these cases of hallucinations occurring in L2 were
explained by the fact that language acquisition took place in different sociocultural,
functional, or temporal contexts, which results in different values attached to each
language. In another case study, Dores et al. (1972) documented the case of a Wolof–
French bilingual patient who exhibited disorganized, incoherent, and disconnected
speech in French (L2), but appeared coherent and calm when speaking Wolof (L1).
Moreover, Lukianowicz (1962) reported on cases when the schizophrenic patients
heard friendly hallucination voices in L1 and the hostile, threatening voices in L2. In
addition, Marcos, Alpert, et al. (1973) reported another case study where a group of
bilingual schizophrenic patients expressed more symptomology in L2. Marcos,
Alpert, et al. (1973) concluded that symptomology expression in L2 serves the func-
tion of affective distancing, which is the tendency to express symptoms in the less
emotionally-loaded second language in most sequential bilinguals. The lack of emo-
tionality and affective expression, however, was interpreted by clinicians as affective
flattening, which is a symptom of schizophrenia (Marcos, Alpert, et al. 1973).
However, these reports rely on case studies of a few patients by reporting clinical
observations rather than systematic measurements in rigorous experimental designs.
No Differential Symptomology Across L1 and L2
Another line of research findings indicates no differences in schizophrenic sympto-

mology across languages for bilingual patients. For example, Bersudsky et al.
(2005) conducted a unique study exploring bilingualism and schizophrenia by
investigating eight Russian immigrants to Israel. These patients had been diagnosed
with schizophrenia prior to immigrating to Israel and being exposed to Hebrew.
Bersudsky et al. (2005) assessed the effects of schizophrenia symptoms on these
patients’ L2 acquisition, specifically assessing syntactic ability, lexicon, and prag-
matics through interviews conducted in Hebrew. Their results showed no evidence
of significant symptom differences across languages; however, when the research-
ers compared bilingual schizophrenic patients to healthy bilingual immigrants on
the same domains, they found almost identical mistakes in L2 syntax and pragmat-
ics (Bersudsky et al., 2005). Bersudsky et al. concluded that schizophrenia might be
only a minor factor impacting L2 acquisition. However, Bersudsky et al. failed to
account for the impact of age at time of L2 acquisition, as well as the impact of
acculturation in their study. Another case study reported that one Afrikaans–English
bilingual diagnosed with schizophrenia exhibited similar symptoms in both lan-
guages (Southwood et al., 2009). Findings from interviews with four bilinguals
diagnosed with schizophrenia using a content analysis approach found evidence of

pragmatic deficits in both languages (Theron et al., 2011). Moreover, Smit et al.
(2011) used data from the Theron et al. (2011) but matched it to comparable control
subjects to detect any differences between L2 samples of schizophrenic patients and
L2 samples of healthy controls. Their analysis showed no significant differences
between schizophrenics and healthy controls’ L2 production (Smit et al., 2011).
However, Wang, Morales and Hsu (1998) reported on six bilingual patients diag-
nosed with schizophrenia and found different symptomology patterns: two experi-
enced auditory hallucinations in both languages, two heard voices mostly in L1 but
sometimes in L2, one heard voices only in L1, and another one only in L2. Wang
et al. (1998) concluded that auditory hallucinations reflected the patients’ thinking
process, which in turn, influenced the content of the hallucinations. Importantly, all
the results presented relied on case studies of patients rather than systematic, experi-
mental methodology, which reduces their validity and reliability.
heories Explaining Differential Symptomology in Bilingual

T
Schizophrenics
Early reports of these differential symptoms in schizophrenia attributed different symp-

tom expression across L1 and L2 to the process of language lateralization, proposing
that an L2 acquired after puberty resulted in most L2 processing occurring in the right
hemisphere (De Zulueta, 1984). However, the theory of hemispheric asymmetry has
lost support due to recent findings using functional magnetic resonance imaging
(fMRI) indicating involvement of both right and left hemispheres in language acquisi-
tion and production (Dugan, 2014; Paradis, 2004). Paradis (2004) proposed a new
model for understanding the causes of differential symptoms among bilingual schizo-
phrenics. His model relied on explaining L2 language production as heavily dependent
on declarative knowledge, which requires bringing attention to the expressions and
words an individual wants to use (Paradis, 2004). Paradis (2004) proposed that L1’s
automaticity means it does not require conscious planning, therefore making L1 more
conducive to hallucinatory content. Using data from neuropsychological assessment in
L1 and L2 for a case study of a schizophrenic patient, Schoeman et al. (2007) proposed
that the different symptoms observed across languages might be due to schizophrenia’s
impact on subcortical structures involved in implicit memory (i.e., not available to con-
scious introspection) processing for L1. Conversely, Schoeman et al. (2007) propose
L2 relies more heavily on explicit memory (i.e., memory that can be consciously
recalled), which is mostly managed by cortical areas (Moretti et al., 2001).
There is limited evidence available on the impact of bilingualism on schizo-
phrenic symptomology, and the few studies available rely on case studies and small
samples. The lifetime prevalence of schizophrenia is estimated at 1% (APA, 2013;
Barlow & Durand, 2015). Thus, the population of schizophrenic patients is limited
even without narrowing it down to only bilingual patients diagnosed with schizo-
phrenia. Moreover, most of the studies reviewed here did not provide reliable
accounts of levels of bilingualism, language acquisition, or bilingual proficiency.

For the most part, these case reports relied on clinical observations. Thus, while it is
possible that bilingualism does impact the expression of schizophrenic symptoms,
it is unclear how or to what extent. More methodologically sound research is needed
to understand bilingualism’s impact on schizophrenia symptoms. For example,
studies could conduct fMRI or MRI scans of bilingual schizophrenics while they are
having an auditory hallucination. Other research may look at other aspects of bilin-
gualism among schizophrenics, such as age at which the languages were acquired,
acculturation, cultural worldviews, and language proficiency as well as how these
variables may interact with the presence and severity of schizophrenia symptoms.
ilingualism as a Barrier to Diagnosis and Treatment

B
of Psychopathology
The main tool for assessing and diagnosing psychopathology is the clinical inter-
view, where a series of questions about symptoms and experiences are used to
gather information about the patient’s mental state (Westernmeyer & Janca, 1997).
The clinical interview focuses on obtaining information about the person’s behav-
ior, attitudes, and emotions, a general history of the individual’s social, interper-
sonal, upbringing and familial experiences, life stressors, educational and medical
history, and cultural concern, as well as the onset, duration, and course of the pre-
senting problems (Barlow & Durand, 2015). Another important tool providing
information to aid in psychopathological diagnosis is psychological testing and
evaluation, which are empirically validated tools to determine the cognitive, emo-
tional, and behavioral responses associated with specific disorders (Barlow &
Durand, 2015). Both tools for assessing and diagnosing psychopathology rely heav-
ily on language in its spoken or written forms (Javier, 2007). The language-
dependent nature of assessment and diagnosis of mental disorders may represent a
barrier for many bilinguals, especially those with limited proficiency on the lan-
guage of the interview or assessment (Malgady & Zayas, 2001). Moreover, when
clinicians do not accurately assess the language proficiency of bilinguals they may
reach inaccurate diagnoses that will be followed by potentially unnecessary treat-
ment and continuing suffering for the affected patients (Malgady, Rogler, &
Constantino, 1987). Indeed, language may be a significant barrier to help-seeking
for many immigrant bilingual groups dealing with mental illness.
Clinician Bias
One of the main issues with current assessment and diagnostic tools is its inherent
bias due to being created, standardized, and validated from an English-speaking,
nonminority, middle-class perspective (Malgady et al., 1987). Even when an
individual claims proficiency in the language in which the interview or psychologi-

cal assessment is conducted, it is possible to find differences in diagnosis and
assessment results depending on the language used (Malgady et al., 1987). Moreover,
age of L2 acquisition is an important factor to consider when assessing bilinguals,
as the older an individual is when acquiring L2, the more likely they are to develop
speech and word distortions that can be misinterpreted by clinicians.
A series of studies conducted by Marcos and colleagues (1976; Marcos, Urcuyo,
Kesselman, & Alpert, 1973) found that clinicians conducting interviews in English
with bilingual Spanish–English patients in psychiatric settings were more likely to
diagnose them with more severe psychopathology than clinicians speaking Spanish
during the clinical interviews. However, it was unclear whether the results of these
studies were related to language or implicit bias on the part of the clinician inter-
viewers (Malgady et al., 1987). Conversely, Price and Cuellar (1981) reported that
bilingual clinicians rated videotaped patient clinical interviews in Spanish (L1) with
more severe psychopathology than those conducted in English (L2). A similar result
was reported by Malgady and Costantino (1998), who found that ethnic and linguis-
tic matching between clinician and patient led to more severe psychopathological
diagnoses. In other words, clinicians of the same ethnic background and speaking
the same language as the patients were more likely to diagnose more severe psycho-
pathology than English-speaking, Anglo clinicians. While it is unclear what dynam-
ics of clinical interviewing influenced the different results obtained, what is clear is
that the language used to conduct the interview did have an impact on the diagnosis
of psychopathology among bilingual patients.
Emotional Expression Across Languages
Research suggests individuals are more emotionally expressive in L1, and thus, are
better able to express discomfort and hostility in clinical interviews using L1
(Gonzalez-Reigosa, 1976). There is evidence that emotional responses to taboo sub-
jects are experienced differently in L1 and L2. For example, Harris, Gleason, and
Aycicegi (2006) found increased physiological arousal in response to taboo words
presented in the participant’s L1. Gonzalez-Reigosa (1976) found that Spanish–
English bilinguals indicated higher state anxiety when reading lists of taboo words
in their L1 than in their second language. Bilingual speakers are also more likely to
feel more inhibited when discussing embarrassing topics in L1 than L2. For exam-
ple, Bond and Lai (1986) found that Chinese–English speakers were more embar-
rassed when talking about sex in Chinese (L1), than in English (L2). Dewaele
(2004) has found that swear and taboo words tend to evoke more intense emotional-
ity when spoken in L1. The evidence to date suggests switches in cultural frames of
reference are likely the result of differing sociolinguistic contexts surrounding the
learning and acquisition of L1 and L2 (Dewaele & Pavlenko, 2002). Specifically, an
individual’s L1 is more frequently used at home and associated with early aware-
ness of a wide range of emotions, while an L2 is usually acquired in the more
emotionally neutral context of schooling and achievement (Guttfreund, 1990;

Harris, Aycicegi, & Gleason, 2003; Opitz & Degner, 2012; but see Heredia &
Cieślicka, 2014).
In addition, there is consistent evidence suggesting bilinguals’ autobiographical
memories and emotions are impacted by the language spoken. Advances in the
study of bilingual memory suggest that memories and emotions are encoded in the
context associated to a particular language, and thus, recall of these memories in the
language in which they were stored is more detailed and emotionally intense (Javier,
Barroso, & Muñoz, 1993; Pérez Foster, 1996; Schrauf, 2000; Heredia & Cieślicka,
2014; see also Heredia & García, this volume). For example, recall of memories
from childhood tends to be more intense when retrieved in the first language, which
is the language associated with the event (Marian & Neisser, 2000; Marian &
Kaushanskaya, 2004; Schrauf, 2000; Laguzzi, 2014). Thus, it is possible that in the
context of diagnostic and assessment interviews patients will recall more details and
emotions about an event in L1 than L2.
Personality Assessment Issues
Studies have found that language can serve as a cue to switch interpretations
about personality and self-perception among bilinguals (Bond & Yang, 1982;
Hong, Chiu, and Kung, 1997; Ramírez-Esparza et al., 2006). For example, a study
by Chen and Bond (2010) found that the language of a personality assessment
instrument impacted fluent English–Chinese bilinguals’ personality ratings (Chen
& Bond, 2010). In addition, Briley et al. (2005) found that language spoken when
giving instructions about a group project requiring Hong Kong Chinese–English
bilinguals to choose from a variety of business strategies predicted whether the
participants engaged in compromising strategies. Therefore, research suggests
that language can trigger a culturally influenced approach to decision-making in
bilinguals.. These choices, in turn, can impact an individual’s understanding and
expression of normal and abnormal behavior, which is essential during a psychi-
atric evaluation.
Conclusion
The literature addressing the relationship of bilingualism to various forms of psy-

chopathology is scant and mostly reliant on case studies. Psychological dysfunction
by definition encompasses impairment in the domains of cognition, emotion, and
behavior. These same domains are linked to L1 and L2 acquisition, which suggests
the possibility of differential expression of psychopathological symptoms given the
differences in processing observed in native and second language structures. While
differential symptoms in various psychological disorders have been observed across
languages, there is not enough methodologically sound research to substantiate that

these differences are due to being bilingual. Moreover, speaking an L2 may be a
barrier to accurate diagnosis and assessment of psychopathology. Studies show that
clinicians speaking different languages during clinical interviews reach different
conclusions about the nature and severity of the bilingual’s psychopathology.
Indeed, research has documented the language- and context-dependent nature of
emotional expression, as well as personality and self-perception. Therefore, it is
possible that differential symptomology reports of psychopathology are, in fact, the
result of faulty, unintentionally biased psychiatric diagnosis.
A major weakness permeating the literature on bilingualism and psychopathol-
ogy lies in its lack of systematic, empirically rigorous methodologies. This can be
partly attributed to the relatively low prevalence of some disorders, but it certainly
is not an insurmountable obstacle. There are plenty of methodologically rigorous
studies exploring these psychopathological disorders among monolingual patients
all over the world. Thus, this absence of sound methodology may be the result of
lack of interest in exploring this area of research. Perhaps the intersection of bilin-
gualism and psychopathology is not an area of research likely to attract attention for
their potential applications to treatment. However, the reality is that the majority of
the world population speaks more than one language and both the fields of bilin-
gualism and psychopathology would gain a more robust knowledge base if more
studies focus on this population.
Clinical Recommendations
A number of recommendations are in order for clinicians working with bilingual

patients. While the research studying the intersection of bilingualism and psychopa-
thology is limited, there are many bilingual patients experiencing mental illness
who are seeking help from a mostly monolingual clinician workforce (Malgady &
Zayas, 2001). Below, I summarize possible recommendations for the treatment and
assessment of psychopathology among bilinguals based on the literature reviewed.
Assessment of language proficiency and dominance: Prior to engaging in any
other type of clinical assessment or diagnosis, clinicians and mental health practi-
tioners might consider conducting a thorough assessment of the patient’s language
proficiency and dominance. A thorough assessment of this type might include ques-
tions related to L2 age of acquisition, the contexts where each language is used
(e.g., home, work), confidence level speaking each language, and confidence read-
ing and writing in each language. Javier (2007) provides an example of a thorough
language proficiency assessment for use in clinical settings. Dunn and Fox Tree
(2009) designed a brief measure of language dominance that can provide a numeri-
cal index of proficiency in each language.
Familiarization with cultural idioms of distress: Whether the clinician is proficient
in the patient’s L2 or relies on translation services, it is important for mental health
practitioners to familiarize themselves with specific idioms of distress commonly
used within the patient’s culture (Kleinman, 1988). While different patients’ cultures
emphasize different idioms to express symptoms of mental illness, it is important that
clinicians are equipped with basic knowledge and the appropriate tools to obtain this
knowledge quickly and efficiently. Most counseling and mental health practitioner
programs mandate taking a course in multicultural considerations of working with
diverse patients, but it is impossible to cover all possible idioms of distress in one
course. However, these courses can emphasize to clinicians-in-training the impor-
tance of acquiring this knowledge when assessing and diagnosing patients of diverse
linguistic and cultural backgrounds. The effective clinician might want to investigate
and research specific contextual variables that can potentially impact the assessment
and diagnosis of psychopathology among bilingual patients.
Carefully select personality assessment instruments: When clinicians are tasked
with assessing and diagnosing personality disorders, they rely on clinical assess-
ment instruments. However, it is very important for clinicians to consider the reason
for referral for assessment prior to selecting an assessment instrument (Javier,
2007). Once the reason for referral has been clarified, it is important to select an
appropriate instrument. In order to make this selection, the clinician might consider
consulting the assessment’s manual to gather information on norming samples,
validity, and reliability indices of the instrument with diverse populations, as well as
consulting research using the assessment instrument with the specific linguistically
and culturally diverse populations of interest. If an assessment instrument is avail-
able in the client’s L2, it may be relevant to consider administering the instrument
to compare results obtained in both languages (Malgady & Zayas, 2001). It is, how-
ever, important to exercise care examining the specific background of the norming
sample for translation of assessment instruments, as there is variation in meanings
and expressions even within the same language.
Provide simultaneous L1 and L2 support to children of immigrants: The research
reviewed here suggests subtractive bilingualism may be linked to increased levels of
externalizing psychopathology among bilingual immigrant children (Toppelberg,
Nieto-Castañón, et al. 2006). This may be the result of emphasizing L2 acquisition
and instruction among these bilingual children before they develop enough concep-
tual skills in L1. Researchers and educators might foster and support continued
learning and experiences in the children’s L1, as it is usually the language spoken
by parents and family. The studies reviewed in this chapter indicate that subtractive
bilingualism is detrimental to the children, specifically impacting their social cogni-
tion and problem solving abilities.
Support L2 language learning among children with developmental disabilities:
Although there is limited research exploring the impact of acquiring L2 among
children with DD, the studies available to date suggest children exhibit similar
impairments and deficits in L1 and L2. Thus, suggesting that learning L2 will not
interfere with their L1 skills and may, in fact, facilitate interaction with important
family and support systems in the child’s life. Families caring for children with DD
should also be made aware that there are variations on the level of L2 proficiency
children with DD will achieve (Kay-Raining Bird et al., 2012).
Assessing schizophrenic symptoms in both L1 and L2: Although the research

reviewed here relied mostly on case studies and small samples, one commonality
was clear: bilingual patients might exhibit different symptoms in each language
they speak. Thus, it is of utmost importance to conduct a clinical interview in both
languages and if available, conduct psychological and neuropsychological assess-
ments in both languages as well. If there are no clinicians who speak the patient’s
language, then it is important to contact a well-trained translation service that can
help the clinician obtain a more complete clinical picture of the patient’s psychopa-
thology (Paradis, 2008; Marcos, 1976).
References
Alarcón, R. D., & Foulks, E. F. (1995a). Personality disorders and culture: Contemporary and
clinical views (Part A). Cultural Diversity and Mental Health, 1, 3–14.
Alarcón, R. D., & Foulks, E. F. (1995b). Personality disorders and culture: Contemporary and
clinical views (Part B). Cultural Diversity and Mental Health, 1, 79–91.
Alba-Ferrara, L., Fernyhough, C., Weis, S., Mitchell, R. L. C., & Hausmann, M. (2012).
Contributions of emotional prosody comprehension deficits to the formation of auditory verbal
hallucinations in schizophrenia. Clinical Psychology Review, 32, 244–250.
Allen, P., & Modinos, G. (2012). Structural neuroimaging in psychotic patients with auditory
verbal hallucinations. In J. D. Blom & I. E. C. Sommer (Eds.), Hallucinations: Research and
Practice. Springer Science+Business: Berlin, Germany.
American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders
(5th ed.). Washington, DC.
Asarnow, J. R., & Callan, J. W. (1985). Boys with peer adjustment problems social cognitive pro-
cesses. Journal of Consulting and Clinical Psychology, 53, 80–87.
Astington, J. W., & Jenkins, J. M. (1999). A longitudinal study of the relation between language
and theory-of-mind development. Development and Psychopathology, 35, 1311–1320.
Barlow, D. H., & Durand, V. M. (2015). Abnormal psychology: An integrative approach (7th ed.).
Melbourne, Australia: Cengage Learning.
Bastiaansen, L., Rossi, G., Schotte, C., & De Fruyt, F. (2011). The structure of personality disor-
ders: Comparing the DSM-IV-TR Axis II classification with the Five-Factor Model framework
using structural equation modeling. Journal of Personality Disorders, 25, 378–396.
Ben-Zeev, S. (1977). The effect of bilingualism in children from Spanish-English low eco-
nomic neighborhoods on cognitive development and cognitive strategy. Working Papers on
Bilingualism, 84, 83–122.
Berenbaum, H., & Oltmanns, T. F. (1992). Emotional experience and expression in schizophrenia
and depression. Journal of Abnormal Psychology, 101, 37–44.
Bersudsky, Y., Fine, J., Gorjalstan, I., Chen, O., & Waters, J. (2005). Schizophrenia and second
language acquisition. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 29,
535–542.
Bhui, K. (1999). Common mental disorders among people with origins in or immigrant from India
and Pakistan. International Review of Psychiatry, 11, 136–144.
Bialystok, E., Craik, F.I., Klein, R., & Viswanathan, M. (2004). Bilingualism, aging, and cognitive
control: Evidence from the Simon task. Psychology of Aging, 19, 290–303.
Bhui, K., Bhugra, D., Goldberg, D., Sauer, J., & Tylee, A. (2004). Assessing the prevalence of
depression in Punjabi and English primary care attenders: The role of culture, physical illness
and somatic symptoms. Transcultural Psychiatry, 41, 307–322.
Bialystok, E., Craik, F. I. M., & Luk, G. (2012). Bilingualism: Consequences for mind and brain.
Trends in Cognitive Science, 16, 240–250.
Bloom, L., & Lahey, M. (1978). Language development and language disorders. New York, NY:
Wiley.
Bond, M. H., & Lai, T. M. (1986). Embarrassment and code-switching into a second language.
Journal of Social Psychology, 126, 179–186.
Bond, M. H., & Yang, K. (1982). Ethnic affirmation versus cross-cultural accommodation: The
variable impact of questionnaire language on Chinese bilinguals from Hong Kong. Journal of
Cross-Cultural Psychology, 13, 169–185.
Briley, D. A., Morris, M. W., & Simonson, I. (2005). Cultural chameleons: Biculturals, conformity
motives, and decision making. Journal of Consumer Psychology, 15, 351–362.
Brion-Meisels, S., Selman, R. L., & Hoffman, A. M. (1996). From fight or flight to collabora-
tion: A framework for understanding individual and institutional development in the school.
In A. M. Hoffman (Ed.), Schools, violence, and society (pp. 163–184). Eastport, CT: Praeger.
Brune, M. (2001). Social cognition and psychopathology in an evolutionary perspective: Current
status and proposals for research. Psychopathology, 34, 85–94.
Bucci, W. (1997). Psychoanalysis and cognitive science: A multiple code theory. New York, NY:
Guilford Publications.
Bucci, W., & Freedman, N. (1981). The language of depression. Bulletin of the Menninger Clinic,
45, 334–358.
Butcher, J. N., Graham, J. R., Ben-Porath, Y. S., Tellegen, A., Gahlstrom, W. G., & Kaemmer,
B. (2001). Minnesota Multiphasic Personality Inventory—2 (MMPI-2) Manual. Minneapolis,
MN: University of Minnesota Press.
Cabassa, L. J., Hansen, M. C., Palinkas, L. A., & Ell, K. (2008). Azucar y nervios: Explanatory
models and treatment experiences of Hispanics with diabetes and depression. Social Science
and Medicine, 66, 2413–2424.
Caplan, R., Guthrie, D., & Komo, S. (1996). Conversational repair in schizophrenic and normal
children. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 950–958.
Castañeda, R., & Franco, H. (1985). Sex and ethnic distribution of borderline personality disorder
in an inpatient sample. American Journal of Psychiatry, 142, 1202–1203.
Centeno, J. G., & Obler, L. K. (2001). Principles of bilingualism. In M. O. Ponton & J. Leon-
Carrion (Eds.), Neuropsychology and the Hispanic patient (pp. 75–86). Mahwah: Lawrence
Erlbaum Associates.
Charnas, J. W., Hilsenroth, M. J., Zodan, J., & Blais, M. A. (2010). Should I stay or should I go?
Personality Assessment Inventory and Rorschach indices of early withdrawal from psycho-
therapy. Psychotherapy: Theory, Research, Practice, Training, 47, 484–499.
Chavira, D. A., Grilo, C. M., Shea, M. T., Yen, S., Gunderson, J. G., Morey, L. C., et al. (2003).
Ethnicity and four personality disorders. Comprehensive Psychiatry, 44, 483–491.
Chen, S. X., Benet-Martinez, V., & Ng, J. C. K. (2014). Does language affect personality perception?
A functional approach to testing the Whorfian hypothesis. Journal of Personality, 82, 130–143.
Chen, S. X., & Bond, M. H. (2010). Two languages, two personalities? Examining language effects
on the expression of personality in a bilingual context. Personality and Social Psychology
Bulletin, 36, 1514–1528.
Cohen, C. I., Natarajan, N., Araujo, M., & Solanki, D. (2013). Prevalence of negative symptoms
and associated factors in older adults with schizophrenia spectrum disorder. American Journal
of Geriatric Psychiatry, 21, 100–107.
Collier, V. (1995). Acquiring a second language for school. Directions in Language and Education,
1(4). Retrieved from www.usc.edu/dept/education/CMMR/CollierThomas_Acquiring_L2_
for_School. Acquiring a second language for school. Vol. 1: National Clearinghouse for
Bilingual Education.
Coy, K., Speltz, M. L., Deklyen, M., & Jones, K. (2001). Social-cognitive processes in preschool
boys with and without oppositional defiant disorder. Journal of Abnormal Child Psychology,
29, 107–119.
Cummins, J. (1979). Linguistic interdependence and the educational development of bilingual
children. Review of Educational Research, 49, 222–251.
Cummins, J. (1980). The cross-lingual dimensions of language proficiency: Implications for bilin-
gual education and the optimal age issue. TESOL Quarterly, 14, 175–187.
Cutting, A. L., & Dunn, J. (1999). Theory of mind, emotion understanding, language, and family
background: Individual differences and interrelations. Child Development, 70, 853–865.
de Bernier, G. L., Kim, Y. R., & Sen, P. (2014). A systematic review of the global prevalence of
personality disorders in Asian populations. Personality and Mental Health, 8, 264–275.
de Bruin, A., Treccani, B., & Della Sala, S. (2015). Cognitive advantage in bilingualism: An example
of publication bias? Psychological Science, 26, 99–107.
de Houwer, A. (2009). Bilingual first language acquisition. Toronto, ON: Multilingual Matters.
De Zulueta, F. I. S. (1984). The implications of bilingualism in the study and treatment of psy-
chotic disorders: A review. Psychological Medicine, 14, 541–557.
De Zulueta, F. I. S., Gene-Cos, N., & Grachev, S. (2001). Differential psychotic symptomatology in
polyglot patients: Case reports and their implications. British Journal of Medical Psychology,
74, 277–292.
Del Castillo, J. C. (1970). Influence of language upon symptomatology in foreign-born patients.
American Journal of Psychiatry, 127, 160–162.
Desmond, J. E., & Glover, G. H. (2002). Estimating sample size in functional MRI (fMRI) neuro-
imaging studies: Statistical power analysis. Journal of Neuroscience Methods, 118, 115–128.
Dewaele, J. (2004). The emotional force of swearwords and taboo words in the speech of multilin-
guals. Journal of Multilingual and Multicultural Development, 25, 204–223.
Dewaele, J. M., & Pavlenko, A. (2002). Emotion vocabulary in interlanguage. Language Learning,
52, 263–322.
Dodge, K. A., Laird, R., Lochman, J. E., & Zeli, A. (2002). Multidimensional latent-construct
analysis of children’s social information processing patterns: Correlations with aggressive
behaviour problems. Psychological Assessment, 14, 60–73.
Dodge, K. A., Lansford, J. E., Burks, V. S., Bates, J. E., Pettit, G. S., Fontaine, R., et al. (2003).
Peer rejection and social information-processing factors in the development of aggressive
behaviour problems in children. Child Development, 74, 374–393.
Dodge, K. A., Pettit, G. S., Bates, J. E., & Valente, E. (1995). Social information processing pat-
terns partially mediate the effect of early physical abuse on later conduct problems. Journal of
Abnormal Psychology, 104, 632–643.
Dores, M., M’Bodj, M., M’Bodj, B., & N’Dao, S. (1972). Bilinguisme et psychopathologie
[Bilingualism and psychopathology]. Psychopathologie Africaine, 8, 425–441.
Dugan, J. E. (2014). Second language acquisition and schizophrenia. Second Language Research,
30, 307–321.
Dunn, A. L., & Fox Tree, J. E. (2009). A quick, gradient bilingual dominance scale. Bilingualism:
Language and Cognition, 12, 273–289.
Endler, N. S., & Kocovski, N. L. (2002). Personality disorders at the crossroads. Journal of
Personality Disorders, 16, 487–502.
Ervin, S. M. (1964). Language and TAT content in bilinguals. Journal of Abnormal and Social
European Commission (2006). Special Eurobarometer 243: Europeans and their languages.
Retrieved from http://ec.europa.eu/public_opinion/archives/ebs/ebs_243_sum_en.pdf
Fabrega, H. (1994). Personality disorders as medical entities: A cultural interpretation. Journal of
Feltmate, K., & Kay-Raining Bird, E. (2008). Language learning in four bilingual children with
Down syndrome: A detailed analysis of vocabulary and morphosyntax. Canadian Journal of
Speech–Language Pathology and Audiology, 32, 6–20.
Fernandez, K., Boccaccini, M. T., & Noland, R. M. (2008). Detecting over-and underreporting of
psychopathology with the Spanish-language Personality Assessment Inventory: Findings from
a simulation study with bilingual speakers. Psychological Assessment, 20, 189–194.
Fillmore, L. W. (1991). When learning a second language means losing the first. Early Childhood
Research Quarterly, 6, 323–346.
Folke, T., Ouzia, J., Bright, P., De Martino, B., & Filippi, R. (2016). A bilingual disadvantage in
metacognitive processing. Cognition, 150, 119–132.
Fontani-Salvador, P., & Rogers, R. (1997). Spanish versions of the MMPI-2 and PAI: An investiga-
tion of concurrent validity with Hispanic patients. Assessment, 4, 29–39.
Fonseca-Pedrero, E., Paino, M., Lemos-Giraldez, S., & Muñiz, J. (2013). Cluster B maladaptive
personality traits in Spanish adolescents. Revista de Psiquiatría y Salud Mental, 6, 129–138.
Gardner, B. O., Boccaccini, M. T., Bitting, B. S., & Edens, J. F. (2015). Personality Assessment
Inventory scores as predictor of misconduct, recidivism, and violence: A meta-analytic review.
Psychological Assessment, 27, 534–544.
Garrido, M. & Velasquez, R. (2006). Interpretation of Latino/Latina MMPI-2 Profiles: Review and
Application of Empirical Findings and Cultural- Linguistic Considerations. In J.N. Butcher
(Ed.) The MMPI-2-: A practitioner’s guide. (pp. 477-504). Washington, DC: American
Psychological Association.
Gonzalez-Reigosa, F. (1976). The anxiety-arousing effect of taboo words in bilinguals. In C. D.
Spielberger & R. Diaz-Guerrero (Eds.), Cross-cultural anxiety (pp. 89–105). Washington, DC:
Hemisphere.
Gross, R., Olfson, M., Gameroff, M. J., Carasquillo, O., Shea, S., Feder, A., et al. (2005).
Depression and glycemic control in Hispanic primary care patients with diabetes. Journal of
General Internal Medicine, 20, 460–466.
Guarnaccia, P. J., Lewis-Fernandez, R., & Marano, M. R. (2003). Toward a Puerto Rican
popular nosology: Nervios and ataque de nervios. Culture, Medicine, and Psychiatry, 27,
337–366.
Gutiérrez, K. D., Zepeda, M., & Castro, D. C. (2010). Advancing early literacy learning for all
children: Implications of the NELP report for dual-language learners. Educational Researcher,
39, 334–339.
Guttfreund, D. G. (1990). Effects of language usage on the emotional experience of Spanish-
English and English-Spanish bilinguals. Journal of Consulting and Clinical Psychology, 58,
604–607.
Hakuta, K. (1986). Mirror of language: The debate of bilingualism. New York, NY: Basic Books.
Hakuta, K., & D’Andrea, D. (1992). Some properties of bilingual maintenance and loss in Mexican
background high school students. Applied Linguistics, 13, 72–99.
Hakuta, K., & Pease-Alvarez, L. (1994). Proficiency, choice and attitudes in bilingual Mexican-
American children. In G. Extra & L. T. Verhoeven (Eds.), The cross-linguistic study of bilin-
gual development (pp. 145–164). Amsterdam: North-Holland.
Harris, C. L., Aycicegi, A., & Gleason, J. B. (2003). Taboo words and reprimands elicit greater
autonomic reactivity in a first language than in a second language. Applied PsychoLinguistics,
24, 561–579.
Harris C. L., Gleason J. B., Ayçiçegi A. (2006). When is a first language more emotional?
Psychophysiological evidence from bilingual speakers. In A. Pavlenko (Ed.), Bilingual
minds: Emotional experience, expression, and representation (pp. 257–283). Clevedon:
Multilingual Matters.
Hemphill, R. E. (1971). Auditory hallucinations in polyglots. South African Medical Journal, 45,
1391–1394.
Heredia, R. R., & Cieślicka, A. B. (2014). Bilingual memory storage: compound-coordinate
and derivatives. In R. R. Heredia & J. Altarriba (Eds.), Foundations of bilingual memory
(pp. 11–39). New York, NY: Springer.
Hofstede, G., & McCrae, R. R. (2004). Personality and culture revisited: Linking traits and dimen-
sions of culture. Cross-Cultural Research, 38, 52–88.
Hong, Y.Y., Chiu, C.Y., & Kung, T.M. (1997). Bringing culture out in front: Effects of cul-
tural meaning system activation on social cognition. In K. Leung, Y. Kashima, U. Kim, &
S. Yamaguchi (Eds.), Progress in Asian Psychology (Vol. 1, pp. 135–146). Singapore: Wiley.
Hughes, G. W. (1981). Neuropsychiatric aspects of bilingualism: A brief review. British Journal
of Psychiatry, 139, 25–28.
Hull, R.A. (2003). How does bilingualism matter? A meta-analytic tale of two hemispheres.
Unpublished doctoral dissertation, Texas A&M University. College Station, TX.
Jaspers, K. (1963). General psychopathology. (J. Hoenig & M.W. Hamilton, Trans. Manchester,
UK: Manchester University Press.
Javier, R. A. (1989). Linguistic consideration in the treatment of bilinguals. Psychoanalytic
Javier, R. A. (2007). The bilingual mind: Thinking, feeling, and speaking in two languages.
New York, NY: Springer.
Javier, R. A., Barroso, F., & Muñoz, M. (1993). Autobiographical memory in bilinguals. Journal
of Psycholinguistic Research, 2, 319–338.
Kashima, E. S., & Kashima, Y. (1998). Culture and language: The case of cultural dimensions in
personal pronoun use. Journal of Cross-Cultural Psychology, 29, 461–486.
Kay-Raining Bird, E. (2006). The case for bilingualism in children with Down syndrome. In
R. Paul (Ed.), Language disorders from a developmental perspective: Essays in honor of Robin
S. Chapman (pp. 249–275). Mahwah, NJ: Lawrence Erlbaum Associates.
Kay-Raining Bird, E., Cleave, P., Trudeau, N., Thordardottir, E., Sutton, A., & Thorpe, A. (2005).
The language abilities of bilingual children with Down syndrome. American Journal of
Speech–Language Pathology, 14, 187–199.
Kay-Raining Bird, E., Lamond, E., & Holden, J. (2012). Survey of bilingualism in autism
spectrum disorders. International Journal of Language and Communication Disorders,
47, 52–64.
Kleinman, A. (1988). The illness narratives: Suffering, healing, and the human condition.
New York, NY: Basic Books.
Kleinman, A., & Good, B. (1985). Culture and depression: Studies in anthropology and cross-
cultural psychiatry of affect and disorder. Berkley, CA: University of California Press.
Kohnert, K., Yim, D., Nett, K., Kan, P. F., & Duran, L. (2005). Intervention with linguistically
diverse preschool children: A focus on developing home language(s). Language, Speech, and
Hearing Services in the Schools, 36, 251–263.
Krueger, R. F., & Markon, K. E. (2006). Understanding psychopathology: Melding behavior genet-
ics, personality, and quantitative psychology to develop an empirically based model. Current
Directions in Psychological Science, 15(3), 113–117.
Laguzzi, A. L. (2014). Bilingüismo y psicoterapia. Psicodebate, 14, 33–44.
Leff, J. (1981). Psychiatry around the globe: A transcultural view. New York, NY: Marcel Dekker.
Levy, R. (1983). Self and emotions. Ethos, 11, 128–134.
Lewis-Fernandez, R., & Kleinman, A. (1994). Culture, personality, and psychopathology. Journal
of Abnormal Psychology, 103, 67–71.
Lindenmayer, J. P., & Khan, A. (2006). Psychological disorder. In J. A. Lieberman, T. S. Stroup,
& D. O. Perkins (Eds.), Textbook of schizophrenia (pp. 187–222). Washington, DC: American
Psychiatric.
Livesley, W. J., Schroeder, M. L., Jackson, D. N., & Jang, K. L. (1994). Categorical distinctions
in the study of personality disorder: Implications for classification. Journal of Abnormal
Lucio, G. M. E., Reyes-Lagunes, I., & Scott, R. L. (1994). MMPI–2 for Mexico: Translation and
adaptation. Journal of Personality Assessment, 63, 105–116.
Lukianowicz, N. (1962). Auditory hallucinations in polyglot subjects. Psychiatria et Neurologia,
143, 274–294.
Lysaker, P. H., Olesek, K. L., Warman, D. M., Martin, J. M., Salzman, A. K., Nicolo, G., et al.
(2011). Metacognition in schizophrenia: Correlates and stability of deficits in theory of mind
and self-reflectivity. Psychiatry Research, 190, 18–22.
Mahler, M. S., Pine, F., & Bergman, A. (1975). The psychological birth of the human infant.
New York, NY: Basic Books.
Malgady, R. G., & Costantino, G. (1998). Symptom severity in bilingual Hispanics as a
function of clinician ethnicity and language of interview. Psychological Assessment, 10,
120–127.
Malgady, R. G., Rogler, L. H., & Constantino, G. (1987). Ethnocultural and linguistic bias in men-
tal health evaluation of Hispanics. American Psychologist, 42, 228–234.
Malgady, R. G., & Zayas, L. H. (2001). Cultural and linguistic considerations in psychodiagnosis
with Hispanics: An empirically informed process model. Social Work, 46, 39–49.
Malo Ocejo, P., Medrano Albeniz, J., & Uriarte Uriarte, J. J. (1991). Alucinaciones auditivas en
sujetos bilingües [Auditory hallucinations in bilingual subjects]. Archivos de Neurobiologiá,
54, 15–19.
Marcos, L. R. (1976). Bilinguals in psychotherapy: Language as an emotional barrier. American

Journal of Psychotherapy, 30, 552–560.
Marcos, L.R., Alpert, M., Urcuyo, L. & Kesselman, M. (1973). The effect of interview language
on the evaluation of psychopathology in Spanish-American schizophrenic patients American
Journal of Psychiatry, 130, 549–553.
Marcos, L. R., Urcuyo, L., Kesselman, M., & Alpert, M. (1973). The language barrier in evaluating
Spanish-American patients. Archives of General Psychiatry, 29, 655–659.
Marcus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion,
and motivation. Psychological Review, 98, 224–253.
Marian, V., & Kaushanskaya, M. (2004). Self-construal and emotion in bicultural bilinguals.
Journal of Memory and Language, 51, 190–201.
Marian, V., & Neisser, U. (2000). Language-dependent recall of autobiographical memories.
Journal of Experimental Psychology: General, 129, 361–368.
Marton, K. A., Abramoff, B., & Rosenzweig, S. (2005). Social cognition and language in children
with specific language impairment (SLI). Journal of Communication Disorders, 38, 143–162.
Mattulis, A. C. (1977). Schizophrenia: Experiment in teaching a new foreign language to inpa-
tients as an analeptic ego aid. Dynamische Psychiatrie, 10, 459–472.
McGilloway, A., Hall, R. E., Lee, T., & Bhui, K. (2010). A systematic review of personality dis-
order, race, and ethnicity: Prevalence, aetiology, and treatment. BMC Psychiatry, 10, 1–14.
Mesquita, B., & Ellsworth, P. (2001). The role of culture in appraisal. In K. R. Scherer, A. Schorr,
& T. Johnstone (Eds.), Appraisal processes in emotion: Theory, methods, research (pp. 233–
248). New York, NY: Oxford University Press.
Montgomery, T., & Orozco, S. (1985). Mexican Americans’ performance on the MMPI as a func-
tion of level of acculturation. Journal of Clinical Psychology, 41, 203–212.
Moretti, R., Bava, A., Torre, P., Antonello, R. M., Zorzon, M., & Zivadinov, R. (2001). Bilingual
aphasia and subcortical-cortical lesions. Perceptual and Motor Skills, 92, 803–814.
Morey, L. C. (2007). The Personality Assessment Inventory: Professional manual. Lutz, FL:
Psychological Assessment Resources.
Muller-Vahl, K. R. (2012). Monolingual coprolalia in bilingual patients with Tourette syndrome.
Movement Disorders, 27, 1468.
Nardelli, A. (2014). Most Europeans can speak multiple languages: UK and Ireland not so much.
Retrieved from http://www.theguardian.com/news/datablog/2014/sep/26/europeans-multiple-
languages-uk-ireland
Oller, D. K., Pearson, B. Z., & Cobo-Lewis, A. B. (2007). Profile effects in early bilingual lan-
guage and literacy. Applied PsychoLinguistics, 28, 191–230.
Opitz, B., & Degner, J. (2012). Emotionality in a second language: It’s a matter of time.
Oquendo, M. A. (1996). Psychiatric evaluation and psychotherapy in the patient’s second lan-
guage. Psychiatric Services, 47, 614–618.
Palaniyappan, L., & Liddle, P. F. (2012). Aberrant cortical gyrification in schizophrenia: A surface-
based morphometry study. Journal of Psychiatry and Neuroscience, 37, 399–406.
Paradis, J., Genesee, F., & Crago, M. B. (2011). Dual language development and disorders: A
handbook on bilingualism and second language learning (2nd ed.). Baltimore, MD: Brookes.
Paradis, M. (2004). A neurolinguistic theory of bilingualism. Amsterdam: John Benjamins.
Paradis, M. (2008). Bilingualism and neuropsychiatric disorders. Journal of Neurolinguistics, 21,
199–230.
Pease-Alvarez, L., & Vasquez, O. (1994). Language socialization in ethnic minority communities.
In F. Genesee (Ed.), Educating second language children: The whole child, the whole commu-
nity (pp. 82–102). Cambridge, UK: Cambridge University Press.
Pennebaker, J. W., Francis, M. E., & Booth, R. J. (2001). Linguistic Inquiry and Word Count
(LIWC): A computerized text analysis program (2nd ed.). Mahwah, NJ: Lawrence Earlbaum.
Pennebaker, J. W., Mehl, M. R., & Niederhoffer, K. G. (2003). Psychological aspects of natural
language use: Our words, our selves. Annual Review of Psychology, 54, 547–577.
Pérez Foster, R. M. (1996). Assessing the psychodynamic function of language in the bilingual
patient. In R. M. Perez Foster, M. Moskowitz, & R. A. Javier (Eds.), Reading across boundar-
ies of culture and class (pp. 246–263). Northvale, NJ: Jason Aronson.
Piaget, J. (1980). Experiments in contradiction. Chicago, IL: University of Chicago Press.
Pincay, I. E. M., & Guarnaccia, P. J. (2007). “It’s like going through an earthquake”: Anthropological
perspectives on depression among Latino immigrants. Journal of Immigrant and Minority
Mental Health, 9, 17–28.
Portes, A., & Hao, L. E. (1998). E pluribus unum: Bilingualism and loss of language in the second
generation. Sociology of Education, 71, 269–294.
Postert, C., Dannlowski, U., Muller, J. M., & Konrad, C. (2012). Beyond the blues: Toward a cross-
cultural phenomenology of depressed mood. Psychopathology, 45, 185–192.
Price, C. S., & Cuellar, I. (1981). Effects of language and related variables on the expression of
psychopathology in Mexican-American psychiatric patients. Hispanic Journal of Behavioral
Sciences, 3, 145–160.
Pyszcynski, T., & Greenberg, J. (1987). Self-regulatory perseveration and the depressive self-
focusing style: A self-awareness theory of depression. Psychological Bulletin, 102, 122–138.
Rait, G. (1999). Counting heads may mask cultural and social factors. British Medical Journal,
318, 302–305.
Ramírez-Esparza, N., Gosling, S. D., Benet-Martínez, V., Potter, J. P., & Pennebaker, J. W. (2006).
Do bilinguals have two personalities? A special case of cultural frame switching. Journal of
Research in Personality, 40, 99–120.
Ratner, N. B. (1997). Atypical language development. In J. B. Grleason (Ed.), The development of
language (pp. 348–397). Needham Heights, MA: Allyn & Bacon.
Reidy, T. J., Sorensen, J. R., & Davidson, M. (2016). Testing the predictive validity of the
Personality Assessment Inventory (PAI) in relation to inmate misconduct and violence.
Rogler, L. H. (1993). Culture in psychiatric diagnosis: An issue of scientific accuracy. Psychiatry,
56, 324–327.
Rude, S. S., Gortner, E., & Pennebaker, J. W. (2004). Language use of depressed and depression-
vulnerable college students. Cognition and Emotion, 18, 1121–1133.
Ryan, C. (2013). Language use in the United States: 2011. American community survey reports..
Retrieved from https://www.census.gov/prod/.../acs-22.pdf
Sandoval, J., & Duran, R. P. (1998). Language. In E. Jonathan et al. (Eds.), Test interpretation
and diversity: Achieving equity in assessment. Washington, DC: American Psychological
Association.
Santiago-Rivera, A. L., & Altarriba, J. (2002). The role of language in therapy with the Spanish-
English bilingual client. Professional Psychology: Research and Practice, 33, 30–38.
Schoeman, R., Chiliza, B., Emsley, R., & Southwood, F. (2007). Bilingualism and psychosis: A
case report. Schizophrenia Research, 103, 333–335.
Schrauf, R. W. (2000). Bilingual autobiographical memory: Experimental studies and clinical
cases. Culture and Psychology, 6, 387–417.
Seung, H.-K., Siddiqi, S., & Elder, J. H. (2006). Intervention outcomes of a bilingual child with
autism. Journal of Medical Speech–Language Pathology, 14, 53–63.
Simsek, O. F., & Cerci, M. (2013). Relationship of the gap between experience and language
with mental health in adolescence: The importance of emotion regulation. The Journal of
Psychology: Interdisciplinary and Applied, 147, 293–309.
Skodol, A. E., Oldham, J. M., Bender, D. S., Dyck, I. R., Stout, R. L., Morey, L. C., et al. (2005).
Dimensional representations of DSM-IV personality disorders: Relationships to functional
impairment. The American Journal of Psychiatry, 162, 1919–1925.
Smit, M., Conradie, S., & Schoeman, R. (2011). A grammatical analysis of the spontaneous L2
English use of schizophrenic bilinguals compared to typical bilinguals. Southern African
Linguistics and Applied Language Studies, 29, 505–513.
Sommer, I. E. C., Diederen, K. M. J., Blom, J. D., Wilems, A., Kushan, L., Slotema, K., et al.
(2008). Auditory verbal hallucinations predominantly activate the right inferior frontal area.
Brain, 131, 3169–3177.
Southwood, F., Schoeman, R., & Emsley, R. (2009). Bilingualism and psychosis: A linguistic
analysis of a patient with differential symptom severity across languages. Southern African
Linguistics and Applied Language Studies, 27, 163–171.
Statistics Canada (2011). Linguistic characteristics of Canadians. Retrieved from http://www12.
statcan.gc.ca/census-recensement/2011/as-sa/98-314-x/98-314-x2011001-eng.cfm
Steiner, G. (1992). After Babel: Aspects of language and translation. Oxford, UK: Oxford
University Press.
Stirman, S. W., & Pennebaker, J. W. (2001). Word use in poetry of suicidal and non-suicidal poets.
Psychosomatic Medicine, 63, 517–522.
Sullivan, H. S. (1953). The interpersonal theory of psychiatry. New York, NY: Norton.
Tabors, P. O. (1997). One child, two languages: A guide for preschool educators of children learn-
ing English as a second language. Baltimore, MD: Brookes.
Tager-Flusberg, H. (1992). Autistic children’s talk about psychological states: Deficits in the early
acquisition of a theory of mind. Child Development, 63, 161–172.
Theron, J., Conradie, S., & Schoeman, R. (2011). Pragmatic assessment of schizophrenic bilinguals’
L1 and L2 use. Southern African Linguistics and Applied Languages Studies, 29, 515–531.
Tomkins, S. S., & Tomkins, E. J. (1947). The Thematic Apperception Test: The theory and tech-
nique of interpretation. New York City, NY: Grune & Stratton.
Toppelberg, C. O., & Collins, B. A. (2010). Language, culture, and adaptation in immigrant chil-
dren. Child and Adolescent Psychiatric Clinics of North America, 19, 697–717p.
Toppelberg, C. O., Munir, K., & Nieto-Castañón, A. (2006). Spanish-English bilingual children
with psychopathology: Language deficits and academic language proficiency. Child and
Adolescent Mental Health, 11, 156–163.
Toppelberg, C. O., Nieto-Castañón, A., & Hauser, S. T. (2006). Bilingual children: Cross-sectional
relations of psychiatric syndrome severity and dual language proficiency. Harvard Review of
Psychiatry, 14, 16–29.
Toppelberg, C. O., Snow, C. E., & Tager-Flusberg, H. (1999). Severe developmental disorders and
bilingualism: Clinical perspectives. Journal of the American Academy of Child and Adolescent
Psychiatry, 38, 1197–1199.
Tse, L. (2001). Resisting and reversing language shift: Heritage-language resilience among U.S.
native biliterates. Harvard Educational Review, 71, 676–709.
Velásquez, R. J., Chavira, D. A., Karle, H. R., Callahan, W. J., Garcia, J. A., & Castellanos,
J. (1997). Assessing bilingual and monolingual Latino students with translations of the MMPI-
2: Initial data. Cultural Diversity and Ethnic Minority Psychology, 6, 65–72.
Velásquez, R. J., Maness, P. J., & Anderson, U. (2002). Culturally competent assessment of Latino
clients: The MMPI-2. In J. N. Butcher (Ed.), Clinical personality assessment (2nd ed., pp.
154–170). New York: Oxford University Press.
Velásquez, R. J., Garrido, M., Castellanos, J., & Burton, M. P. (2004). Culturally competent assess-
ment of Chicana/os with the Minnesota Multiphasic Personality Inventory-2. In R. J. Velasquez,
L. M. Arellano, & B. McNeill (Eds.), The handbook of Chicano psychology (pp. 153–174).
Mahwah, NJ: Lawrence Erlbaum Associates.
Vygotsky, L. S. (1962). Thought and language. Cambridge, MA: Massachusetts Institute of
Technology Press.
Wang, J. H., Morales, O., & Hsu, L. K. G. (1998). Auditory hallucinations in bilingual immigrants.
The Journal of Nervous and Mental Disease, 186, 501–503.
Westernmeyer, J., & Janca, A. (1997). Language, culture, and psychopathology: Conceptual and
methodological issues. Transcultural Psychiatry, 34, 291–311.
Whitworth, R. H. (1988). Anglo- and Mexican-American performance on the MMPI administered
in Spanish or English. Journal of Clinical Psychology, 44, 891–897.
Widiger, T. A., & Simonsen, E. (2005). Alternative dimensional models of personality disorder:
Finding a common ground. Journal of Personality Disorders, 19, 110–130.
Williams, R., & Hunt, K. (1997). Psychological distress among British South Asians: The contri-
bution of stressful situations and subcultural differences in the West of Scotland. Psychological
Medicine, 27, 1171–1181.
Yang, J., Bagby, R. M., Costa, P. T., Ryder, A. G., & Herbst, J. H. (2002). Assessing the DSM-IV
structure of personality disorder with a sample of Chinese psychiatric patients. Journal of
Yang, J., McCrae, R. R., Costa, P. T., Dai, X., Yao, S., Cai, T., et al. (1999). Cross-cultural personal-
ity assessment in psychiatric populations: The NEO-PI-R in the People’s Republic of China.
Yang, J., McCrae, R. R., Costa, P. T., Yao, S., Dai, X., Cai, T., et al. (2000). The cross-cultural
generalizability of Axis-II constructs: An evaluation of two personality disorder assessment
instruments in the People’s Republic of China. Journal of Personality Disorders, 14, 249–263.
Yanping, Z., Leyi, X., & Qijie, S. (1986). Styles of verbal expression of emotional and physical
experiences: A study of depressed patients and normal controls in China. Culture, Medicine,
and Psychiatry, 10, 231–243.
Yeates, K. O., Schultz, L. H., & Selman, R. L. (1991). The developmental interpersonal negotiation
strategies in thought and action: A social-cognitive link to behavioural adjustment and social
status. Merrill-Palmer Quarterly, 37, 369–405.
Zadeh, Z. Y., Im-Bolter, N., & Cohen, N. J. (2007). Social cognition and externalizing psycho-
pathology: An investigation of the mediating role of language. Journal of Abnormal Child
Chapter 11
Personality Traits in Bilinguals
Contents
Introduction 259
Personality Dimensions in Young Spanish–English Bilinguals 262
Results 264
Conclusions 265
References 266
Introduction
Speaking two or more languages influence the way we feel and behave. Studies have
found that bilinguals frequently report feeling like a different person when speaking
the first language (L1) compared to the second language (L2; Dewaele, 2004, 2008);
this is particularly true when the two languages correspond to cultures different in
values (Ramirez-Esparza, Gosling, Benet-Martinez, Potter, & Pennebaker, 2006).
Thus, learning two or more languages may be associated with marked differences in
behavior, emotions, and personality, depending on which language the bilingual is
expressing at the moment of the assessment (Ramirez-Esparza et al., 2006; See
Rosselli, Vélez-Uribe & Ardila; Terrazas-Carrillo, this volume). This chapter reviews
the evidence that supports the association between the bilingual experience and
M. Roselli (*)
e-mail: mrossell@fau.edu
I. Vélez-Uribe
Florida Atlantic University, Davie, FL, USA
A. Ardila

personality dimensions and presents a new study conducted with Spanish–English

bilinguals from South Florida.
In individuals who speak two languages, the switching of languages is accompa-
nied by switching in other aspects of their behavior. Bicultural individuals, those
raised under the influence of two cultures, behave according to one culture or to the
other depending on the context they are in. This phenomenon of switching between
cultures has been interpreted as a cultural frame switching effect (CFS; Hong, Chiu,
& Kung, 1997; Hong, Ip, Chiu, Morris, & Menon, 2001), where bicultural individuals
shift values and attributions in the presence of culture-relevant stimuli. This switching,
which can be considered as homologous to code-switching in language, occurs in
individuals who are bicultural and who have internalized the two cultures’ values,
feelings and actions (Hong et al., 2001; LaFromboise, Coleman, & Gerton, 1993) and
would imply high levels of linguistic and cultural proficiency. Since bilinguals are
frequently bicultural, it is not surprising that they exhibit this switching phenomena.
Dewaele and van Oudenhoven (2009) analyzed the combined influence of culture and
language over personality traits in a study with multicultural teenagers in London. The
Multicultural Personality Questionnaire (MPQ; van Oudenhoven & van der Zee,
2002) was used to assess participants in five dimensions (Cultural Empathy, Open-
Mindedness, Social Initiative, Emotional Stability, and Flexibility). Their findings
indicated that multicultural teenagers who were born abroad but had settled down in
London during their childhood, scored higher on the dimension of Open-Mindedness
(defined as the open and unbiased attitude towards members of a different culture),
and Cultural Empathy (the overall ability to empathize with the emotional aspects of
persons from a different culture). In an opposite manner, they scored lower on
Emotional Stability (referring to the level of calmness or strong emotional reactions
during stress provoking situations) compared to teenagers born and raised in London.
In addition, these authors found that language experience had an effect on the person-
ality profile; with participants who were dominant in more than one language (i.e.,
multi-dominance), scoring significantly higher on Open-Mindedness, marginally
higher on Cultural Empathy and significantly lower on Emotional Stability than par-
ticipants dominant in only one language. The authors’ findings support the notion that
personality is shaped by social factors such as multiculturalism and language domi-
nance. Higher multicompetence scores correlated to higher levels of Open-
Mindedness, while lower multicompetence scores correlated to lower levels of
Emotional Stability. The correlation between the scores and Cultural Empathy was
marginally positive. Therefore, it is reasonable to conclude that the experience of
adapting to a new environment and being in contact with different languages and cul-
tures results in higher levels of Cultural Empathy and Open-Mindedness.
Ramirez-Esparza et al. (2006) demonstrated the direct influence of bilingualism
on personality traits in a group of Spanish–English bilingual adults living in Texas,
when comparing their The Big Five Inventory (BFI) scores in both languages. The
BFI is a widely used 44-item scale that classifies participants based on five dimen-
sions of personality (openness, extraversion, agreeableness, neuroticism, and consci-
entiousness), with valid and reliable versions available in English and Spanish
(Benet-Martinez & John, 1998; John, Donahue, & Kentle, 1991). Since this study
comprises a very large sample size (i.e., 168,451 in the US and 1031 in Mexico),
11 Personality Traits in Bilinguals 261
which biases the analysis resulting on significant differences in all comparisons, the
authors focused the report only on the directionality of the differences. Bilinguals
in Ramirez-Esparza et al.’s (2006) study were more extroverted, agreeable, and
conscientious in English than in Spanish and these differences were consistent with
the personality displayed by previously studied monolinguals in each culture. The
cross-language personality differences for Neuroticism were relatively small but
scores were higher in Spanish. The differences for Openness showed higher scores in
Spanish than in English, but they were not consistent with the cross-cultural differ-
ences reported in monolinguals in which scores were higher in English speakers.
Interestingly, the authors found a strong positive correlation between the Spanish and
English versions of the personality questionnaire suggesting that individuals tended
to retain their rank ordering within a group, but the group as a whole changed. In
other words, an extrovert bilingual does not suddenly become an introvert as she/he
switches languages; instead, a bilingual becomes more extroverted when speaking in
English rather than in Spanish, remaining in the same rank order within each lan-
guage/culture of the groups. In the interpretation of their results, Ramirez-Esparza
et al. (2006) included language experience changes over time as a relevant variable,
and even a possible confounding variable on personality scores. Perhaps, according
to the authors, some of bilinguals spent a significant part of their early lives in
Spanish-speaking environments, later becoming bilingual when learning English. If
this were true, the fact that a participant has one personality in one language and
another personality in the other language, would be more a function of age-related
personality differences, than a function of culture, correlating to the corresponding
language used during different life periods. Therefore, it could be possible that their
responses in Spanish would reflect their childhood personality and experiences and
their responses in English would reflect their adult personality and experiences.
Ramirez-Esparza et al.’s (2006) sample however, comprised very balanced bilinguals
who were active in the use of English and Spanish at the time of the assessment rul-
ing out this interpretation for their sample. The authors also excluded the possibility
of the influence of differences in the translations of the Spanish and English versions
of the BFI on their personality scores. An item analysis of both versions showed no
item bias (i.e., only one item from the Openness scale behaved anomalously) which
lead the authors to conclude that the cross-language differences found in the bilin-
guals’ personality subscales could not be attributed to translation differences of the
assessment inventory. This conclusion is consistent with the results reported and the
methodology applied by Bennet-Martinez and John (1998) for the translation of the
BFI and the consequent validity and reliability analyses.
Veltkamp, Recio, Jacobs, and Conrad (2012) wanted to investigate if the differ-
ences found by Ramirez-Esparza et al. (2006) in early bilinguals would extend to late
bilinguals. However, their study not only contrasted two languages but also two dif-
ferent cultures. Participants were late German–Spanish bilinguals who scored higher
in extraversion and neuroticism when tested in Spanish, and higher in agreeability in
German in the Neuroticism Extraversion Openness-Five Factor Inventory (NEO-
FFI), which similarly to the BFI, assesses five dimensions of personality (openness,
extraversion, agreeableness, neuroticism, conscientiousness), including 60 items

equally divided among subscales. These results are partially consistent with the study
of Ramirez-Esparza et al. (2006), with similar effects in the dimensions of neuroticism
(higher in Spanish than in German), with an inverted effect in the dimension of extra-
version, which in the Ramirez-Espaza et al. study (2006) was higher in English than
in Spanish. Comparing the results from these two studies could be problematic
because the samples are not completely equivalent: early Spanish–English and late
German–Spanish bilinguals. Veltkamp et al. (2012) considered it possible that these
differences could respond to the need for adaptation to the main culture while grow-
ing up within a minority subcultural group. According to the authors, their results
continue to support the cultural frame switch hypothesis.
More recently, Gonzalez, Vélez-Uribe and Rosselli (2015) presented preliminary
results of a study in which Spanish–English bilinguals were asked whether they felt
like a different person when speaking Spanish or English; 42.5% of the bilingual
participants responded yes to this question. All the participants included in the sam-
ple were simultaneous bilinguals born in South Florida but exposed to both lan-
guages since childhood or early bilinguals born in Latin America who arrived to the
US before the age of 12 years.
The next section of this chapter describes a study developed to further analyze
the association between bilingualism and personality characteristics in a sample of
young active and proficient Spanish–English bilinguals.
Personality Dimensions in Young Spanish–English Bilinguals
Participants were 134 college students and members of the community (96 females)
who were Spanish–English bilinguals and were living in South Florida. All participants
(Mage = 21.46, SD = 5.65) acquired Spanish as a native language, that is, being exposed
to it since birth except for 11 participants who reported both languages as native. The
mean age of language acquisition of English was 5.29 years (SD = 4.42). All partici-
pants were living in a bilingual environment using Spanish 32% and English 68% of
the day. However, 85% described English as the dominant language. Detailed informa-
tion on the characteristics of the sample is presented on Tables 11.1 and 11.2.
Table 11.1 Sample characteristics

Characteristics n = 134 Minimum Maximum M SD
Age 18.00 44.00 21.46 5.65
Age of arrival to the US (n = 72)a 1.00 37.00 9.66 6.33
Total years in the US (n = 134) 0.50 34.00 16.29 5.65
Total years in US (n = 72)a 0.50 26.00 12.65 4.44
Percent of life spent in the US 2.63 100.00 77.12 25.9
Age of acquisition of English 1.00 30.00 5.29 4.42
Age of acquisition of Spanish 1.00 1.00 1.00 0.00
Total years of education 12.00 27.00 14.71 2.54
Participants born in the USA not included
a
Table 11.2 Sample Count Percent

characteristics: gender,
Gender
education, and language
factors Males 38 28.4
Female 96 71.6
Highest level of education achieved
High school 78 58.3
Associates degree (2 years of college) 43 32.0
Bachelor’s degree (4 years college) 13 9.7
Native language
Spanish 123 91.8
Both 11 8.2
Dominant language
English 114 85.1
Spanish 20 14.9
Table 11.3 Self-reported levels of proficiency and cultural identity subscales

Minimum Maximum M SD
Level of proficiency
Speaking English 4.00 10.00 9.47 1.00
Understanding English 6.00 10.00 9.63 0.72
Reading English 5.00 10.00 9.50 0.91
English proficiency 5.33 10.00 9.32 0.96
Speaking Spanish 2.00 10.00 7.92 1.83
Understanding Spanish 1.00 10.00 8.72 1.49
Reading Spanish 1.00 10.00 7.60 1.95
Spanish proficiency 3.33 10.00 8.37 1.53
Cultural identity subscales
Latino cultural identity 8.00 24.00 20.65 3.40
US cultural identity 6.00 24.00 19.30 3.40
Language proficiency was assessed using the self-rated language experience and
proficiency questionnaire (LEAP-Q; Marian, Blumenfeld, & Kaushanskaya, 2007).
LEAP-Q proficiency scores were divided in three domains (speaking, understand-
ing spoken language, and reading) on a 0–10 scale as follows: 0 = none, 1 = very
low, 2 = low, 3 = fair, 4 = slightly less than adequate, 5 = adequate, 6 = slightly more
than adequate, 7 = good, 8 = very good, 9 = excellent, 10 = perfect. From this ques-
tionnaire, the total mean scores for the two subscales related to English and Spanish
proficiency were used. The proficiency score for each language was obtained by
calculating the average of the three subscales. As seen in Table 11.3, the group was
highly proficient in both languages although higher scores are observed for English,
consistent with the fact that most of their education was received in English.
In addition, each participant received the Abbreviated Multidimensional
Acculturation Scale (AMAS-ZABB) that yields three distinct factors from 42 items:
cultural identity, language competence, and cultural competence in six differenti-

ated subscales (U.S. cultural identity—six items, Latino cultural identity—six
items, English language—nine items, Spanish language—nine items, U.S. cultural
competence—six items, Latino cultural competence—six items). The AMAS-
ZABB has been shown to have good internal reliability and adequate concurrent
validity on a sample of 246 participants composed of Central American immigrants
and native Spanish speakers living in the United States. Scores were given on a scale
of minimum 1 and maximum 4 and calculated individually for each scale (Zea,
Asner-Self, Birman, & Buki, 2003). For the current study two subscales were con-
sidered for sample characterization purposes: US cultural identity and Latino cul-
tural identity. As seen in Table 11.3 the cultural identity scores for the two cultures
were very similar suggesting biculturalism in the sample.
Finally, personality dimensions were evaluated using BFI, which has been shown
to have good consistency and reliability across different age and cultural groups
(John et al., 1991). The Spanish version of the BFI has shown high reliability and
convergence across languages, as well as applicability in bilingual samples within
the United States (Benet-Martinez & John, 1998). The test was administered in both
English (John et al., 1991) and Spanish (Benet-Martinez & John, 1998) by fluent
Spanish–English bilinguals in counterbalanced order.
Results
A 2 (Language: English vs. Spanish) × 5 (BFI subscales: openness, extraversion,

agreeableness, neuroticism, conscientiousness) repeated measures ANOVA, using
the GLM procedure, was conducted. Results indicated a significant main effect of
language of administration, F(1, 133) = 425.59, p < .001, η2 = .76, and of BFI sub-
scale, F(4, 130) = 40.01, p < .001, η2 = .55. Total BFI scores (adding all scales) were
higher in English than in Spanish (see Table 11.4); total BFI subscales (sum of
Spanish and English scores) showed the highest total scores for agreeableness and
the lowest for neuroticism. In addition, a significant interaction between language
and personality scale was detected, F(4, 130) = 42.64, p < .001, η2 = .57. Higher
scores in English compared to Spanish were revealed for extraversion, F(1,
133) = 26.64, p < .001, η2 = .17, agreeableness, F(1, 133) = 238.92, p < 001, η2 = .64,
conscientiousness F(1, 133) = 238.28, p < .001, η2 = .64, and openness, F(1,
133) = 131.02, p < .001, η2 = .5. In an opposite manner, neuroticism scores were
higher in Spanish than in English F(1, 133) = 9.51, p < .01, η2 = .07.
A bivariate correlations analysis was conducted to examine the relationship
between the BFI subscales. Some significant correlations were found between the
English and the Spanish BFI subscales: Extraversion (r = .19; p = .031) and
Neuroticism (r = .30; p = .000) presented significant correlations, with conscien-
tiousness (r = .17; p = .049) and Openness (r = .17; p = .054) being marginally
significant, while agreeableness (r = −.1; p = −.171) did not correlate significantly
between languages.
Table 11.4 BFI descriptive Mean SD

for all subscales in both
English
languages (n = 134)
Extraversion 3.51 0.83
Agreeableness 3.98 0.60
Conscientiousness 3.76 0.60
Neuroticism 2.80 0.77
Openness 3.73 0.60
Spanish
Extraversion 3.14 0.33
Agreeableness 3.02 0.33
Conscientiousness 2.83 0.47
Neuroticism 3.00 0.32
Openness 3.06 0.43
46.2% of our participants (n = 62)
reported feeling like a different per-
son when switching languages, while
53.8% (n = 72) reported not feeling
any different
Additional correlations analyses were conducted to examine the influence of

acculturation on the personality subscale scores; correlations between each of the
AMAS-ZABB subscales (Latino Cultural Identity and US Cultural Identity) and the
five BFI scores in all five subscales were performed. Neuroticism in English was
significantly correlated with US cultural identity (r = −.21) meaning that higher
neuroticism was associated with less cultural identity to the US culture; no other
significant correlations were found.
Conclusions
The results of this study, as well as previous ones provide support to the effect of
language of administration on the results of personality inventories as reflected in
the differences in the personality profile of Spanish–English bilinguals. Participants
scored higher in levels of extraversion, agreeableness, conscientiousness, and open-
ness when the questions were in English and higher in levels of neuroticism when
the questions were in Spanish. These results are comparable to the findings by
Ramirez-Esparza et al. (2006) in a bilingual sample from two states in the United
States. They also found more extraversion, agreeableness, and consciousness in
English than in Spanish and higher levels of neuroticism in Spanish than in English.
Different from the Ramirez-Esparza et al. study, results from our study showed
higher scores in openness in English compared to Spanish. The English and Spanish
Openness score in our bilingual sample was 3.73 and 3.06 respectively compared to
4.00 and 4.25 in the Texas sample described by Ramirez-Esparza et al. (2006). The
correlational analysis of the relationship between the scales in both languages
indicated that higher levels of extraversion, neuroticism, conscientiousness, and

openness in Spanish corresponded to higher levels in English in all four scales.
However, higher levels of agreeableness in Spanish corresponded to lower levels in
English.
As mentioned above, the question of whether scoring differently in personality
measures in the bilinguals’ two languages can be attributed to actual personality
differences remains to be seen. In the current sample, 46.2% of our participants
(n = 62) reported feeling like a different person when switching languages, while
53.8% (n = 72) reported not feeling any different. However, feeling different did not
correlate significantly with scores in any of the subscales in either language, which
might indicate either a subjective feeling induced by other factors, or to limitations
in the ability to notice subtle changes in behavior, rather to the lack of them
(Ożańska-Ponikwia, 2012).
Dewaele and van Oudenhoven (2009) found that being dominant in more than
one language seemed to heighten some personality features (open-mindedness and
cultural empathy). When we tested the effect of language dominance on the differ-
ent scales, however, we found that it did not exert a significant influence and the
directionality of the differences did not change. So, this question remains unsettled
and controversial.
Future research is clearly needed in larger samples controlling for different cul-
tural and linguistic factors as well as conditions of second language acquisition and
second culture immersion. The inclusion of additional variables can potentially con-
tribute to the clarification of this question. While it seems clear that there are some
differences in the results of personality inventories depending on the language of
administration, the extent to which personality traits are modulated by language is
unclear. Perhaps it will be necessary to develop new methods to attempt to measure
more precisely the extent of the influence of cultural factors.
Acknowledgement Our gratitude to Deven Christopher for her editorial support and valuable
comments.
References
Benet-Martinez, V., & John, O. (1998). Los cinco grandes across cultures and ethnic groups:
Multitrait multimethod analysis of the big five in Spanish and English. Journal of Personality
and Social Psychology, 75(3), 729–750.
Dewaele, J. (2004). The emotional force of swearwords and taboo words in the speech of multilin-
guals. Journal of Multilingual and Multicultural Development, 25(2–3), 204–222.
Dewaele, J. (2008). The emotional weight of “I love you” in multilinguals’ languages. Journal of
Pragmatics, 40(10), 1753–1780.
Dewaele, J. M., & Van Oudenhoven, J. P. (2009). The effect of multilingualism/multicultural-
ism on personality: No gain without pain for third culture kids? International Journal of
Multilingualism, 6(4), 443–459.
Gonzalez, K., Vélez-Uribe, I., & Rosselli, M. (2015). A cross-linguistic analysis of language
effects on personality measures in Spanish-English bilinguals. In 27th Annual meeting of the
Association for Psychological Sciences (APS). New York.
Hong, Y. Y., Chiu, C. Y., & Kung, T. M. (1997). Bringing culture out in front: Effects of cul-
tural meaning system activation on social cognition. In K. Leung, Y. Kashima, U. Kim, &
S. Yamaguchi (Eds.), Progress in Asian social psychology (Vol. 1, pp. 135–146). Singapore:
Wiley.
Hong, Y. Y., Ip, G., Chiu, C. Y., Morris, M. W., & Menon, T. (2001). Cultural identity and dynamic
construction of the self: Collective duties and individual rights in Chinese and American cul-
tures. Social Cognition, 19, 251–268.
John, O. P., Donahue, E. M., & Kentle, R. L. (1991). The Big Five Inventory: Versions 4a and 54,
institute of personality and social research. Berkeley, CA: University of California.
LaFromboise, T., Coleman, H. L., & Gerton, J. (1993). Psychological impact of biculturalism:
Evidence and theory. Psychological Bulletin, 114, 395–412.
Marian, V., Blumenfeld, H. K., & Kaushanskaya, M. (2007). The language experience and profi-
ciency questionnaire (LEAP-Q): Assessing language profiles in bilinguals and multilinguals.
Journal of Speech, Language and Hearing Research, 50, 940–967.
Ożańska-Ponikwia, K. (2012). What has personality and emotional intelligence to do with ‘feeling
different’ while using a foreign language? International Journal of Bilingual Education and
Bilingualism, 15(2), 217–234.
Ramirez-Esparza, N., Gosling, S., Benet-Martinez, V., Potter, J., & Pennebaker, J. (2006). Do bilin-
guals have two personalities? A special case of culture frame switching. Journal of Research
in Personality, 40, 99–120.
van Oudenhoven, J. P., & van der Zee, K. I. (2002). Predicting multicultural effectiveness of
international students: The Multicultural Personality Questionnaire. International Journal of
Intercultural Relations, 26(6), 679–694.
Veltkamp, G. M., Recio, G., Jacobs, A. M., & Conrad, M. (2012). Is personality modulated by lan-
guage? International Journal of Bilingualism, 17 (4), 496–504. doi:10.1177/1367006912438894.
Zea, M. C., Asner-Self, K. K., Birman, D., & Buki, L. P. (2003). The abbreviated multidimentional
acculturation scale: Empirical validation with two Latino/Latina samples. Cultural Diversity
and Ethnic Minority Psychology, 9, 107–126.
Chapter 12
Cognitive Abilities in Bilinguals in L1 and L2
Andy V. Pham, Sara Castro-Olivo, Heejung Chun, and Anisa N. Goforth
Contents
ognitive Abilities in Bilinguals in L1 and L2
C 269
Models of Bilingualism 271
Cognitive Test Performance in L1 and L2 272
Cognitive Testing of Bilingual Children in School and Clinical Settings 273
L1 and Bilingual Cognitive Assessment 275
L2 Cognitive Assessment 275
Nonverbal Cognitive Assessment 278
Best Practices in Cognitive Assessment of Bilinguals 280
Conclusion and Future Directions 287
References 288
Cognitive Abilities in Bilinguals in L1 and L2
It is estimated that ethnic minorities make up one third of the total U.S. population
and that they will become the majority by 2050 (U.S. Census Bureau, 2013).
National census data also show that over 20% of the U.S. population speaks a
A.V. Pham, Ph.D. (*)

e-mail: avpham@fiu.edu
S. Castro-Olivo, Ph.D.
Texas A&M University, College Station, TX, USA
e-mail: s.castro-olivo@tamu.edu
H. Chun, Ph.D
New Mexico State University, Las Cruces, NM, USA
e-mail: chun@nmsu.edu
A.N. Goforth, Ph.D.
University of Montana, Missoula, MT, USA
e-mail: Anisa.Goforth@mso.umt.edu

270 A.V. Pham et al.
language other than English at home, albeit with varying degrees of linguistic
fluency in both native language (L1) and the second language (L2), or in most cases,
English. From these households, three out of four report speaking English well
or very well (Shin & Ortman, 2011). Among individuals ages five or older, the
languages most frequently spoken at home are: Spanish (14.5% of the overall U.S.
population), Chinese (0.9%); French (0.7%), Tagalog (0.5%), and Vietnamese
(0.5%; Shin & Ortman, 2011). Due to these rapidly changing demographics,
this can present unique challenges when assessing bilingual or multilingual popula-
tions, along with providing culturally competent services across various settings
(e.g., pediatric/adult, clinical/school).
With increasing percentages of multilingual families in the U.S., children in these
households are also likely to develop varying linguistic proficiencies depending on
age of acquisition, setting (e.g., home, school), and the amount of exposure to the
language(s) spoken by parents, teachers, and peers. The effects of second language
acquisition on academic and cognitive development of bilingual children and
English-language learners (ELLs) are therefore complex. In particular, Hispanics
are considered one of the fastest growing ethnic groups in the nation (U.S. Census,
2013); however, the Spanish-speaking community in the U.S. is comprised of a het-
erogeneous group with variability in country of origin, language dialect and profi-
ciency, educational attainment, immigration and acculturation patterns, religion, and
other sociocultural factors. This may also include those who speak Spanglish, which
is described as loosely borrowing and combining English or Spanish words, and/or
code-switching (Poplack, 1980). Educators and researchers, therefore, must consider
bilingual development in the context of both neurocognitive and sociocultural per-
spectives when assessing children’s cognitive abilities.
It is not uncommon for psychologists and neuropsychologists to use poorly
translated tests or employing interpreters to administer English-language tests
along with using norms obtained from English-only speaking populations.
However, significant efforts have been made to improve the quality and utility of
cognitive and neuropsychological tests used for assessing bilingual individuals or
those with limited English proficiency. For example, many current neuropsycho-
logical tests have been developed, adapted, and modified particularly for Spanish-
speaking adults (Strutt, Burton, Resendiz, & Peery, 2016). Despite these efforts,
however, many of them lack specificity or generalizable normative data. This is in
part due to the significant degree of variability that exists among Spanish sub-
groups (e.g., Mexican, Cuban, Spaniards). At the same time, obtaining geographi-
cal distribution of each Spanish subgroup residing in specific regions of the U.S.
is nearly impossible.
The purpose of this chapter is to present the issues and research pertaining to
assessment of cognitive abilities of bilingual children. In addition to practical
considerations, theoretical models of bilingualism are discussed in relation to L1
and L2 acquisition. We outline the issues pertaining to bilingualism in education,
specifically addressing sociocultural factors (e.g., acculturation) contributing to
the discrepancies in cognitive test performance of bilingual children. Then, we
highlight some specific tests that assess cognitive abilities of bilinguals (notably
12 Cognitive Testing 271
English–Spanish) used in clinical and school settings. Finally, we provide rec-

ommendations for best practices in test selection and cognitive assessment of
bilingual children.
Models of Bilingualism
Bilingual children are the majority in many countries around the world, yet exploration
of this topic is still emerging in the U.S. Although the neural mechanisms of learning
two languages remain unclear, several proposed theoretical models have attempted
to explain L1 and L2 acquisition. One supported model, the dual language system
hypothesis (Genesee, 1989), suggests that bilinguals have separate linguistic systems
allowing them to develop two sets of vocabulary, or one set from each language.
Consequently, children can fall under two types of bilingual groups, simultaneous
and sequential/successive, depending on when second language acquisition occurs
(Paradis, Genesee, & Crago, 2011).
Simultaneous acquisition of both languages (L1 and L2, or both L1 in some
cases) can occur as early as birth, when two languages develop concurrently; this
process does not significantly differ from single language acquisition (Paradis et al.,
2011). Parental knowledge and command of both languages at home often deter-
mine whether a young child would also demonstrate simultaneous bilingualism.
Although many speculate that these children would exhibit expressive language
delays due to language mixing or language confusion, current research suggests that
there are no differences in vocabulary size when both languages are considered as
opposed to a single language (Hoff et al., 2011), and this finding is supported by the
dual language system hypothesis (Genesee, 1989). Some studies also purport that
bilingual children pass through the same phases and milestones of language devel-
opment similar to monolingual children (e.g., Genesee & Nicoladis, 2007), and
demonstrate executive control advantages, including cognitive flexibility, response
inhibition, attentional control, and metalinguistic awareness (Bialystok, 2010;
Bialystok & Craik, 2010). However, some critics assert these cognitive advantages
are nonexistent, cannot be replicated in other studies, or only occur in restricted
circumstances (Paap, Johnson, & Sawi, 2015). Nevertheless, those who demon-
strate simultaneous language acquisition may also exhibit balanced bilingualism, or
relatively equal knowledge and command of both languages, often with near native
fluency. However, it is relatively rare for children, as well as for adults, to demon-
strate balanced bilingualism, as most bilinguals typically have a preferred or domi-
nant language. This phenomenon is also contextually dependent on peers or adults
to whom bilinguals communicate in their environment; in other words, b ilinguals
are likely to exhibit balanced bilingualism only when there is equal exposure to two
languages.
Sequential or successive bilinguals, on the other hand, develop L1 during infancy,
and then L2 at a later point in their development, typically by 3 years of age or older
(Valdes and Figueroa, 1994). Two different theories have been proposed to account
for why and how the timing of L2 exposure affects bilinguals’ language acquisition
and development. The critical period hypothesis (Johnson & Newport, 1989) suggests
that there are different outcomes for early and late bilinguals, since late bilinguals
(after age 6) often do not acquire native-like fluency in L2. On the other hand,
Hernandez et al. (2005) suggested that differences between early bilinguals and late
bilinguals are due to the competition and entrenchment hypothesis, which highlights
the L2 acquisition process as dependent on brain plasticity and first-language
entrenchment. For example, children’s earliest L1 exposure is represented neuro-
logically as phonological representations. Early bilinguals who are exposed to two
languages from infancy, or simultaneous bilinguals, spend a considerable amount of
time processing language streams from two different languages (Hernandez et al.,
2005). By the time early bilinguals begin speaking L1 and L2, this early exposure
facilitates their ability to keep both languages separate. However, for late bilinguals,
L2 acquisition occurs after L1 has already established an intricate lexical (sound),
syntactic (rules of language) and semantic (word meanings) neural network.
Because L1 has already become entrenched, L2 acquisition primarily occurs with
the assistance of L1. For example, in order for the late bilingual to say a word in L2,
escuela, the child must think of the word in L1, school (Hernandez et al., 2005).
Therefore, words from L2 will pair closely with relevant phonological and represen-
tational information from L1.
Sequential bilingualism is typically seen in children who immigrated to the U.S.,
or who come from a non-English-speaking household and are enrolled in school.
Many school-age children who are ELLs (or those with limited English proficiency)
are also late bilinguals. For these children, L1 is typically the dominant or preferred
language. L2, however, may eventually become the dominant language for some
children over time (Salinas, Bordes-Edgar, and Puente, 2016), to the point where
some children exhibit subtractive bilingualism or L1 attrition (Cummins & Swain,
1986). On the other hand, additive bilingualism occurs when L1 continues to
develop alongside L2 (Cummins & Swain, 1986). As mentioned earlier, bilingual
proficiency is largely dependent on the amount of the L1 and L2 exposure across
environments, the child’s metalinguistic awareness, and the child’s available neural
plasticity to create a separate L2 network with limited L1 interference. It will also
vary across domains (e.g., listening, speaking, reading, writing) and across contexts
(e.g., social vs. academic language) as children start to acquire L2 (e.g., English) in
schools. When children appear to lack proficiency in either L1 or L2, parents and
educators are likely to refer these children for neuropsychological or speech/lan-
guage evaluations as they may suspect potential language or learning delays.
Cognitive Test Performance in L1 and L2
Results from limited research studies are varied in the degree to which bilingual
individuals score differently on cognitive tests compared to monolinguals when
tested in L1 or L2. Cognitive abilities have traditionally been assessed through the
use of intelligence tests based on the general g ability theory, which tend to measure
intelligence through the use of broad cognitive tasks that include verbal and nonver-
bal demands. The use of verbal tests among individuals who are not proficient in the
language (L2 or English) has been criticized due to high cultural loading and lin-
guistic demands (Naglieri & Bornstein, 2003; Suzuki & Valencia, 1997), and has
been found to put individuals from minority cultures and with limited English
proficiency at a significant disadvantage. Additional research is needed in this area
to first determine which cognitive tests (of the ones available) are culture and lan-
guage appropriate for individuals who are not proficient in L1 or L2. To this date,
research has shown that when bilingual individuals are tested using cognitive tests
that have been found to be culturally loaded, they perform significantly worse than
monolingual control groups (Garratt and Kelly 2007; Karlsson et al., 2015).
Most recent research is finding that differences between bilingual and monolin-
gual individuals are not as evident when less culturally or linguistically loaded tests
are used or when the bilingual participants have had sufficient number of years to
gain full proficiency/practice in their L2. For example, research has shown that cog-
nitive performance of bilingual children (age 10) did not differ significantly from
monolingual children on overall cognitive performance, but differences were found
for younger children who have had fewer opportunities to learn L2 (Karlsson, et al.,
2015). Similarly, Jarvis, Danks, and Merriman (1995) found no relationship between
the cognitive performance of bilingual children and monolingual children, using
nonverbal cognitive tests. Naglieri, Otero, DeLauder, and Matto (2007) also found
no significant differences in cognitive abilities in English or Spanish when a pro-
cessing test, with minimal language usage, was employed to test abilities in L1 and
L2 among Hispanic bilingual children.
Among adults, Rosselli et al. (2000) found similar performance between bilin-
gual and monolingual participants on all tests except for semantic verbal fluency
tasks (e.g., word generation, categorical fluency). They further specified that bilin-
guals who learned English before age 12 performed significantly better on the
English repetition test and produced a higher number of words in the description of
a picture than the bilinguals who learned English after age 12. Similarly, Kalia,
Wilbourn, and Ghio (2014) found that individuals who are exposed to two lan-
guages early in life (before age 6) perform similarly to monolingual individuals than
those who are exposed to an L2 later in development (after age 6).
ognitive Testing of Bilingual Children in School and Clinical

C
Settings
One of the areas that have been controversial and understudied is the influence of
dual language functioning on the cognitive development and academic trajectory
(e.g., reading, math) of bilingual children. Historically, sequential bilinguals, and in
particular ELLs, have been overrepresented in special education (Yates, Ortiz &
Anderson, 1998). Researchers have attributed that overrepresentation to biased and
discriminatory cognitive testing practices that inaccurately identified ELLs as
possessing less than normal/average cognitive ability when compared to a
standardized sample (Rhodes, Ochoa & Ortiz, 2005). According to the Individuals
with Disabilities Education Act (IDEA; 2004), a child must not be determined to
have a specific learning disability if the lack of achievement is primarily attributed
to cultural factors or limited English proficiency. Thus, comprehensive psychologi-
cal or neuropsychological evaluations must take into consideration the impact of cul-
ture and language exposure along with L1 and L2 proficiency when examining
learning difficulties. On the other hand, bilingual children may also be underidenti-
fied especially if there are true language disorders or learning difficulties. This pro-
file can be especially difficult to determine, given that sequential bilinguals may
learn to develop L2 slowly and thus be perceived as delayed in language develop-
ment. Generally, a language disorder is diagnosed in bilingual children when there
are significant delays in both L1 and L2.
Nevertheless, research and practice in cognitive assessment have changed over
the years to be more responsive to the language development of bilinguals. Indeed,
these changes have proven to be effective in at least reducing the overrepresentation
of ELLs in special education services, particularly for specific learning disability.
For example, it is estimated that 13% of the overall U.S. public school population
receive some type of special education service; however, only 9% of that population
are ELLs (NCES, 2016). Yet, there are continuing concerns about the achievement
gaps between ELLs and non-ELL counterparts. According to the National
Assessment of Educational Progress (NAEP), since 1998, ELL fourth and eighth
graders have performed worse on annual reading achievement tests than their non-ELL
counterparts (Kena et al., 2015).
Currently, there is no set standard for conducting cognitive assessments on bilin-
gual children. The American Psychological Association (APA, 2010) and the
American Educational Research Association (AERA, 2014) provide statements
within their Codes of Ethics and the Standards for Educational and Psychological
Testing that require only those psychologists who have received specialized training
and have demonstrated competency in administering cognitive assessments to be
allowed to administer and interpret the results of these tests. Although a bilingual
assessment may be considered the ideal approach, it can present logistical chal-
lenges. Primarily, there are a limited number qualified bilingual psychologists in the
U.S. who are specifically trained to conduct bilingual assessments. According to the
National Association of School Psychologists (NASP), only 6.2% of all practicing
school psychologists provide services in a language other than English (Charvat,
2008). Although data of other psychologists (i.e., counseling and clinical) who pro-
vide services in a language other than English are scarce, APA (2010) estimated that
only 5% of all new doctorates in psychology in their most recent Employment
Survey were identified as Latino or Hispanic. Despite lack of current data, it is clear
that the proportion of bilingual practitioners in psychology is considerably lower
than the proportion of bilingual children and families residing in the U.S.
APA (2010) and AERA (2014) require that language proficiency of the examinee
should be considered prior to conducting cognitive assessments. Assessment and
understanding of the examinee’s language proficiency often dictate the selection of
the most appropriate cognitive test(s) to administer and ensure valid results and
interpretations. However, just like any other standardized testing procedure, each
approach is not without its challenges. Thus, it is necessary to document the limita-
tions of the assessment tools when communicating test results. This process prompts
psychologists and neuropsychologists to critically evaluate measures before using
them. Table 12.1 provides a nonexhaustive list of available language and cognitive
tests that can be used to develop a test battery for children of different ages. The fol-
lowing section will discuss the strengths and limitations of conducting assessment
in the native language (L1), assessment in the second language (L2), bilingual
assessment, and nonverbal assessment (Ochoa & Ortiz, 2005).
L1 and Bilingual Cognitive Assessment
L1 and bilingual assessments can be useful, and even necessary, in cases when the
child is exposed to some formal education in their native language. There are a
number of cognitive tests that have been developed and adapted for assessing indi-
viduals in their first language. The Batería III Woodcock-Muñoz (Woodcock,
Muñoz-Sandoval, McGrew, and Mather, 2007), for example, was the adapted and
translated version of the Woodcock-Johnson III Tests of Cognitive Ability and Tests
of Achievement. It is the Spanish assessment battery most widely used by psycholo-
gists. It measures general intellectual ability, specific cognitive abilities, oral lan-
guage, and academic achievement. The Batería III can be given to children and
adults (ages 2–90+) and is comprised of 33 cognitive tests based on Catell-Horn-
Carroll (CHC) theory. The standardization and calibration data were gathered from
a sample of 1413 native Spanish-speakers representing several countries including
Mexico, Costa Rica, Panama, Argentina, Colombia, Puerto Rico and Spain, and
were equated to the Woodcock-Johnson III (WJ-III) norms. Caution should be used
when interpreting results of ELLs in the U.S, since differences in levels of Spanish
language proficiency, regional dialects, and educational backgrounds can contribute
to linguistic bias. When used with the WJ-III, it is the most comprehensive bilingual
test of cognitive abilities for children and adults.
L2 Cognitive Assessment
Although this may be considered the least desirable approach when testing bilingual
individuals or ELLs, L2 assessment can still be conducted adequately provided that
caution should be used when interpreting results, particularly if the individual
performed considerably low on cognitive tests. One consideration is to select tests
that have reduced cultural loadings and linguistic demands, based on the CHC
Cross-Battery approach (Flanagan, Ortiz, & Alfonso, 2013; Ortiz & Ochoa, 2005).
Tasks that are typically low in both cultural loading and linguistic demand are ones
that predominantly assess visual–spatial skills and fluid reasoning (i.e., novel problem
Table 12.1 Sample test selection for English–Spanish and/or bilingual individuals
276
Language development tests Age Language Domain/Ability

• Preschool Language Scale, 5th Edition (PLS-5; Birth-7:11 English, Spanish Receptive/expressive language, pragmatic
Zimmerman, Steiner, & Pond, 2012) language, morphology, syntax, vocabulary,
phonological awareness
• Expressive and Receptive One-Word Picture 2–70+ English, Spanish Receptive/expressive language, vocabulary
Vocabulary Tests, 4th edition (EOWPVT-4,
ROWPVT-4; Martin and Brownell, 2010a, 2010b,
2012a, 2012b)
• Peabody Picture Vocabulary Test, 4th Edition 2:6–90+ English Receptive language, vocabulary
(PPVT-4; Dunn & Dunn, 2007)
• Test de Vocabulario en Imagenes Peabody (TVIP; 2:6–17:11 Spanish Receptive language, vocabulary
Dunn, Lugo, Padilla, & Dunn, 1986)
• Woodcock-Munoz Language Survey—Revised 2–90+ English, Spanish Receptive/expressive language, reading,
(WMLS-R; Woodcock, Muñoz-Sandoval, Ruef, & writing, speaking, listening
Alvarado, 2005)
• Woodcock-Johnson Tests of Oral Language, 4th 2–90+ English, Spanish Receptive/expressive language, vocabulary,
Edition (WJ-IV OL; Schrank, Mather, & McGrew, phonological awareness
2014)
• Clinical Evaluation of Language Fundamentals, 3–6:11 English, Spanish Receptive/expressive language, pragmatic
Preschool, 2nd edition (CELF-Preschool-2; Semel, language, morphology, syntax, vocabulary,
Wiig, & Secord, 2006) phonological awareness
• Clinical Evaluation of Language Fundamentals, 5–21:11 English, Spanish Receptive/expressive language, pragmatic
4th edition (CELF-4; Semel, Wiig, & Secord, 2006) language, morphology, syntax, vocabulary,
phonological awareness, language memory,
working memory
• Bilingual Verbal Abilities Test, Normative Update 5–90+ English, Spanish, Arabic, Chinese, Receptive/expressive language, vocabulary
(BVAT-NU; Muñoz-Sandoval, Cummins, Alvarado, French, German, Haitian-creole,
& Ruef, 2005) Hindi, Hmong, Italian, Japanese,
Korean, Navajo, polish, Portuguese,
Russian, Turkish, Vietnamese
A.V. Pham et al.
(continued)
Language development tests Age Language Domain/Ability
Cognitive ability tests—Comprehensive
• Differential Abilities Scale, 2nd edition (DAS-2; 2:6–17:11 English, Spanish supplement Verbal ability, fluid reasoning, visual–spatial
Elliot, 2007) ability, working memory, long-term retrieval,
processing speed
• Kaufman assessment battery for children, 2nd 3–18 English Verbal ability, fluid reasoning, visual–spatial
edition (KABC-2; Kaufman & Kaufman, 2004) ability, working memory, long-term retrieval,
12 Cognitive Testing
processing speed
• Cognitive Assessment System, 2nd edition (CAS-2; 5–18:11 English Verbal ability, fluid reasoning, visual–spatial
Naglieri, Das, & Goldstein, 2014) ability, working memory, long-term retrieval,
attention shifting, processing speed
• Wechsler Intelligence Scale for Children, 5th 6–16:11 English, Spanish Verbal ability, fluid reasoning, visual–spatial
edition (WISC-V; Wechsler, 2014) ability, working memory, long-term retrieval,
processing speed
• Batería III Woodcock-Muñoz, Pruebas de 2–90+ Spanish Verbal ability, fluid reasoning, visual–spatial
habilidades cognitivas (Muñoz-Sandoval, ability, working memory, long-term retrieval,
Cummins, Alvarado, and Ruef, 2005) processing speed, attention
• Reynolds Intellectual Assessment Scales, 2nd 3–94 English Verbal ability, fluid reasoning, nonverbal
edition (RIAS-2; Reynolds & Kamphaus, 2015) memory, visual–spatial ability and memory,
processing speed
Cognitive ability tests—nonverbal
• Universal Nonverbal Intelligence Test, 2nd edition 5–21:11 English instructions Fluid reasoning, nonverbal memory, visual–
(UNIT-2; Bracken & McCallum, 2015) spatial ability and memory
• Leiter International Performance Scale, 3rd edition 3–75+ English instructions Fluid reasoning, nonverbal memory,
(Leiter-3; Roid, Miller, Pomplun, & Koch, 2013) visual–spatial ability and memory, visual
attention, attention shifting
• Comprehensive Test of Nonverbal Intelligence, 2nd 6–89:11 English instructions Fluid reasoning (e.g., pictures and geometric
edition (CTONI-2; Hammill, Pearson, and designs)
Weiderholt, 2009)
• Wechsler Nonverbal Scale of Ability (WNV; 4–21:11 English instructions Fluid reasoning, nonverbal memory, visual–
Wechsler & Naglieri, 2006) spatial ability, and memory
277
solving). Flanagan et al. (2013) developed the Cultural-Language Interpretative

Matrix (C-LIM) which provides an interpretive framework for clinicians to deter-
mine whether valid inferences may be drawn from the collected data, taking into
consideration cultural loading and linguistic demands of the subtests. If the effect of
cultural loading and linguistic demand is present in the test results, then the clinician
cannot attribute the low scores to cognitive weaknesses, but likely the individual’s
performance is attributed to cultural factors or limited English proficiency.
Nevertheless, further research is needed to validate the utility of this framework and
its ability to distinguish typically developing individuals from those with disabilities
in diverse populations.
Nonverbal Cognitive Assessment
Ideally, psychologists and neuropsychologists should use assessment methods that

are appropriate to the individual’s preferred and/or most competent language (APA,
2010). However, in cases where the clinician is not proficient in the examinee’s
native language, or if the test is not available in that language, then nonverbal assess-
ment of cognitive abilities should be considered. More aptly called language-reduced
assessment, this approach minimizes the influence of language during test adminis-
tration (e.g., instructions given in pantomime rather than verbally), during presenta-
tion of items or stimuli (e.g., pictorial content rather than written), and when
showing or responding to items (e.g., pointing rather than answering verbally).
However, most tests that are described as nonverbal still require verbal and linguis-
tic knowledge, and thus are never completely nonverbal (Ortiz, 2011). Studies have
shown that nonverbal abilities are also affected by schooling and culture (e.g.,
Arentoft et al., 2012). Most of all, language-reduced assessments measure a narrow
range of abilities and thus ignore other important abilities, such as auditory process-
ing and crystallized intelligence (i.e., the ability to use acquired skills, knowledge,
and experience), that are highly correlated with academic skills, particularly reading
and writing. Given these challenges, clinicians must always consider the purpose of
the assessment and examine whether the constructs in question are being accurately
assessed. Several of the most commonly used multidimensional nonverbal measures
are discussed below.
The Universal Nonverbal Intelligence Test-Second Edition (UNIT-2; Bracken
and McCallum, 2015) is unique because all instructions are given entirely in ges-
tures and pantomime. Thus, no verbal receptive language and expressive language
is assessed. It is given to children and young adults (ages 5–21) and is comprised of
six subtests that rely on quantitative, analogical, spatial reasoning, and visual mem-
ory. The norms consist of a national sample of 1802 individuals based on the 2014
U.S. Census. Reliability data suggest that it is a strong measure with their ELL
sample (.95–.99), although correlations with academic achievement are moderate at
best (.58–.71).
The Comprehensive Test of Nonverbal Intelligence-Second Edition (CTONI-2;

Hammill, Pearson, and Weiderholt 2009) measures analogical, sequential, and cat-
egorical reasoning (e.g., selecting options that fit a pattern, sequence, or category),
and can be given to children and adults (ages 6–89). Tasks are presented by either
pictorial or geometric figures, where the individual must determine the pattern or
relationship. Instructions can be presented orally or via gestures. It may be easier
for examiners to administer the CTONI-2 than the UNIT-2, since the former requires
no manipulatives, and the examinee responds by pointing to the correct response
from a list of options provided in the stimulus book. The norms of the CTONI-2
consist of a national sample of 2827 individuals in the U.S.
The Wechsler Nonverbal Scale of Ability (WNV; Wechsler & Naglieri, 2006) is
given to children and young adults (ages 4–21) and provides pictorial directions
instead of pantomime directions, although it does provide minimal verbal instructions
in multiple languages. It is comprised of six tasks including Matrices (e.g., selecting
missing portion of a figural puzzle), Recognition (e.g., recalling geometric designs
from a list), Spatial Span (e.g., tapping blocks in a particular order as demonstrated),
Picture Arrangement (e.g., sequencing pictures to present a logical story), Object
Assembly (e.g., forming a puzzle into a meaningful whole), and Coding (e.g., copying
symbols paired with numbers/shapes using a key). The norms of the WNV consist of
1323 individuals in the U.S. and 875 individuals in Canada.
Lastly, the Leiter International Performance Scale-Third Edition (Leiter-3; Roid,
Miller, Pomplun, & Koch, 2013) is given to young children and adults (ages 3–75+
years) and provides game-like tasks that assess fluid reasoning and visual–spatial
skills, along with attention and memory. Instructions are given via pantomime and
gestures. Compared to the other nonverbal measures, the Leiter-3 has many engaging
manipulatives (e.g., blocks, foam shapes) particularly for young examinees to
respond, in addition to pointing and making marks with a purple marker. The Leiter-3
is also unique, in that it yields Growth Scores (e.g., charting small improvements)
that can be used to monitor progress over time, particularly those with significant
intellectual disabilities. It is standardized with a norm sample of 1603 typical
individuals and 548 atypical individuals.
Nonverbal composite scores. Many cognitive or intellectual ability tests feature
nonverbal composites or clusters that are comprised of language-reduced tasks and
are purported to be fairer estimates of overall ability. Clinicians often report and
emphasize the nonverbal composite when explaining the individual’s overall ability,
particularly if the clinicians find significantly low performance scores on verbal
tasks and attribute these scores to limited English proficiency or developmental dis-
ability. For example, the Wechsler Individual Scale for Children-Fifth Edition
(WISC-V; Wechsler, 2014), which is given to children and adolescents (ages 6–16),
includes a Nonverbal score as part of their ancillary index scales. The WISC-V
Nonverbal score comprises six subtests: Matrix Reasoning (e.g., selecting missing
portion of a figural puzzle), Figure Weights (e.g., selecting best option to balance
two weights using quantitative reasoning), Block Design (e.g., constructing a design
using blocks and visual-motor skills), Visual Puzzles (e.g., selecting multiple pieces
that mentally construct a whole picture), Picture Span (e.g., recalling visual objects
in specified order), and Coding (e.g., copying symbols paired with numbers/shapes
using a key). This ancillary score places more emphasis on visual and spatial reason-
ing than the Full Scale IQ. Another commonly used test is the Kaufman Assessment
Battery for Children-Second Edition (KABC-2; Kaufman & Kaufman, 2004),
which provides the Nonverbal Index (NVI) score. The KABC-2 is administered to
children and adolescents (ages 3–18) and is appropriate for individuals whose native
language is not English, due to decreased linguistic demands. Additionally, the
Differential Abilities Scales-Second Edition (DAS-2; Elliot, 2007) provides the
Special Nonverbal Composite (SNC) for both preschool and school-aged children
(ages 2 ½–17). The SNC is also a measure of general nonverbal ability, and is
similarly comprised of tasks that assess fluid reasoning and visual–spatial skills,
making it appropriate to use for children with limited English proficiency. Although
these measures are helpful when the purpose is to assess overall cognitive ability
while minimizing the role of language proficiency, they do not completely eliminate
its influence on test results.
Best Practices in Cognitive Assessment of Bilinguals
Accurately assessing the cognitive abilities of bilinguals has been, and will continue
to be, a challenge for psychologists. More research is needed to identify best prac-
tices in assessment, interpretation, and intervention practices. In the following sec-
tion, we provide several recommendations for practitioners based on current
research and best practice. Future research should focus on testing these models for
their accuracy in identifying cognitive abilities of bilinguals and the implications of
these tests on their socio-emotional and academic success.
We first must emphasize that psychologists and other clinicians who work with
bilingual populations should have developed cultural understanding and compe-
tence before they start working with members of this population. Psychologists
must be mindful of potential ethical issues in assessment of diverse learners and
consider how cultural backgrounds and context may play a role in cognitive assess-
ment and interpretations. We propose a four-step process for test selection when
conducting cognitive assessments for bilingual adults and children and adults (see
Fig. 12.1). First, clinicians should observe behaviors related to the individual’s
acculturation at home, school, or work settings, including interactions with family
members and peer groups. Second, clinicians should evaluate the individual’s lan-
guage proficiency and dominance before deciding which language(s) should be
used for testing. Third, clinicians should become familiar with the population from
which normative data of standardized tests were developed and decide if the indi-
vidual (i.e., examinee) is comparable to the norm. Lastly, clinicians should select
the most appropriate test(s) based on the information collected from the previous
steps.
Step 1: Assessing Behavioral and Linguistic

Acculturation
• Consider the individual's dominant language in the context
of the child's level of acculturation.
• Observe the individual's behavior and language use in the
natural environment within multiple settings.
• Consider the individual's language proiciency in Basic
Interpersonal Communication Skills (BICS) and cognitive-
academic langauge proiciency (CALP; Cummins, 1984).
Step 2: Asessing Language Proiciency

• Assess L1 and/or L2
• Standardized language proiciency assessments can be
administered in groups or individually.
• Consider whether to use bilingual assessment tools (e.g., Bilingual
Verbal Abilities Test, Normative Update; Munoz-Sandoval,
Cummins, Alvarado, & Ruef, 2005).
Step 3: Determining Normative Issues Prior to

Testing Cognitive Abilities
• Review test manuals for norms
• Consider strengths and limitations of the norms in relation to
child's background and acculturation (e.g., monolingual speakers
from countries who are educated in the native language).
Step 4: Selecting the Appropriate Test(s)

• Referral questions often dictate the language(s) of the evaluation
along with test availability (e.g., referral for language disorder vs.
current and expected functioning in classroom).
• Caution should be taken using translators or interpreters to
administer English-language assessments
Fig. 12.1 Summary of recommendations for best practices in cognitive assessment of bilinguals
Step 1: Assessing Behavioral and Linguistic Acculturation
A person’s first language may not necessarily be his or her dominant language.
For instance, depending on language history and acculturation, a young child’s first
words may be spoken in the native language, but when immigrated at an early age
to a new language environment, the child’s L2 may then become the child’s dominant
language. Berry (1997) defined acculturation as the degree to which the person
maintains their heritage or native culture (or L1, in the case of language) and adapts
to the host or mainstream culture (or acquisition of L2). Although the degree to
which individuals use L2 is widely used as one of the proxies of acculturation, it
should not be automatically equated with acculturation into a new culture (Pazos &
Nadkarni, 2010).
Observations of individuals’ behavior and language use in the natural environment
can provide insight into the person’s acculturation (Rhodes, Ochoa, & Ortiz, 2005).
These observations can also provide substantive information during the assessment
process by contextualizing performance from, or in contrast to, standardized tests
and also allow for qualitative descriptions of the individuals’ strengths and weak-
nesses to be noted. Clinicians can observe the person’s language preference and
usage across multiple settings. They can also observe how the person adjusts to
the mainstream culture and the degree to which the person is comfortable speaking
L1 versus L2 in different settings (e.g., home versus school settings).
Cummins (1984) distinguishes between two types of language proficiency:
Basic Interpersonal Communication Skills (BICS) and Cognitive-Academic
Language Proficiency (CALP). BICS is described as a context-embedded language
skill that occurs primarily in social environments and in everyday communication.
These aspects of language proficiency seem to be acquired naturally and can be
developed without formal schooling. A person who uses BICS may do so when
interacting with peers, conversing at recess or lunch, or talking on the phone. BICS
is used when there are contextual supports and meaningful cues (e.g., facial expres-
sions, immediate feedback); thus, individuals can acquire BICS relatively quickly
(approximately three to 5 years).
CALP, on the other hand, is described as a more cognitively demanding skill that
emerges in formal school learning. These context-reduced language skills are essen-
tial for engaging in academically oriented tasks such as reading content area text-
books, completing statewide achievement tests, and listening to teacher lectures.
CALP requires higher order thinking skills (e.g., analysis, synthesis, and evaluation)
and increased cognitive demands, of which the development may take at least 7 years
for a child to become proficient (Cummins, 1984). When assessing language develop-
ment and progression, the child’s CALP can fall along any of these five (or in some
cases six) levels: (1) preproduction, (2) early production, (3) speech emergence, (4)
intermediate fluency, and (5) advanced fluency. Thus, clinicians may overestimate a
child’s overall language proficiency based on his or her BICS. A child may appear to
speak fluently in the social environment because he or she is able to use contextual
support (e.g., peer facial expressions), but may be several years away from attaining
the language proficiency needed to excel in academic content areas. In other words,
the child may speak fluently with friends at recess but be unable to speak fluently in
the classroom setting when learning science or history.
Clinicians should observe the child’s acculturation in both instructional settings
(e.g., reading lesson) and social contexts (e.g., recess) in order to estimate the child’s
BICS and CALP in both L1 and L2 (Pham, Goforth, Oganes, Medina-Pekofsky,
& Fine, 2016). Behavioral indicators of acculturation may include the child’s
preferred language at school, the cultural and linguistic backgrounds of peers with
whom they socialize, and retention of native or cultural traditions, customs, and
values. Clinicians, when possible, should also observe the child in the home
environment in order to understand the parent–child relationship and to observe the
acculturation of other family members. It is relatively common for the child to be
more acculturated to the mainstream culture than parents, and this pattern has been
demonstrated across immigrant groups (e.g., Costigan & Dokis, 2006; Goforth,
Pham & Oka, 2015; Telzer, Yuen, Gonzalez, & Fuligni, 2016). Thus, observations
of family dynamics and acculturation can provide insight into the development of
the child’s cultural identity, where behavioral changes can include shedding of
native cultural practices, either accidentally or deliberately, while adopting main-
stream cultural values (Berry, Poortinga, Segall, and Dasen 2002). Evaluating BICS
and CALP informally and formally yields information that would aid the clinician
in determining the best approach when conducting a psychological evaluation, and
in particular, when assessing cognitive functioning.
Step 2: Assessing Language Proficiency
Although assessment in the dominant language yields the most valid results for
diagnostic purposes, we also recommend screening or assessing the proficiency of
the less dominant language. Standardized language proficiency assessment can be
individually or group-administered to measure CALP growth in children and ado-
lescents who are bilingual or English ELLs. As an example of group-administered
assessment of CALP, recently, schools in Florida are using the Assessing
Comprehension and Communication in English State-to-State (ACCESS) for ELLs,
a standards-based assessment, developed by the World-Class Instructional Design
and Assessment (WIDA, 2012). This large-scale assessment is systematically
aligned with the English Language Development Standards (WIDA, 2012), which
is designed to improve teaching and learning of ELLs. It assesses four general lan-
guage domains: listening, speaking, reading, and writing. It also describes student
performance in terms of six English language proficiency levels: (1) Entering, (2)
Emerging, (3) Development, (4) Expanding, (5) Bridging, and (6) Reaching. This
group-administered state-level criterion-referenced assessment is helpful for deter-
mining the level of English proficiency among bilingual children and to gather data
on growth and development of students’ English language proficiency over time
(e.g., across the school year or multiple years).
In addition to state-level, group-administered assessments, clinicians can use

norm-referenced measures to determine language proficiency in both L1 and L2.
Unfortunately, some clinicians choose to translate or use interpreters to administer
English-language assessments, while also using norms developed for English-
speakers (Strutt et al., 2016). As we indicated earlier, this practice may lead to
potentially discriminatory practices due to issues of validity and reliability.
Nevertheless, there has been increasing availability of bilingual assessment tools,
although mainly for Spanish–English-speaking individuals. For example, the
Bilingual Verbal Abilities Tests, Normative Update (BVAT-NU; Muñoz-Sandoval,
Cummins, Alvarado, & Ruef, 2005) is an individually administered test used to
assess overall verbal ability and language proficiency of bilingual children and
adults (ages 5–90+) in 18 languages, including English. All items are first adminis-
tered in English (L2); however, any items answered incorrectly would be subse-
quently readministered in the person’s L1. The BVAT-NU provides information on
the individual’s CALP levels (levels 1–5), along with an English Language
Proficiency score, an overall Bilingual Verbal Ability score, and subtest scores. If a
person’s Bilingual Verbal Ability score is significantly higher than the English
Language Proficiency score, then this discrepancy can be attributed to limited
English proficiency.
The BVAT-NU is comprised of three subtests that allows the person to identify
objects and respond to items mainly using single word responses. In the first subtest,
Picture Vocabulary, the person is required to identify the name of a pictured object.
This is an expressive language task that involves word retrieval and measures com-
prehensive/knowledge or crystallized intelligence. The second subtest, Oral
Vocabulary, requires the person to provide a synonymous or antonymous word asso-
ciation of an orally presented stimulus. This task measures the person’s word know
ledge and word meaning. The third subtest, Verbal Analogies, requires the person to
recognize similarities between words by completing a logical word relationship.
The BVAT-NU provides a snapshot of the person’s CALP, although there are limita-
tions when using this particular measure. These limitations include (1) some items
being too culturally specific, (2) narrow assessment of specific language domains
(e.g., single word-level responses and no reading or writing tasks), and (3) some
items may have multiple translations due to regional differences (e.g., Mexican–
Spanish vs. Cuban–Spanish). Despite these limitations, the BVAT-NU does provide
useful information when determining general language proficiency and comprehen-
sion between L1 and L2 in many languages.
Another language measure is the Woodcock Muñoz Language Survey–Revised
(WMLS-R; Woodcock, Muñoz-Sandoval, Ruef, and Alvarado, 2005), which are sets
of individually administered tests designed to provide a broad sampling of English
and Spanish proficiency in receptive and expressive language, reading, and writing.
Several subtests are similar to the BVAT-NU, although the WMLS-R includes formal
testing of brief reading and writing skills. The WMLS-R can be administered to
individuals from ages 2–90+. In contrast with the BVAT-NU, the clinician must
administer both the English and Spanish versions separately to compute CALP lev-
els, which can be time consuming. Quantitative information is organized by aca-

demic language clusters that include Oral Language, Reading-Writing, Broad
English Ability, Oral Expression, Reading, Writing, Language Comprehension,
Applied Language Proficiency, Oral-Language Total, and Broad English Ability-
Total. These scores are often used in conjunction with the Batería III Woodcock-
Muñoz (Woodcock, Muñoz-Sandoval, McGrew, and Mather, 2007), which are
translated and adapted versions of the WJ-III (Woodcock, McGrew & Mather, 2001).
The WMLS-R also allows the clinician to gather qualitative information, such as the
person’s native language, the number of years exposed to the language of the test,
and the degree to which the person has been exposed to the language in home, school,
and social environments.
The WLMS-R consists of seven English subtests, which also include their
corresponding Spanish subtests: (1) Picture Vocabulary, (2) Verbal Analogies, (3)
Letter-Word Identification, (4) Dictation, (5) Understanding Directions, (6) Story
Recall and (7) Passage Comprehension. The WMLS-R yields six levels of CALP
in both English and Spanish to describe the individual’s language proficiency.
Additionally, the levels include regions of uncertainty to offset concerns regard-
ing seemingly arbitrary classification levels and cut-off scores. This is true in
cases where decisions are made regarding decisions in ELL classroom placement
or bilingual education. Thus, the WMLS-R provides two regions of uncertainty:
the first between CALP level 3 (limited) and level 4 (fluent) known as level 3–4,
and the second between CALP level 4 (fluent) and level 5 (advanced), known as
level 4–5.
A third test is the Woodcock-Johnson IV, Tests of Oral Language (WJ-IV OL,
Schrank, Mather, & McGrew, 2014), the most updated version of the Woodcock-
Johnson tests, which is used to gain information regarding the influence of language
on a person’s academic and cognitive performance. The WJ-IV OL is suitable for
ages 2–90+ and includes subtests from the previous WJ-III Tests of Cognitive
Abilities and Tests of Achievement, as well as new subtests that measure particular
language domains. These subtests include (1) Picture Vocabulary (e.g., recognizing
pictures with lexical knowledge); (2) Oral Comprehension (e.g., understanding and
supplying word(s) in an oral passage); (3) Segmentation (e.g., sound or phoneme
separation); (4) Rapid Picture Naming (e.g., naming facility of common pictures);
(5) Sentence Repetition (e.g., listening and recalling sentences using short-term
working memory); (6) Understanding Directions (e.g., listening to instructions and
pointing to objects in pictures); (7) Sound Blending (e.g., synthesizing language
sounds); (8) Retrieval Fluency (e.g., naming facility of acquired knowledge); and
(9) Sound Awareness (e.g., understanding and manipulating sounds within words).
Three of the subtests have parallel Spanish forms: Picture Vocabulary, Oral
Comprehension, and Understanding Directions, which only evaluate receptive and
expressive language in contrast with the WMLS-R, which includes reading and
writing. To assess bilingual learners, the WJ-IV OL provides CALP levels and the
Comparative Language Index (CLI) when using both English and Spanish parallel
versions of the subtests. The CLI is comprised of the numerators computed as part
of the Relative Proficiency Index (RPI) of two specific clusters: The Broad Oral
Language Cluster (English) and the Amplio languaje oral cluster (Spanish). The
larger of the two numerators indicates higher proficiency of that language.
Although more commonly used by speech-language pathologists, the following
tests may be useful for psychologists and neuropsychologists to examine receptive
and/or expressive language in both L1 and L2. The Preschool Language Scale-Fifth
Edition (PLS-5; Zimmerman, Steiner & Pond, 2012) is especially useful for infants,
toddlers, and young children (ages birth-7) due to norms provided in one-month
increments. In contrast with the BVAT-NU, items from the PLS-5 are administered
in Spanish first and any incorrect responses would be subsequently administered in
English. The Clinical Evaluation Language Fundamentals-Fourth Edition (CELF-
4; Semel, Wiig & Secord, 2006) can be administered to school-age children up
through college-age (ages 5–21) and evaluates both BICS and CALP along with
specific components of language, including morphology, syntax, semantics, and
pragmatics. It also offers the Preschool Version, 2nd edition (CELF-Preschool-2) in
both English and Spanish. The Peabody Picture Vocabulary Test-Fourth Edition
(PPVT; Dunn & Dunn, 2007) can be given to young children and adults (ages
2½–90+) but only assesses receptive vocabulary at the word-level. The PPVT’s
Spanish counterpart, Test de Vocabulario en Imágenes Peabody (TVIP; Dunn,
Lugo, Padilla, & Dunn, 1986) is administered to young children and young adults
(ages 2.5–18) to evaluate Spanish vocabulary, although its norms have not been
updated since the late 1980s. Ardila, Ostrosky-Solis, Rosselli, and Gómez (2000)
recommend testing receptive language in both English and Spanish, when possible.
All of these measures provide essential information that should be collected prior
to testing the child’s cognitive abilities.
tep 3: Determine Normative Issues Prior to Testing Cognitive

S
Abilities
Norms for many tests consist of only monolingual speakers from countries who are
educated in the native language. There have been frequent attempts to create bilingual
norms, yet there is still great diversity of acculturation and linguistic backgrounds in
a cultural group, such as those among Spanish-speakers. Even tests that purport to
include bilinguals in their norm groups do not consider the variations in English
proficiency during second language acquisition. It is important for normative data to
be culturally sensitive in addition to language equivalent, as any tests that are high in
cultural loadings and/or linguistic demands would not yield valid or interpretable
results. Salvia and Ysseldyke (1991) described this as assumption of comparability,
where cognitive tests and construction of norm samples should be based on shared
similarities between the acculturative and linguistic histories and experiences of
the sample norm and the child to whom the test is administered. However, this
assumption is rarely met due to widespread cultural and linguistic differences
(Ortiz & Ochoa, 2005), often leading to inequitable or discriminatory outcomes.

Thus, it is important to follow through with the first two steps of this process
(i.e., assess language proficiency and acculturation) in order to select the most
appropriate test and minimize unfair comparisons or bias.
Step 4: Selecting the Appropriate Test(s)
Collecting information regarding the child’s language and acculturation history is

important for selecting, administering, and interpreting tests for cognitive abilities
(AERA, 2014). Nevertheless, referral questions most often dictate the language(s)
of the evaluation along with test availability (Salinas, Bordes-Edgar, and Puente,
2016). For example, if the referral question asks whether the bilingual child exhibits
a language disorder, then the evaluation should be conducted in both L1 and L2.
If the referral question, however, asks about the current level of functioning in the
child’s mainstream classroom, then the focus of evaluation may be assessment of
L2, especially if that is the language in which competence will be evaluated.
Administering tests in both L1 and L2 separately can take a substantially longer
amount of time in the evaluation process than administering a test in one language,
but ultimately may yield more valid results for interpretation and decision-making.
There has been some evidence that balanced bilinguals tend to score higher on tests
when allowed to use both languages during the assessment (Paradis, 2008).
Additionally, the examiner must provide a rationale for determining which language
should be used during test administration and document any deviations from the
standardization, or testing the limits to account for these decisions.
Caution should also be taken if translators or interpreters are used to translate
English-only test items. The Standards for Educational and Psychologists Tests
(2014) suggest the use of interpreters should be the last possible option, due to inher-
ent challenges and difficulties resulting from this practice. These include: (a) lack of
linguistic and cultural equivalence between translation and test items, (b) lack of
appropriate training of the interpreter in test administration methods and knowledge
of psychometric terminology and concepts, and (c) potential miscommunication or
misinterpretation of information to and from the examiner. Thus, introducing an
interpreter who does not possess the proper training or experience working with such
individuals may pose a threat to the overall integrity (e.g., validity and reliability) of
the evaluation.
Conclusion and Future Directions
Assessment of cognitive abilities among bilingual/multilingual individuals continues

to pose many challenges for practitioners. Current research supports the use of L1
assessment among individuals who are still learning L2. Newer research (Karlsson,
et al., 2015) has shown that cognitive abilities could also be tested in L2 when the
participant/client has already obtained proficiency in the second language. However,
research in this area is still scarce and little has been done to determine the implica-
tions for clinical practice. Performance on tests of cognitive and language abilities
has typically been used to determine educational and diagnostic placements of
clients, which subsequently leads to appropriate access of educational support,
interventions, or services. Given the large proportion of the U.S. population who
report speaking a language other than English at home, researcher and practitioners
are urged to conduct more research in this area. They should also study the impact
of endorsing broad models that determine the effects of language dominance on the
clients’ performance on a cognitive test. We propose that practitioners should follow
a culturally and linguistically responsive model for cognitive assessment of bilin-
guals. This model must examine language dominance, acculturation level, and
similarities with the normative sample of the cognitive test to be used. Psychologists
have the ethical obligation to ensure they protect their clients’ overall well-being
by selecting valid and reliable tests and by engaging in culturally and linguistic
responsive practices as they make diagnostic and intervention recommendations
based on the results of the tests they select and administer. Engaging in culturally
and linguistically responsive practice allows us to be able to determine if an indi-
vidual not only possesses the abilities that will help them be successful in a given
society, but also in life.
References
American Educational Research Association. (2014). Standards for educational and psychological
testing. Washington, DC: AERA.
American Psychological Association. (2010). Ethical principles of psychologists and code of
conduct. Retrieved from http://www.apa.org/ethics/code/index.aspx
Ardila, A., Ostrosky-Solis, F., Rosselli, M., & Gómez, C. (2000). Age-related cognitive decline
during normal aging: The complex effect of education. Archives of Clinical Neuropsychology,
15, 495–513.
Arentoft, A., Byrd, D., Robbins, R., Monzones, J., Miranda, C., Rosario, A., et al. (2012).
Multidimensional effects of acculturation on English-language neuropsychological test
performance among HIV+ Caribbean Latinas/os. Journal of Clinical and Experimental
Neuropsychology, 34, 814–825. doi:10.1080/13803395.2012.683856
Berry, J. W. (1997). Immigration, acculturation, and adaptation. Applied Psychology: An
International Review, 46, 5–6.
Berry, J. W., Poortinga, Y. H., Segall, M. H., & Dasen, P. R. (2002). Cross-cultural psychology:
Research and applications (2nd ed.). New York: Cambridge University Press.
Bialystok, E. (2010). Bilingualism. Wiley Interdisciplinary Reviews: Cognitive Science, 1,
559–572.
Bialystok, E., & Craik, F. I. (2010). Cognitive and linguistic processing in the bilingual mind.
Current Directions in Psychological Science, 19, 19–23.
Bracken, B. A., & McCallum, R. S. (2015). The Universal Nonverbal Intelligence Test (2nd ed.).
Austin, TX: PRO-ED.
Charvat, J. (2008). Fluency and use of languages other than English among school psychologists:
Data from the NASP membership survey. Retrieved from http://www.nasponline.org/advocacy/
LanguageFluency_NASPSurvey05.pdf
Costigan, C. L., & Dokis, D. P. (2006). Relations between parent–child acculturation differences
and adjustment within immigrant Chinese families. Child Development, 77, 1252–1267.
Cummins, J. (1984). Bilingualism and special education: Issues in assessment pedagogy. San
Diego: College-Hill.
Cummins, J., & Swain, M. (1986). Bilingualism in education: Aspects of theory, research and
practice. New York: Longman.
Dunn, L. M., & Dunn, D. M. (2007). Peabody Picture Vocabulary Test (4th ed.). Minneapolis,
MN: Pearson.
Dunn, L. M., Lugo, D. E., Padilla, E. R., & Dunn, L. M. (1986). Test de Vocabulario en Imágenes
Peabody. Circle Pines, MN: American Guidance Service.
Elliot, C. D. (2007). Differential ability scales (2nd ed.). San Antonio, TX: Pearson Education.
Flanagan, D. P., Ortiz, S. O., & Alfonso, V. C. (2013). Essentials of cross-battery assessment (3rd ed.).
Hoboken, NJ: Wiley.
Garratt, L. C., & Kelly, T. P. (2007). To what extent does bilingualism affect children’s performance
on the NEPSY? Child Neuropsychology, 14(1), 71–81.
Genesee, F. (1989). Early bilingual development: One language or two? Journal of Child Language,
16, 161–179.
Genesee, F., & Nicoladis, E. (2007). Bilingual acquisition. In E. Hoff & M. Shatz (Eds.), Handbook
of language development (pp. 324–342). Oxford: Blackwell.
Goforth, A. N., Pham, A. V., & Oka, E. R. (2015). Parent–child conflict, acculturation gap, accul-
turative stress, and behavior problems in Arab American adolescents. Journal of Cross-Cultural
Hammill, D. D., Pearson, N. A., & Weiderholt, J. L. (2009). Comprehensive Test of Nonverbal
Intelligence, Second Edition (CTONI-2). Austin, TX: PRO-ED.
Hernandez, A., Li, P., & MacWhinney, B. (2005). The emergence of competing modules in bilin-
gualism. Trends in Cognitive Sciences, 9, 220–225. doi:10.1016/j.tics.2005.03.003
Hoff, E., Core, C., Place, S., Rumiche, R., Señor, M., & Parra, M. (2011). Dual language exposure
and early bilingual development. Journal of Child Language, 39, 1–27.
Individuals with Disabilities Education Act § 1400 (2004).
Jarvis, L. H., Danks, J. H., & Merriman, W. E. (1995). The effect of bilingualism on cogni-
tive ability: A test of the level of bilingualism hypothesis. Applied PsychoLinguistics, 16(3),
293–308. doi:10.1017/S0142716400007311
Johnson, J. S., & Newport, E. L. (1989). Critical period effects in second language learning: The
influence of maturational state on the acquisition of English as a second language. Cognitive
Kalia, V., Wilbourn, M. P., & Ghio, K. (2014). Better early or late? Examining the influence of
age of exposure and language proficiency on executive function in early and late bilinguals.
Journal of Cognitive Psychology, 26, 699–713.
Kaufman, A. S., & Kaufman, N. L. (2004). Kaufman assessment battery for children (2nd ed.).
Circle Pines, MN: American Guidance Service.
Karlsson, L. C., Soveri, A., Räsänen, P., Kärnä, A., Delatte, S., Lagerström, E., et al. (2015).
Bilingualism and performance on two widely used developmental neuropsychological test
batteries. PLoS One, 10(4), e0125867.
Kena, G., Musu-Gillette, L., Robinson, J., Wang, X., Rathbun, A., Zhang, J., et al. (2015). The condition
of education 2015. NCES 2015–144. National Center for Education Statistics. Retrieved from
http://files.eric.ed.gov/fulltext/ED556901.pdf
Martin, N., & Brownell, R. (2010a). Expressive One Word Picture Vocabulary Test, Fourth Edition
(EOWPVT-IV). Austin, TX: PRO-ED.
Martin, N., & Brownell, R. (2010b). Receptive One Word Picture Vocabulary Test, Fourth Edition
(ROWPVT—IV). Austin, TX: PRO-ED.
Martin, N., & Brownell, R. (2012a). Expressive One Word Picture Vocabulary Test, Fourth Edition:
Spanish (EOWPVT—IV). Austin, TX: PRO-ED.
Martin, N., & Brownell, R. (2012b). Receptive One Word Picture Vocabulary Test, Fourth Edition:
Spanish (EOWPVT—IV). Austin, TX: PRO-ED.
Muñoz-Sandoval, A. F., Cummins, J., Alvarado, C. G., & Ruef, M. L. (2005). Bilingual verbal
ability tests. Itasca, IL: Riverside Publishing.
Naglieri, J. A., & Bornstein, B. T. (2003). Intelligence and achievement: Just how correlated are
they? Journal of Psychoeducational Assessment, 21, 244–260.
Naglieri, J. A., Das, J. P., & Goldstein, S. (2014). Cognitive assessment system (2nd ed.). Austin,
TX: PRO-ED.
Naglieri, J. A., Otero, T., DeLauder, B., & Matto, H. (2007). Bilingual Hispanic children’s
performance on the English and Spanish versions of the cognitive assessment system. School
Psychology Quarterly, 22(3), 432–448.
National Center for Educational Statistics (2016). Children and youth with disabilities. Retrieved
from http://nces.ed.gov/programs/coe/indicator_cgg.asp
Ochoa, S. H., & Ortiz, S. O. (2005). Language proficiency assessment: The foundation for
psychoeducational assessment of second-language learners. In R. L. Rhodes, S. H. Ochoa, &
S. O. Ortiz (Eds.), Assessing culturally and linguistically diverse students: A practical guide
(pp. 137–152). New York: Guilford Press.
Ortiz, S. O. (2011). Separating cultural and linguistic differences (CLD) from specific learning
disability in the evaluation of diverse students: Difference or disorder. In D. P. Flanagan &
V. C. Alfronso (Eds.), Essentials of specific learning disability identification (pp. 299–324).
Hoboken, NJ: Wiley.
Ortiz, S. O., & Ochoa, S. H. (2005). Cognitive assessment of culturally and linguistically diverse
individuals: An integrated approach. In R. Rhodes, S. H. Ochoa, & S. O. Ortiz (Eds.), Assessing
culturally and linguistically diverse students: A practical guide (pp. 168–201). New York:
Guilford Press.
Paap, K. R., Johnson, H. A., & Sawi, O. (2015). Bilingual advantages in executive functioning
either do not exist or are restricted to very specific and undetermined circumstances. Cortex,
69, 265–278.
Paradis, J., Genesee, F., & Crago, M. B. (2011). Dual language development and disorders:
A handbook on bilingualism and second language learning (2nd ed.). Baltimore, MD: Paul H
Brookes.
Paradis, M. (2008). Bilingualism and neuropsychiatric disorders. Journal of Neurolinguistics, 21,
199–230.
Pazos, H., & Nadkarni, L. (2010). Competency with linguistically diverse populations. In J. A.
Erickson-Cornish, B. A. Schreier, L. I. Nadkarni, L. Henderson Metzger, & E. R. Rodolfa (Eds.),
Handbook of multicultural counseling competencies (pp. 153–194). Hoboken, NJ: Wiley.
Pham, A. V., Goforth, A., Oganes, M., Medina-Pekofsky, E., & Fine, J. G. (2016). Nondiscriminatory
neuropsychological assessment of children with learning disabilities. In F. R. Ferraro (Ed.),
Minority and cross-cultural aspects of neuropsychological assessment (2nd ed., pp. 359–378).
New York, NY: Taylor & Francis.
Poplack, S. (1980). Sometimes I’ll start a sentence in Spanish y termino en espanol: Toward a
typology of code-switching. Linguistics, 18, 581–618.
Reynolds, C. R., & Kamphaus, R. W. (2015). Reynolds Intellectual Assessment Scales (RIAS-2)
Lutz. FL: Psychological Assessment Resource.
Rhodes, R. L., Ochoa, S. H., & Ortiz, S. O. (2005). Assessing culturally and linguistically diverse
students: A practical guide. New York: Guilford Press.
Roid, G. H., Miller, L. J., Pomplun, M., & Koch, C. (2013). Leiter International Performance
Scale (3rd ed.). Los Angeles, CA: Western Psychological Services.
Rosselli, M., Ardila, A., Araujo, K., Weekes, V. A., Caracciolo, V., Padilla, M., et al. (2000).
Verbal fluency and repetition skills in healthy older Spanish-English bilinguals. Applied
Neuropsychology, 7, 17–24.
Salinas, C. M., Bordes-Edgar, V., & Puente, A. E. (2016). Barriers and practical approaches to
neuropsychological assessment of Spanish speakers. In F. R. Ferraro (Ed.), Minority and
cross-cultural aspects of neuropsychological assessment (2nd ed., pp. 229–258). New York,
NY: Taylor & Francis.
Salvia, J., & Ysseldyke, J. E. (1991). Assessment in special education and remedial education
(5th ed.). Boston, MA: Houghton Mifflin.
Schrank, F. A., Mather, N., & McGrew, K. S. (2014). Woodcock-Johnson IV Tests of Oral
Language. Rolling Meadows, IL: Riverside Publishing.
Semel, E. M., Wiig, E. H., & Secord, W. (2006). Clinical Evaluation of Language Fundamentals
(CELF-4) Spanish. San Antonio, TX: The Psychological Corporation.
Shin, H. B., & Ortman, J. M. (2011). Language projections: 2010 to 2020.. Retrieved October 4,
2016, from http://www.census.gov/hhes/socdemo/language/data/acs/Shin_Ortman_FFC2011_
paper.pdf
Strutt, A. M., Burton, V. J., Resendiz, C. V., & Peery, S. (2016). Neurocognitive assessment of
Hispanic individuals residing in the US: Current issues and potential solutions. In F. R. Ferraro
(Ed.), Minority and cross-cultural aspects of neuropsychological assessment: Enduring and
emerging trends (2nd ed., pp. 201–228). NY: Taylor & Francis.
Suzuki, L. A., & Valencia, R. R. (1997). Race-ethnicity and measured intelligence. American
Psychologists, 52, 1103–1114.
Telzer, E. H., Yuen, C., Gonzales, N., & Fuligni, A. J. (2016). Filling gaps in the acculturation gap-
distress model: Heritage cultural maintenance and adjustment in Mexican–American families.
Journal of Youth and Adolescence, 45, 1412–1425.
U.S. Census Bureau. (2013). Language use in the United States: 2011. American community survey
reports. Retrieved from https://www.census.gov/prod/2013pubs/acs-22.pdf
Valdes, G., & Figueroa, R. A. (1994). Bilingualism and testing: A special case of bias. Norwood,
NJ: Ablex.
Wechsler, D. (2014). Wechsler Intelligence Scale for Children (5th ed.). Bloomington, MN:
Pearson.
Wechsler, D., & Naglieri, J. A. (2006). Wechsler Nonverbal Scale of Ability (WNV). San Antonio,
TX: Harcourt Assessment.
WIDA Consortium. (2012). 2012 Amplification of the English language development standards
kindergarten–grade 12. Madison, WI: Board of Regents of the University of Wisconsin System.
Woodcock, R. W., McGrew, K. S., & Mather, N. (2001). Woodcock-Johnson Tests of Cognitive
Abilities and Tests of Achievement (3rd ed.). Itasca, IL: Riverside.
Woodcock, R. W., Muñoz-Sandoval, A. F., Ruef, M. L., & Alvarado, C. G. (2005). Woodcock-Muñoz
language survey–Revised. Itasca, IL: Riverside Publishing.
Woodcock, R. W., Muñoz-Sandoval, A. F., McGrew, K. S., & Mather, N. (2007). Batería III
Woodcock-Muñoz. Itasca, IL: Riverside Publishing.
Yates, J., Ortiz, A., & Anderson, R. (1998). Issues of race, ethnicity, disability and culture.
Enhancing Diversity: Educators with Disabilities, 21–37.
Zimmerman, I. L., Steiner, V. G., & Pond, R. E. (2012). Preschool Language Scales (5th ed.).
Bloomington, MN: Pearson.
Index
A Active-passive, 157–183
Abbreviated Multidimensional Acculturation Additive bilingualism, 237, 272
Scale (AMAS-ZABB), 263–265 Adulthood, 29, 34, 35, 39, 193, 194, 196, 199,
Abnormal aging, 189, 197–205, 211, 212 202, 213, 232
Abnormal aging process, 189, 198–212 Adult learners, 203
Abstract, 7, 8, 34, 76, 81, 84, 101, 146, 159, Advanced fluency, 282
160, 163, 164, 175, 228, 229 AERA. See American Educational Research
Abstract concepts, 5–9, 20, 21, 76, 228 Association (AERA)
Academic Affect, 33, 43, 66, 68, 77, 101–105, 129, 131,
achievement, 238, 275, 278 142, 151, 167, 175, 177, 179, 184, 189,
language, 272, 285 195–197, 199–201, 203, 204, 215, 216,
skills, 278 221, 229, 230, 240, 272
Accents, 44, 163, 172 Affective
Acceptability judgment tests, 181 attitude, 96
ACCESS. See Assessing Comprehension and flattening, 240, 243
Communication in English State-to- structure, 164–167, 171, 176
State (ACCESS) system, 164, 166, 170, 171, 175, 177,
Accessibility, 148, 151, 169, 170, 228 178, 184
Acculturation, 40, 236, 238, 243, 245, 263, Affective-conceptual, 164, 176, 177
265, 270, 280–283, 286–288 Affective-conceptual interface, 164, 176, 177
Acculturative stress, 236, 238 African-American, 44
Accuracy, 63, 96, 97, 102, 126, 127, 130, 138, Afrikaans, 220
150, 241, 280 Afrikaans–English, 220, 243
Acquisition by processing theory (APT), Age, 11, 12, 29, 30, 34, 35, 40, 41, 45, 46, 48,
169, 170 50–55, 57–59, 61, 64, 93, 120,
Across-language, 4, 10, 11, 41, 49, 70, 86, 123–127, 130, 131, 143, 145, 151, 157,
101, 104, 228, 229, 239, 241, 243, 244, 175, 189–191, 193–197, 200, 202, 203,
246–247, 264 205, 213, 218–221, 234, 239, 243, 245,
Activation, 5, 8, 21, 33, 52, 53, 55, 56, 60, 65, 246, 248, 261, 262, 264, 270, 272, 273,
68, 77, 79–81, 83, 85, 89, 91, 93, 94, 275–280, 282, 284–286
96, 98–100, 103–105, 108, 120, 121, Age of acquisition (AoA), 40, 43, 44, 47, 51,
123, 126, 127, 129, 163, 166–168, 53–55, 60, 66, 69
172–174, 180, 183, 212 Age of language acquisition, 213, 262
Activation theory of consciousness, 172 Age-related decline, 197
Active bilingual, 45–50, 53, 54, 59, 124, Aging, 124–125, 189–206, 211, 212, 218,
176, 194 220, 221

Brain Book Series, DOI 10.1007/978-3-319-64099-0
294 Index
Agraphias, 213, 215 Assessment, 231, 233–236, 242, 244–250,

Agreeableness, 260, 262, 264–266 259, 261, 274–278, 280–284, 287, 288
Alcohol, 137, 235 Asymmetry, 106, 121, 244
Alexia, 205, 211–215, 221 Attention, 51, 60, 66, 85, 123, 124, 127–129,
Allocentric coordinates, 10, 13 144, 145, 172, 192, 193, 217, 236, 238,
Alogia, 240 244, 248, 271, 277, 279
Alzheimer’s disease, 97, 197–204, 218–220 Attentional mechanisms, 60
Amazonian, 33 Attentional network task, 123, 124, 129
American culture, 44, 232 Attention deficit hyperactivity disorder
American Educational Research Association (ADHD), 236
(AERA), 274, 287 Attrition, 196, 272
American Psychological Association (APA), Auditory
244, 274, 278 hallucinations, 241, 244, 245
Amount, 10, 15, 19, 60, 127, 201, 202, 270, input, 91, 161, 169, 170, 180
272, 287 module, 161
interference, 15 processing, 278
metaphors, 10, 15, 19 structure, 161, 162, 168–170, 179
Amplio languaje oral cluster, 286 verbal imagery, 28
Amygdala, 64, 68 Australian, 13, 218
Analogical, 278, 279 Australian aboriginal, 13
Analyzability, 77, 82, 83 Autism, 239
Anger, 29, 54, 164, 229 Autism spectrum disorders (ASDs), 239
Anglo, 246 Autobiographical memories, 29, 148, 247
Anhedonia, 230, 240 Automatic
Animals, 21, 88, 122, 190, 220 processing, 52, 66
Anomia, 213, 220 skill, 200
Antagonistic, 214 Automaticity, 54–56, 244
Anterior Autonomic, 42, 56, 57, 59, 167
insula, 68, 127 Autonomic approaches, 56, 60
temporal atrophy, 220 Autonomous, 162
Anterior cingulate cortex, 125, 127 Availability, 21, 89, 103, 107, 148, 166, 281,
Antisaccade, 130 284, 287
Antonyms, 284 Avoidance
Anxiety, 29, 42, 230, 231, 235, 246 behavior, 167
Anxiety-provoking, 42 mechanisms, 51
AoA. See Age of acquisition (AoA) Avolition, 240
APA. See American Psychological Association Axes, 10, 13
(APA) Azari–Farsi, 204
Aphasia, 198, 205, 211–214, 220, 221
Aphasia profile, 213
Aphasic Agraphia, 213 B
Applied Language Proficiency, 285 Baby, 162
Arabic, 92, 276 Balanced bilingualism, 131, 271
Argentina, 275 Balanced bilinguals, 53, 67, 95, 122, 129, 143,
Aristotelian, 79 150, 153, 161, 287
Arithmetic, 31–35, 191 Basal ganglia, 125, 204
equations, 32 Basic color terms, 4
memory, 31, 35 Basic interpersonal skills (BICS), 281–283, 286
Arousal, 41, 42, 44, 246 Basque, 243
Arrangement task, 12, 14, 15, 17, 18 Batería III, 275, 277, 285
Asians, 137, 230 Batería III Woodcock-Muñoz, 275, 277, 285
Assessing Comprehension and Behavioral, 31, 33, 44, 64, 69, 91, 119, 120,
Communication in English 123–127, 131, 132, 150, 217, 228, 229,
State-to-State (ACCESS), 283 238, 245, 281–283
Index 295
Behavioral methods, 44, 127 Canonical, 13

Belgium, 203 Canonical condition, 13
Benjamin Whorf, 4 Cantonese–English, 122
Berinmo, 5 Cardinal direction, 10, 13, 14, 19
Between-language, 4, 5, 9, 16, 31, 33, 41, 43, Cardiovascular pathology, 198
48, 53, 57, 65, 66, 70, 103, 120, 128, Career of metaphor theory, 84
129, 138, 139, 146, 151, 153, 167, 172, Caregiver, 199, 200, 219, 221
179, 193, 216, 264 Case studies, 42, 228, 241, 243, 244, 247, 250
BICS. See Basic interpersonal Cat, 76–78, 86, 91, 139
skills (BICS) Catalan, 46, 53, 70, 125–127, 200
Bicultural, 260 Catalan–Spanish, 46, 125, 126, 200
Big Five, 233, 260 Catatonia, 240
Big Five Inventory (BFI), 260, 261, 264, 265 Categorical perception, 4, 5
Bilingual advantages, 125, 128–130, 132, 193, Categorical reasoning, 279
194, 202–204 Category-fluency tasks, 122, 197
Bilingual cognitive advantage, 193 Category information, 7
Bilingual cognitive Assessment, 275 Catell-Horn-Carroll (CHC), 275, 277
Bilingual dementia, 199–202, 219 Caudate, 126, 127
Bilingual dual code theory, 139 Caudate nuclei, 127
Bilingual experience, 119, 128, 131, 259 CELF-Preschool-2, 276, 286
Bilingualism-specific advantages, 205 Central American, 264
Bilingual language effects, 147, 149 Central-processor, 160
Bilingual memory, 101, 138–140, 144, 146, Centroparietal, 67
147, 149, 151–153, 247 Cerebrovascular disease, 192, 198
Bilingual mode, 101, 171, 183, 220 Charanda, 137, 148
Bilingual norms, 153, 286 CHC Cross-Battery approach, 275
Bilingual Verbal Abilities Tests, Normative Childhood, 34, 35, 39, 51, 52, 93, 191, 193,
Update (BVAT-NU), 276, 284, 286 194, 196, 199, 201, 228, 237, 239, 243,
Bilingual Verbal Ability, 276, 281, 284 247, 260–262
Block design, 279 Children, 15, 29, 31, 34, 55, 120, 129, 130,
Blocked conditions, 122 167, 175, 176, 193, 212, 215–218, 221,
Bodily states, 6 237–239, 249, 270–280, 283, 284, 286
Boston Naming Test (BNT), 197 China, 13, 17, 18, 148, 150, 151, 230
Braille text, 174 Chinese, 9–13, 16, 18, 19, 28, 29, 32, 33,
Brain 40, 47, 55, 63, 67, 68, 102–104, 148,
atrophy, 198, 200, 202 150, 215, 218, 221, 230, 231, 246, 247,
correlates, 83 270, 276
function, 92 Chinese–English, 11–13, 16, 18, 32, 33, 48,
pathology, 211, 212 55, 63, 65, 67, 68, 102, 104, 246, 247
representation, 212 Chinese idioms, 102–104
reserve, 192, 202 Chinese-translated idioms, 102–104
size, 192 Chronological, 41, 218, 239
Bridging, 283 Chronological age (CA), 218
Broad English Ability, 285 Chunks, 85, 89, 93, 108
Broad English Ability-Total, 285 Cingulate cortex, 125, 127
Broca’s area, 28, 35, 126, 193, 213, 240 Circumlocation errors, 190
Brodmann Area 44, 28 Circumstantial bilingualism, 40
BVAT-NU. See Bilingual Verbal Abilities Clinical
Tests, Normative Update (BVAT-NU) assessment, 233, 248, 249
interview, 229, 242, 245, 246, 248, 250
populations, 97, 233, 239
C studies, 42
CALP. See Cognitive-academic language Clinical Evaluation Language Fundamentals-
proficiency (CALP) Fourth Edition (CELF-4), 276, 286
Canada, 227, 279 Clinician Bias, 245–246
296 Index
Clockwise, 18, 21 Comprehension, 69, 75, 77–86, 90, 92, 95–97,

Clusters, 279, 285, 286 100, 101, 105, 123, 161, 173, 176, 179,
Co-activate, 161, 163–164, 168, 169, 173 190, 191, 196, 198, 200, 204, 212, 220,
Co-activated chain, 163 238, 241, 283–285
Co-activation, 8 Comprehensive Test of Nonverbal
Codes of Ethics, 274 Intelligence-Second Edition (CTONI-
Code-switches, 105 2), 277, 279
Codeswitching, 122, 199, 201, 205 Computerized tomography (CT) scan, 202
Coding, 98–100, 106, 138, 279, 280 Computer screen, 86, 88, 91
Cognition, 4, 6, 20, 21, 30, 66, 69, 76, Conceptual
191–195, 197, 236, 237, 239, 241, domains, 7, 76, 81
247, 249 grammatical knowledge, 172
Cognitive judgement task, 7
abilities, 132, 191, 192, 195, 211, 269–288 metaphor, 3–22, 76, 81
approaches, 44 metaphor theory, 6, 8, 21, 76
assessment, 271, 274, 275, 277–281, 288 principles, 162
decline, 195, 200, 202, 211, 218 semantic, 141, 162, 163
impairment, 200, 204, 205, 218, 221 structure, 162, 164–166, 171, 174,
level, 202 177–180, 183
performance, 124, 130, 194, 195, 205, 221, system, 7, 76, 81, 160–162, 164, 166, 172,
273, 285 175–180
processes, 92, 237, 241 Conceptually-driven, 140
psychology, 6, 146 Concrete, 5–8, 21, 76, 81, 92, 101, 142, 146,
reserve, 192, 193, 195, 202–205 153, 229
reserve hypothesis, 192, 193 Concreteness effect, 146
science, 158 Concrete words, 142, 146, 229
screening test, 195 Conduct disorder (CD), 236, 237
tasks, 192–195, 273 Configurative hypothesis, 82, 86, 93
test performance, 270, 272–273 Conflict monitoring, 35, 128, 130
tests, 205, 270–288 Conflict tasks, 194
Cognitive-academic language proficiency Confrontational naming, 219, 220
(CALP), 281–286 Congruent, 9, 54, 56, 104, 129, 234
Co-index, 164, 165 Connectionist approach, 120
Collocations, 75, 85, 96, 106, 182 Conscientiousness, 260, 262, 264–266
Colombia, 275 Conscious awareness, 140, 152, 166, 171,
Colombian, 76, 83, 97 172, 174
Color, 4, 5, 54, 55, 126, 130, 193 Conscious experience, 173
Color processing, 5 Consciousness, 166, 171–173, 265
Color-shape switching, 126, 130 Constraint-based model, 83, 102
Commander-in-chief, 160 Container, 7, 10, 15
Communicative interaction, 201 Content analysis approach, 244
Community language, 176 Context, 29, 32, 40, 41, 43, 44, 52, 53, 55, 60,
Compensatory pathways 69, 70, 78–81, 84, 88, 90, 92, 94, 98,
Competence, 69, 85, 95, 96, 98, 108, 219, 264, 100–104, 107, 108, 120, 128, 131, 140,
280, 287 149–151, 168, 174, 177, 216, 228, 230,
Competition, 120, 126, 169–171, 173, 182, 272 235, 237, 238, 243, 246–249
Competitive, 167, 170 Context-dependent, 216, 228, 248
Complex psychological processes, 28, 35 Context-dependent memory, 228
Complex task, 122, 127 Context-embedded, 282
Composite scores, 279 Context of acquisition, 41, 43, 53, 55
Compositional, 76, 82, 83, 86, 88, 89, 93, 96, Continuum, 39, 41, 96, 121, 233
106–108 Control groups, 130, 200, 204, 273
Compound, 145, 151, 153 Control processes, 126, 128, 132
Index 297
Convergent validity, 131, 132 proficiency, 260

Coordinate, 10, 13, 145, 153 understanding, 280
Coprolalia, 239 Cultural-Language Interpretative Matrix
Core grammar, 173 (C-LIM), 278
Core language system, 161–164, 167, 172, Culturally loaded, 273
174, 178–180, 182, 183 Culture, 10–12, 16, 19, 33, 40, 43, 44, 51, 53,
Correlation, 12, 33, 44, 55, 125, 203, 219, 235, 66, 69, 70, 131, 150, 229, 232–234,
236, 241, 260, 261, 264, 265, 278 236, 249, 259–262, 264–266, 273, 278,
Correlation analysis, 44 282, 283
Correspondence rules, 163, 214 Culture-relevant stimuli, 260
Cortex, 28, 68, 125–127, 131, 192, 204 Cyrillic, 214
Cortical areas, 35, 60, 212, 213, 244
Cortical brain, 33
Cortical networks, 41 D
Costa Rica, 275 Danish, 29, 217
Counseling, 231, 234, 249, 274 Decimal point, 30
Count, 31, 86, 87, 192, 231 Decimals, 30
Counterclockwise, 18, 21 Decision-making, 65, 123, 241, 247, 287
Counting, 31, 45, 52, 140, 141, Declarative knowledge, 199, 244
168, 199, 200 Declarative learning, 204
Crick and Koch, 166 Decline, 190–193, 196, 198–200, 202, 205,
Cross-cultural, 76, 230, 231, 261 206, 211, 218, 219
Cross-domain, 8, 272 Decomposable, 77, 82, 90
Cross-language, 33, 86, 101, 102, 104, 105, Dedicated memory, 160
107, 122, 199, 218, 219, 261 Deep
interference, 218 processing, 53, 140–142, 144, 150, 153
intrusions, 122 semantic processing, 141
overlap, 86, 101, 102 Deeper level of processing, 51
Cross-linguistic, 4, 5, 11, 76, 102, 103, 120, Deeper processing levels, 66
129, 214, 216, 231 Deeper semantic analysis, 66
analyses, 214 Delusions, 240, 242
competition, 120 Dementia, 189, 198–203, 211, 212, 218–221
effects, 179–183 Dementia syndromes, 211
interactions, 102 Demographic data, 130
Cross-modal lexical, 91 Denmark, 29
Cross-modal lexical priming, 91 Depth-of-processing, 53, 139–141, 153
Cross-modal lexical priming paradigm Developmental, 12, 31, 170, 175, 177, 211,
(CMLP), 91 214–217, 221, 228, 229, 238–239, 249,
Cross-modal priming, 94 279
Cross-talk, 158 Developmental dyslexia, 211, 214, 216–218,
Crystallized intelligence, 278, 284 221
Cuban, 270 Development disorders (DD), 239, 249
Cued-recall task, 101 Development dysphasia, 211, 212, 215–216,
Cued switching, 194, 205 221
Cultural Diagnostic and Statistical Manual of Mental
competence, 264 Disorders 5th Edition (DSM-5), 232
empathy, 260, 266 Diagnostic tool, 245
frame switching effects, 260 Dialect, 44, 130, 163, 171, 172, 230, 270, 275
identity, 44, 263–265, 283 Dialect-switching, 130
identity scores, 44, 264 Dialogue, 122
idioms, 229–232, 236, 248 Dictation, 285
linguistic background, 43 Differential Abilities Scales-Second Edition
loading, 273, 275, 278, 286 (DAS-2), 277, 280
298 Index
Differential aphasia recovery, 205 Early posterior negativity (EPN), 60–63, 66,
Dimensions, 14, 40, 77, 83, 93, 231, 233, 67
260–264, 278 Early production, 282
Direct East, 10, 14, 15
access model, 80, 92, 93 Ecologically, 87
memory tasks, 152 Education(al)
retrieval, 102, 103 background, 275
Direction, 10, 13, 14, 16–21, 43, 47, 53, 108, level, 219
131–132, 141, 144, 145, 153, 228, 261, EEG. See Electroencephalogram (EEG)
266, 279, 285, 287–288 Egocentric coordinates, 10
Discriminant validity, 153 Egocentric speech, 28
Disease, 97, 189, 192, 198–204, 218–220 Ego-moving, 8, 10, 11, 16
Disgust, 164 Elective bilingualism, 40
Disorganized symptoms, 240, 241 Electrical activity, 34, 92
Dissociated aphasia, 212, 213 Electrodes, 60, 62, 67
Dissociated dyslexia, 217, 218 Electroencephalogram (EEG), 62, 63, 92
Dissociated language disorders, 205, 211–221 Electrophysiological, 42, 60–67, 80, 83, 105
Distance ELLs. See English-language learners (ELLs)
interference, 15 Embarrassing, 41, 42, 246
metaphors, 10, 15 Embodiment, 6–7, 14, 16, 18, 21–22
Distinctive, 28, 139, 144 Emerging, 271, 283
Distributed-practice, 138, 145–147, 152, 153 Emotion(al)
Divided attention, 128 association, 39–70
Divided visual field, 98–100 dissociation, 41, 51, 69
Dividing, 30, 128 distance, 41, 42
Domain, 4, 7, 8, 21, 29, 33, 39, 76, 81, 83, 84, inner speech, 29, 30, 35
96, 100, 106, 147, 160, 162, 175, 193, intensity, 43, 50–53, 59, 140
201, 216, 228, 229, 240, 243, 247, 261, interference tasks, 31, 54–56, 69
272, 276, 277, 281, 284, 285 processes, 70, 228, 229
Domain-general, 160 ratings, 41
Dominant language, 30, 34, 43, 55, 56, 105, stability, 260
121–126, 142, 143, 171, 197, 200, 201, state, 40, 229
219, 262, 263, 271, 272, 281–283 stimuli, 50, 54, 56, 60, 68, 70
Dorsolateral, 126 Stroop task, 48, 54, 69
Dot-drawing task, 14 valence, 51, 61, 63, 64, 67–69
Down’s Syndrome, 239 value, 40, 41
Dreaming, 30 word processing, 56, 60
Dual coding theory, 138 words, 40, 47, 50–56, 60, 62, 63, 66–70
Dual language, 128, 194, 195, 228, 271, 273 Emotion–memory effect, 45, 46, 51–54, 140,
Dual language functioning, 273 141
Dual language system hypothesis, 271 Emotive decision task, 96
Dual-processing model, 82 Empirical regularity, 147
Duration reproduction, 15 Encoding, 35, 51, 105, 138–147, 149–153
Dutch, 9, 18, 107, 182, 183 specificity, 149–151, 153
Dysarthria, 204 England, 55
Dyslexia, 205, 211, 212, 214, 216–218, 221 English
Dyslexics, 216, 217 proficiency, 44, 61, 263, 270, 272–274,
Dysphasia, 205, 211, 212, 215–216, 221 278–280, 283, 284, 286
speakers, 5, 10–12, 15, 17, 19, 20, 40, 47,
53, 59, 88, 103, 104, 230, 246, 261,
E 284
Early bilingualism, 40, 93, 123, 124 English–Afrikaans, 220
Early bilinguals, 35, 40, 46, 54, 69, 93, 123, 124, English–Chinese, 247
138, 153, 203, 212, 213, 261, 262, 272 English-dominant, 94, 95, 142
Index 299
English–French, 101, 122, 212 Family dynamics, 283

English Language Development Standards, Farsi, 9, 204
283 Fear, 54, 64, 164, 230
English-language learners (ELLs), 270, Feline, 76
272–275, 278, 283, 285 Females, 29, 35, 47, 48, 59, 213, 234, 262,
English Language Proficiency, 283, 284 263
English-language tests, 270 Figurative competence, 85, 98, 108
English-learning, 237 Figurative expressions, 75, 76, 81, 84–86, 89,
English-only, 14, 218, 270, 287 91, 92, 95, 96, 98–101, 106, 107
English–Spanish, 28, 29, 95, 121, 123, 129, Figurative-first hypothesis, 81
146, 152, 153, 212, 236, 271, 276, 277 Figurative-first model, 80–81, 86, 95
English-speaking, 40, 52, 55, 67, 217, 230, Figurative language, 75–108
231, 242, 245, 246, 272, 284 Figurative meaning, 76–83, 85, 86, 89, 91–96,
Entering, 283 99, 100, 103, 106–108
Environment-based direction, 10 Figurative processing, 79, 81, 85–108
Episodic memory, 137–153 Figurative tropes, 78–80, 85, 93, 94, 96, 99,
EPN. See Early posterior negativity (EPN) 107, 108
Equivalence hypothesis, 81 Figure weights, 279
ERPs. See Event related potentials (ERPs) File-drawer problem, 130
Erratic behaviors, 240 Filipino–English, 32
Ethnic, 34, 230, 246, 270 Fine/Coarse Coding Theory (FCT), 98, 99,
Ethnic minorities, 269 106
European, 216, 217, 228 Finnish–English, 47, 54
Evaluative processing, 96 Finnish, 90, 217, 219
Event related potentials (ERPs), 34, 60–67, 92, Finnish-speaking, 219
105, 126 Finnish–Swedish bilinguals, 219
Executive function, 123, 126, 130, 131, 191, First fixation, 86, 95, 103
192, 241 First person, 231, 232
Executive processing ability, 123 Five-Factor model, 233
Expanding, 283 Fixation, 86, 87, 94
Experiential knowledge, 76 Fixation durations, 86, 87, 95, 103
Explicit Fixed phrase, 77
knowledge, 172, 182 Flankers, 130
memory, 244 Flanker task, 124, 128, 130
Explosive words, 239 Flexibility, 34, 36, 83, 192, 237, 260, 271
Expressive Florida, 260, 262, 283
abilities, 178 Fluency, 14, 51, 90, 91, 119–132, 147, 192,
language tasks, 284 196, 218–221, 228, 270–273, 282, 285
Externalizing psychopathology, 236–238, 249 Fluent aphasia, 213
External speech, 28, 169 Fluid reasoning, 275, 277, 279, 280
Extralinguistic, 97, 107 fMRI. See Functional magnetic resonance
Extraversion, 260–262, 264–266 imaging (fMRI)
Eyes Forced bilingualism, 40
measures, 87 Foreign language, 176
movements, 86, 104 Forensic, 235
tracking, 86, 87, 91, 93, 95, 100, 102, 104, Forgetting, 149, 151
105, 129 Formal-semantic, 96
Formulaic language, 85, 87, 88, 90, 93, 97,
100, 104, 108
F Frame, 21, 262
Facial expressions, 240, 282 Frame of reference, 10, 14, 174, 246
False friends, 95 Framework, 76, 79–84, 94, 95, 107, 108, 138,
Familiarity, 45, 46, 49, 50, 52, 57–59, 84, 157–184, 278
89–91, 96, 104, 105, 107, 108, 148, 153 France, 200
300 Index
Free recall, 45, 46, 51, 140, 146, 151, 152 Grapheme-to-phoneme correspondence, 214
Free spontaneous fluency tasks, 196 Grasping, 162
French, 62, 90, 120, 145, 163, 200, 214–218, Greek, 10, 15, 90, 181, 217
235, 243, 270, 276 Greek–English, 48, 55, 59
French–English, 122, 145, 235, 239 Green-blue boundary, 5
French–German, 66 Ground, 6–8, 16, 19, 20, 76, 77, 149, 182
French–Spanish, 217 Grounded cognition, 6
French-speaking, 200 Group-administered, 283, 284
Frequency, 22, 34, 43, 45, 53, 55, 69, 83, 84, 89, Growth, 169–171, 283
92, 122, 124, 131, 132, 143, 147, 148, Growth Scores, 279
153, 170, 196, 197, 200, 219, 221, 231 GSH. See Graded Salience
Frequent structures, 171 Hypothesis (GSH)
Friulian–Italian, 204 Gyrus, 5, 28, 35, 64, 65, 68, 126
Frontal area, 33, 213
Frontal cortical areas, 213
Frontal gyrus, 28, 35, 65, 126 H
Frontal lobe, 213 Hak, 14
Front/back, 8, 10, 13, 14 Hallucinations, 240–245
Frontotemporal dementia, 198, 203 Hangul, 215
Full Scale IQ, 280 Hanja, 215
Functional magnetic resonance imaging Happiness, 68, 164, 230
(fMRI), 33, 61, 63–65, 68, 126, 127, Head noun, 161
212, 240, 244, 245 Healthy, 75, 97, 190, 192, 197, 200–202, 204,
Functional magnetoencephalography 206, 231, 238, 240, 243, 244
(MEG), 193 Hebrew, 17, 20, 98, 100, 243
Height interference task, 9
Helplessness, 230
G Hemisphere, 97–99
g ability theory, 272 Hemisphere-specific, 98
Gender, 43, 61, 62, 64, 130, 194, 233, 263 Hemispheric asymmetry, 244
General cognitive abilities, 191 Hemorrhage, 213
Generalizability, 120, 138, 150, 153 Hemorrhagic brain lesions, 198
Generation, 28, 39, 138, 144, 197, 203, 273 Heritage language, 176
Generation effect, 144–145, 153 Higher-order, 86, 141, 143, 241, 282
Generative linguistics, 162 High-frequency words, 196, 197, 200
Geometric figures, 279 Highly frequent, 171
German, 44, 60–62, 64, 66, 68, 90, 182, 183, Hijacking, 91
215, 219, 220, 261, 262, 276 Hindi, 217, 276
German–English, 68 Hispanic(s), 234, 236, 270, 273, 274
German–French, 66 Historical facts, 34, 150
German–Spanish, 60, 261, 262 Hmong, 230, 276
Gestures, 179, 278, 279 Holistic, 86–89, 107
GLM, 264 Holistic Hypothesis, 86, 88, 89, 92, 106
Global aphasia, 213 Holocaust, 44
Goal maintenance, 128 Home environment, 176, 283
Goluboy, 5 Hong Kong, 247
Go/noGO, 128 Hopelessness, 230
Graded Salience Hypothesis (GSH), 84, 89, Horizontal, 9–13, 17–20
94, 98–100, 103, 106 Horizontal dimension, 14
Grammar, 3, 161, 172, 182 Hospitalization, 234
Grammatical Hybrid model, 83, 86, 89, 107
complexity, 191 Hyderabad, 203
structures, 172, 200 Hypoperfusive, 198
Index 301
I Intellectual ability, 275, 280

Iconic, 22 Intellectual testing, 212
Identifying tag, 161 Intelligence quotient (IQ), 191, 194
Ideograms, 215 Intelligence tests, 124, 128, 272
Idiomatic constructions, 83 Intentional, 138, 140, 144, 145, 147, 149,
Idiomatic expressions, 78, 80, 82, 83, 86, 87, 151–153
89, 91, 93, 99, 100, 102, 181 Interaction of language, 63, 143, 144
Idiomatic key, 83 Interactively, 158
Idiomatic Processing Model, 81 Interdependent memory, 152
Idiom Decomposition Model, 82 Interfaces, 160–166, 168–170, 173, 175–178,
Idiom Diffusion Model Of Second Languages, 180, 182
90 Interference, 5, 9, 15, 32, 44, 47, 48, 54–56,
Idiom-final word, 101, 102, 105 59, 66, 67, 69, 95, 120, 121, 123, 128,
Idiom List Hypothesis, 80, 82, 90 139, 151, 171, 179, 192, 194, 199, 218,
Idioms, 75–91, 93, 94, 96, 99, 100, 102–105, 219, 272
107, 229–232, 236, 248, 249 effect, 48, 54, 55, 95
I know, 137 suppression, 128, 192
Illiterate, 203, 217 Interlocutor, 97, 201
Illness, 198, 229, 231–233, 235, 245, 248, 249 Intermediate fluency, 282
Immersion, 43, 47, 50, 52, 53 Internalization, 28
Immersive, 43, 55 Internalized social speech, 28
Immigrants, 29, 34, 55, 59, 131, 202–204, Internal reliability, 264
219, 236, 237, 243, 245, 249, 264, 283 International, 181, 277, 279
Immigration, 29, 195, 202, 203, 236, 270 Interpersonal behavior, 229
Immigration status, 236 Interpreters, 123, 127, 129, 270, 284, 287
Implicit Introspective, 42–44
association task, 12 Involuntary, 44, 167
memory, 244 Involuntary responses, 162
Incidental, 105, 138, 145, 147, 152, 212, 221 IQ. See Intelligence quotient (IQ)
Incidental free recall task, 140 I remember, 137, 138
Incongruent, 9, 54, 56 Ironic, 78, 96, 97
Independently, 153, 158 Irony, 76, 78, 84, 85, 96, 97, 99, 107, 108, 241
Independent systems, 159 Irregular items, 199
Index, 92, 161–166, 169, 180, 248, 279 Irregular writing systems, 214, 215, 217
India, 203 Irrelevant lexical information, 128
Indian languages, 217 Israel, 243
Indigenous, 14, 231 Italian, 40, 123, 130, 214, 216, 217, 276
Indirect, 152, 158, 215 Italian–English, 123
Individuals with Disabilities Education Act
(IDEA), 274
Indo-European, 214 J
Infants, 4, 9, 21, 286 Jackendoff-style architecture, 163
Inferior frontal gyrus, 28, 35, 126 Japanese, 163, 183, 199, 214, 215, 218, 221,
Inferior parietal lobule, 5, 33 276
Inhibition, 97, 121, 126, 129, 130, 192, 194, Japanese–English, 183
205, 206, 236, 271 Jewish, 44
Inhibitory control, 121, 123, 125, 126, 128, Journalist, 214
130, 192–194 Journey, 7, 8, 76, 81
Innate, 4
Inner speech, 27–36
Instinct, 162 K
Instinctive, 162 Kana, 199, 214, 215, 218
Instructional setting, 43, 50, 53, 283 Kanji, 199, 214, 215, 218, 221
302 Index
Kannada, 217 Language-related, 33, 162, 167

Kaufman Assessment Battery for Children- Language-sensitive areas, 213
Second Edition (KABC-2), 277, 280 Language-specific, 32, 33, 35, 121, 139, 205
Keyboard, 142, 144, 193 Language-specific deficits, 120–121
Knowledge-formation, 158 Language-switching, 31, 120, 122, 123,
Korean, 60, 61, 215, 276 125–129, 131, 205
Korean–English, 60, 212 Last in, first out theory, 199
KuukThaayorre, 13, 14 Late bilingualism, 40, 93, 123–124
Late bilinguals, 45, 53, 54, 59, 60, 67, 69, 93,
123–124, 195, 203–205, 211–212, 261,
L 272
LA–, 35 Latencies, 35, 48, 54, 55, 60, 66, 67, 80
L2 acquisition, 45, 93, 124, 212, 237, 239, Late positive complex (LPC), 61, 66, 67
241, 243, 246, 247, 249, 270–272 Lateralization, 244
Lag, 145, 146 Latin, 44, 181, 214
Language Latin America, 30, 262
appropriate, 273 Latino(s)
attitude, 42, 44, 176–177 cultural identity, 44, 263–265
choice, 30, 34, 42, 43, 201, 219–221 culture, 44, 234, 264
competence, 264 Latvia, 228
comprehension, 78, 79, 82–84, 92, 105, Learning arithmetic (LA+), 35
198, 285 Learning disability, 274
confusion, 271 Least significant difference (LSD), 142
control, 125–128, 192, 199, 201 Left hand, 17, 123
deficits, 198, 204, 238 Left hemisphere, 5, 97, 98, 100, 244
delays, 239, 271 Left posterior, 5, 213, 215
deterioration, 198, 199, 201 Left-to-right/left/right, 10, 12–15, 17, 18
disorders, 205, 211–221, 239, 274, 287 Leiter International Performance Scale-Third
dominance, 42, 43, 53, 56, 69, 93–95, 105, Edition (Leiter-3), 277, 279
142, 143, 145, 147, 153, 200, 201, 248, Lesions, 28, 35, 99, 198, 205, 213
260, 266, 288 Letter fluency-tasks, 190, 196
impairment, 198, 215, 216, 219, 221, 237, Letter identification, 215
239 Letter–number sequencing, 191
independent system, 139 Levels of processing, 52, 138–144, 150, 153
loss, 198, 200 Lewy bodies, 198, 203
mixing, 271 Lexical access, 67, 84, 86, 119, 122–124
mode, 171, 183 Lexical decision(s), 51, 91, 98, 99, 103, 152,
processor, 79, 95, 108 196
similarity, 47 Lexical decision task, 46, 47, 52, 60, 61, 63,
skill, 201, 282 65–68, 120
Language-choice abilities, 201 Lexical diversity, 191
Language-dependent, 31, 139, 147, 150–153, Lexical item, 82, 90, 96, 108, 148, 162
245 Lexicalization Hypothesis, 81
Language experience and proficiency Lexicalized, 84, 89–91, 93, 96, 106
questionnaire (LEAP-Q), 263 Lexicalized chunks, 85
Language faculty broadly defined (LFB), 163 Lexical network, 96
Language faculty narrowly defined (LFN), Lexical priming, 91
162, 163 Lexical representation, 67, 82, 86, 90, 91, 121,
Language-free, 151 123, 131
Language-neutrally, 182 Lexical Representation Hypothesis, 81, 92
Language-reduced Lexical retrieval
assessment, 278–280 abilities, 196
tasks, 279 task, 197
Index 303
Lexical robustness, 121, 131 Memory

Lexical-semantic processing, 204 effect, 52, 138, 139, 146, 150
Lexical semantic systems, 220 encoding, 51
Lexicon, 78–86, 89–91, 93, 96, 99, 100, 102, networks, 32, 35, 36
105–108, 120, 163, 190, 191, 196, 243 performance, 151
Lifespan, 124–125, 190–195, 203 retrieval, 139, 140, 147, 149, 153, 192
Lifestyle, 192, 194, 195, 202, 203 Mental age, 239
Limited English proficiency, 270, 272–274, Mental arithmetic, 31
278–280, 284 Mental calculations, 29–31, 33–36
Linguistic(s) Mental context, 174
abilities, 189, 191, 204 Mentalese, 28, 29
activity, 171 Mentalese hypothesis, 28
bias, 275 Mental illness, 231–233, 235, 245, 248, 249
deficits, 215 Mental language, 27
demand, 273, 275, 278, 280, 286 Mental lexicon, 78, 79, 81–86, 89, 91, 93, 96,
determinisn, 4 99, 105–108, 163
errors, 204 Mental line, 11, 13, 17–21
fluency, 270 Mental mathematics, 36
input, 161 Mental operations, 171, 174
metaphors, 7–10, 20, 21 Mental representation(s), 4, 7, 9
relativity, 3–22 Mental representation of time, 9–12, 15, 16,
Listening, 9, 59, 85, 127, 175, 238, 272, 276, 19, 20
282, 283, 285 Metacognition theory, 240
Literal analysis, 77, 79, 80, 82, 83, 90, 93, 96, Metalinguistic, 172, 174, 178, 181, 271, 272
107 Metaphor
Literal-first view, 80, 93 processing, 21, 22, 83, 84
Literally plausible, 77, 93, 96, 99, 103 training, 15
Literal meaning, 76–85, 87, 90, 92–95, 97, Metaphorical competence, 95, 96
99–101, 107 Metaphoric association, 6, 8, 9, 21, 22
Literalness, 77, 84, 107 Metaphoric concepts, 7
Literal Salience Model, 90, 91 Methodological, 88, 144, 245, 248
Logographic, 214, 215, 218, 221 Metonymy, 76, 78, 80, 85, 107, 108
Longitudinal designs, 130, 131 Mexican, 270
Longitudinal study, 193, 194 Mexico, 10, 234, 235, 260, 275
Long term memory (LTM), 32, 138, 168 Miami, 39
Los Angeles, 227 Mianmin village, 14
Low frequency, 122, 124 Mian speakers, 10, 15
Low-frequency words, 196, 197 Middle-aged, 124
Low-literalness, 77 Middle-class, 245
LPC. See Late positive complex (LPC) Migration, 29, 40, 195, 196
Luxemburg, 228 Milestones, 271
Mind, 6, 28, 75, 76, 105, 125, 152, 158–160,
164, 166, 167, 170, 173–175, 177–180,
M 183, 184, 228
Magnetic resonance imaging (MRI), 68, 220 Mini Mental-State Examination (MMSE),
Main effect, 58, 63, 65, 140, 264 195, 203
Mandarin, 11, 150 Minnesota Multiphasic Personality
Mapping(s), 7–9, 11, 76, 81 Inventory-2 (MMPI-2), 233–236
Massed presentations, 145 Minority groups, 44
Matched on age, 130, 202 Mixed
Math, 30, 32–35, 273 conditions, 122
Matrix Reasoning, 279 context, 43
Mayan, 14 dementia, 203
304 Index
Mix languages, 179, 219, 220 N

MMPI-2. See Minnesota Multiphasic N2, 67, 126
Personality Inventory-2 (MMPI-2) N200, 60
Modality-specific, 6 N400, 35, 67, 92, 105
Model of Dual Idiom Representation, 90 N(oun), 165
Moderators, 131 NAEP. See National Assessment of
Modular, 160, 164–166, 172, 173, 177, 178 Educational Progress (NAEP)
Modular architecture, 159, 162 Naming defect, 213
Modular-internal, 161 Naming tasks, 120, 123, 124, 126,
Modular Online Growth and Use of Language 128, 190, 194
(MOGUL), 157–176, 182 Narrative production, 191
Modules, 160–165, 167–174, 178–180 National Assessment of Educational Progress
Monitoring, 35, 62, 66, 67, 127–130, 170, 192, (NAEP), 274
193 National Association of School Psychologists
Monitoring task, 62, 66, 67, 129 (NASP), 274
Monolingual(s) National census, 269
memory, 138, 153 Native
mode, 107, 171 language, 12, 29, 30, 34, 43–45, 51–56, 58,
Monologue, 122, 231 69, 86, 90, 98, 102, 106, 121, 127, 148,
Mood, 6, 233, 236 176, 213, 216, 219, 231, 238, 242, 262,
Moral dilemma, 41 263, 270, 275, 278, 280, 282, 285, 286
Morphemes, 217 speaker, 4, 39, 40, 45, 47, 52, 59, 66, 85, 87,
Morphological awareness, 218 89, 90, 99, 103, 104, 106, 107, 172, 239
Morphology, 276, 286 Native-like, 90, 165, 170, 201, 272
Morpho-syntactic deficits, 204 Nativeness, 122
Morpho-syntactic impairments, 204 Natural reading, 87
Morpho-syntactic module, 161 N2 component, 67, 126
Morpho-syntactic structure, 162 Negative
Morpho-syntax, 204 effects, 192
Motivated forgetting, 149, 151 symptoms, 240
Motivation, 83, 167, 235 valence, 61, 96, 97
Motor area, 28, 127 Neologisms, 213
Motor programming, 167 Nerves, 230
Motor structure (MoS), 163, 167 Nervios, 230
Motor system, 159, 163, 164, 167, 178, 179 Network, 21, 32, 35, 36, 41, 85, 96, 120, 123,
MRI. See Magnetic resonance imaging (MRI) 124, 129, 159–161, 163–167, 173, 180,
Multicompetence, 260 181, 202, 272
Multiculturalism, 29, 260 Neural circuit, 76
Multicultural Personality Questionnaire Neural correlates, 33
(MPQ), 260 Neural count, 192
Multi-dominance, 260 Neural level, 160, 168
Multilingual(s), 29, 34, 41, 127, 157, 159, 163, Neurally, 158
171, 172, 174, 178–180, 202, 205, 220, Neural structures, 212
270, 287 Neural substrate(s), 63, 68, 192
Multilingualism, 29, 144, 163, 195, 203 Neuro basis, 5, 125, 126, 202
Multilingual patients, 220 Neurocognitive, 125–128, 270
Multiplication calculations, 34 Neurodegenerative disease, 198
Multiplying, 30 Neuroimaging, 5, 60, 83, 119, 125, 131, 132
Multistage Model, 79 Neuroimaging approaches, 42, 68
Multi-word Neurological, 28, 35, 41, 97–101, 124, 221
expressions, 89, 90, 93, 102 Neuropsychological, 97, 221, 242, 244, 250,
phrase, 80 270, 272, 274
Musical, 4, 9 Neuropsychological conditions, 212
Index 305
Neuropsychologists, 270, 275, 278, 286 Numerosity, 31

Neuro theory of metaphor, 76 NVI. See Nonverbal Index (NVI) score
Neuroticism, 233, 260–262, 264–266
Neuroticism Extraversion Openness-Five
Factor Inventory (NEO-FFI), 261 O
New face recognition, 85, 183 Occipital lobe, 215
Noncanonical, 13, 103 Occipitotemporal, 62, 66
Non-canonical condition, 13 Old age, 190, 195–197
Noncompositional, 83, 107 Older, 31, 33, 123–125, 139, 175, 192–197,
Non-cued switching, 194 205, 218, 219, 246, 270, 271
Nondecomposable, 77, 79, 82, 90, 93 Older adults, 125, 190–195, 197, 198, 206
Non-dominant language, 122, 125, 143, 171, One-dimensional, 11–15
197, 200, 201 One-dimensional space, 10, 15
Non-educated, 195 One-to-one correspondence, 77
Non-elective bilingualism, 40 Online
Non-English speaking, 272 monitoring, 192
Nonfluent aphasia, 213 questionnaire, 29
Nonformulaic, 85–88, 97, 104, 106 Onset, 11, 66, 92, 120, 123, 124, 127, 131,
Noninvasive, 92 198, 202, 205, 232, 240, 245
Non-linguistic, 4, 9, 12, 15, 17, 167, Onset of dementia, 198, 201–205
192, 193 Opacity, 216, 221
Non-linguistic tasks, 7–9, 15, 19, 20 Open-mindedness, 260, 266
Nonliteral, 79–81, 85, 86, 92, 94, 95, 97–101 Openness, 260, 261, 264–266
Non-minority, 245 Opercularis, 28
Non-movement, 21 Oppositional defiant disorder (ODD), 237
Nonnative language, 85, 86, 88, 90, 92, Oppression, 237
97–100, 102, 104, 107 Oral Expression, 285
Non-pathological personality, 233 Oral language, 211, 238, 275, 276, 285, 286
Non-switchers, 128 Oral Vocabulary, 284
Nonverbal, 124, 139, 194, 229, 237, 238, 273, Order of acquisition, 43, 60, 67, 200, 201, 205
275 Organs, 167, 178
assessment, 275, 278 Orthographic, 66–68, 70, 214, 216, 218, 221
composite scores, 279 Orthography, 18, 216–218
intelligence test, 124
Nonverbal Index (NVI) score, 280
Nonwords, 66, 91, 98, 120, 216, 217 P
Normal aging, 189–199, 220 P1, 67
Normal bilinguals, 198, 220, 221 P300, 60
Normative data, 270, 280, 286 PAI. See Personality Assessment Inventory
Normative issues, 286–287 (PAI)
North, 10, 14 Panama, 275
Nosology, 231 Papua New Guinea, 5, 14
Noun phrase, 161 Parachute reflex, 162
Nouns, 7, 55, 66, 126, 161, 165, 180, 183, Paradigms, 9, 11, 12, 16, 42, 44–50, 57, 61,
190, 220 62, 64, 70, 86, 87, 91, 96, 126–128,
Novice interpreters, 127 131, 138, 149, 152
Null differential effects, 53 Parahippocampal cortex, 68
Null effects, 130 Parallel, 14, 31, 53, 95, 107, 159, 166, 168,
Number, 15, 30–33, 35, 52, 75, 79, 81–90, 93, 173, 174, 200, 204, 205, 211, 212, 214,
94, 97, 101, 102, 105–107, 120, 121, 285
131, 142, 144, 161, 165, 168, 169, 178, Parallel aphasia, 212
179, 182, 190, 191, 194–196, 202–205, Parallel processing, 81–83, 86, 88, 91, 96, 103,
214, 231–233, 242, 248, 273–275, 279, 106, 107
280, 285 Parallel processing account, 80
306 Index
Paraphasias, 213 Vocabulary, 124, 238, 276, 284–286

Parasitic strategy, 91 Picture-naming tasks, 120, 123, 124, 128
Parent-child relationship, 283 Pitch, 4, 6, 9, 241
Parents, 55, 130, 176, 177, 216, 235, 238, 239, Pitch reproduction task, 9
249, 270, 272, 283 Pitres’ law, 213
Parietal cortex, 125 Pitres’ rule, 213
Parietotemporal, 213 Planning, 30, 126, 191, 244
Parkinson’s disease, 97, 198, 204 Plasticity, 34, 272
Pars, 28 PLS-5, 276, 286
Passage Comprehension, 285 Plural nouns, 232
Passive bilingual, 174, 176 Poetics, 79, 80, 99
Pathways, 160, 192, 238 Pointing, 12, 14, 16, 83, 100, 278, 279, 285
Patient, 42, 91, 97, 99, 199–202, 212–214, Polish, 29, 86, 96, 97, 99, 102, 215, 220, 276
218–221, 231, 239–250 Polish–English, 96, 99, 102
Pattern recognition, 152 Polyglot, 213, 219
Peabody Picture Vocabulary Test-Fourth Pormpuraaw, 13, 14, 20
Edition (PPVT), 276, 286 Portuguese, 199, 200, 217, 276
Peak latency, 35, 66 Positive, 6, 8, 12, 35, 41, 43–49, 51–53,
Pedagogy, 85 56–64, 66–68, 78, 96, 97, 140, 141,
Pentalinguals, 29 164–167, 175–178, 193, 195, 230, 237,
Perceptual, 7, 76, 92, 152, 161, 163, 166, 171, 240, 241, 260, 261
172, 174, 179, 215 Positive symptoms, 240
judgment task, 7 Positive valence, 7, 96, 97
modules, 161, 179 Post-childhood, 191
Personality Posterior, 5, 60, 61, 67, 213, 215
disorder, 232–236, 249 Post-stroke recovery, 205
profile, 260, 265 PPA. See Primary progressive aphasia (PPA)
traits, 232, 233, 259–266 PPVT. See Peabody Picture Vocabulary
Personality Assessment Inventory (PAI), 233, Test-Fourth Edition (PPVT)
235, 236 Pragmatic information, 78, 199
Personal word meanings, 28 Pragmatics, 78–80, 182, 199, 201, 243, 244,
Phonemic, 140, 151, 218 276, 286
Phonemic cues, 190 Predictability, 83, 92
Phonograms, 215 Preferential-looking paradigm, 9
Phonographic systems, 215 Prefrontal cortex, 125–127
Phonological structure (PS), 161, 162, 167, Pre-frontal cortex, 125–127, 192
169 Prelinguistic, 4, 21, 28
Phonology/phonological Premotor cortex, 28
awareness, 218, 276 Pre-production, 282
knowledge, 161 Preschool Language Scale-Fifth Edition
module, 161, 163, 170, 179 (PLS-5), 276, 286
principles, 160 Preschool Version, 2nd edition (CELF-
system, 160, 162, 168, 214 Preschool-2), 286
tasks, 216 Primary progressive aphasia (PPA), 198–200
Phrasal verbs, 85, 91–93, 96, 106, 108 Priming, 11, 17, 44, 91, 94, 98, 100, 103, 104
Phrase, 43, 59, 76–78, 80, 82, 83, 85–90, 92, Primitive elements, 173
93, 96, 100, 102–108, 161, 183, 219, Primitive structures, 173
231 Private speech, 28, 30, 35
Physiological arousal, 246 Proactive interference, 139
Pictorial content, 278 Problem solving, 30, 33, 128, 237, 249
Picture Procedural knowledge, 199
description tasks, 190, 196 Processing, 5, 33, 42, 75–108, 123, 138, 158,
Span, 279 191, 212, 229, 272
superiority effect, 151 Processing speed, 277
Index 307
Processor, 79, 95, 108, 160, 163, 168, 169, threshold, 217
173, 179 time, 86–88, 94, 95, 102–104
Production, 28, 91, 101, 125, 161, 167, 173, Reading level (RL), 217, 218
175, 176, 178, 179, 190, 191, 198, 204, Reading-Writing, 285
212, 220, 240, 244, 282 Real world, 5, 6
Proficiency, 39, 42–44, 52–56, 61, 66, 67, 69, Recall, 29, 31, 44, 45, 80, 101, 137–141, 145,
89, 90, 93, 96, 102, 104, 107, 108, 119, 147, 149, 150, 163, 240, 244, 247, 285
120, 122, 124, 127, 129, 131, 132, 139, Recall tasks, 45, 51–54, 69, 140, 146,
143, 147, 153, 157, 195–197, 201, 203, 151, 152
205, 238, 239, 245, 248, 249, 260, 263, Recency, 200
270, 272–275, 278–280, 282–288 Receptive, 167, 175, 177, 179, 238, 276, 278,
Prosody comprehension, 241 284–286
Prototypicality, 84 abilities, 175, 177, 238, 276, 279
Proverbs, 76, 80, 85, 99, 105, 107, 108, 123 bilingualism, 175–177
Pseudo-words, 62, 217 Receptive-only, 175
Psychiatric, 233, 234, 242, 246, 247 Recognition, 45, 51, 52, 82, 89, 91, 101, 105,
diagnosis, 228, 229, 248 129, 138, 140, 142, 143, 145, 152, 190,
disorders, 231 196, 279
symptom, 238 Reference, 10, 14, 21, 45–50, 57, 59, 61–63,
ward, 242 94, 95, 100, 103, 107, 123, 171, 174,
Psycholinguistic models, 214 246, 283, 284
Psychological, 41, 124, 158, 168, 174, 177, Referents, 5, 6, 13
231, 235, 238, 245, 246, 250, 274, 283 Referral, 249, 281, 287
dysfunction, 228, 229, 247 Regressions, 44, 86, 87, 94, 104, 201, 218,
processes, 28, 35, 231 219, 221
Psychopathology, 198, 205, 227–250 Regular orthographic systems, 214
Psychophysical task, 9 Regular words, 199
Psychophysiological, 56 Relational factors, 43
Psychotic, 234, 240, 242 Relative Proficiency Index (RPI), 286
Psychotypology, 180–183 Repetition effects, 138, 145, 146, 153
Puberty, 242, 244 Repetition tasks, 196
Publication bias, 130 Representation(al), 4, 6–21, 28, 29, 31, 32, 34,
Public school, 274 43, 60, 67, 69, 82, 85, 86, 88, 90–92,
Puerto Rico, 275 103, 104, 121, 123, 126, 131, 132, 138,
Punjabi, 230 147, 159, 161–163, 165, 169, 170, 173,
Puzzle assembly, 279 176, 178–183, 196, 200, 201, 212, 229,
240, 272
Reprimands, 45, 46, 51, 52, 56–59
Q Researcher biases, 130
Quadrilinguals, 29 Resting level of activation, 169, 170
Quantitative, 29, 66, 67, 231, 242, 278, 279, Resting levels, 168–171, 173, 174, 182
285 Retention hypothesis, 95, 100
Quechua, 176, 179–181 Retirement, 195, 196, 218
Retrieval
cue, 148, 149, 151
R failure, 148
Rapid naming, 218 Fluency, 285
Rapid serial visual presentation (RSVP), 55, 56 Retrieval-induced interference, 32
Reaching, 56, 283 Retroactive interference, 151
Reaction time (RTs), 54, 91 Reverse-iconic, 22
Reading RH. See Right hemisphere (RH)
routes, 214 Rhesus monkeys, 21
task, 17, 87, 216 Rhyming, 140
308 Index
Ribot’s law, 213 Self-reported, 34, 44, 121, 143, 263

Ribot’s rule, 213 Self-reports, 41, 42
Right hand, 17, 123, 213 Semantic(s)
Right-handed, 123, 213 access, 83
Right hemisphere (RH), 97–101 analyzability, 83
Right-to-left, 17, 18 anomia, 220
Right visual field, 5, 98 categorization tasks, 126
Risk factors, 192, 238 decomposition, 77
Rotated orthography, 18 dementia, 198–200, 219, 220
RPI. See Relative Proficiency Index (RPI) dementia subtype, 198, 200
RSVP. See Rapid serial visual presentation embeddedness, 28
(RSVP) knowledge, 190
Rum, 137 memory, 137, 138, 148, 198
Russian, 4, 5, 215, 232, 243, 276 system, 149, 198, 220
Russian–English, 31, 32, 232 verbal fluency tasks, 220, 273
Russian–German, 128, 215 Semantically-driven, 141, 144, 152
Sensations, 164, 178, 230
Sensorimotor, 6–8, 19–21
S Sensorimotor system, 6
Sadness, 41, 164, 229, 230, 235 Sentence repetition, 196, 199, 200, 285
Sagittal, 13 Sequential, 62, 69, 108, 174, 183, 271, 279
Salience view, 89, 108 Sequential bilinguals. See Successive
Saliency, 84, 89 Sequential second language
Salient, 52, 84, 87, 89, 94, 95, 98–101, 103, acquisition, 183
106, 107 Serial model, 79
Sapir, Edward, 4 Shadowing, 127
Sapir-Whorf hypothesis, 4 Shallow, 52, 53, 139–143, 147, 150, 216
Sarcasm, 76, 78, 99, 107 Shallowing processing task, 52, 140
Scales, 124, 143, 153, 230, 231, 233–236, Shallow task, 144
260–266, 276, 277, 279, 280, 283, 286 Shedding, 283
Scalp-recorded, 92 Shortcoming, 52, 88
Scanner, 68 Short-cuts, 28
Schemas, 33, 34, 84 Signifiers, 27
Schizophrenia, 233–235, 239–245 Sign language, 167
School environment, 197 Silence, 28
School psychologists, 274 Silent period, 175
School setting, 271, 282 Similarity, 7, 47, 53, 101–103, 105, 183
Scots, 217 Similarity-judgement, 7
SCRs. See Skin conductance response (SCRs) Simon
Second language, 11, 31, 32, 39, 44, 79, 84, effects, 49, 130
88, 101, 120, 121, 123, 127, 131, 138, task, 49, 55, 56, 70, 129, 130, 193
171, 175, 183, 195, 215, 227, 243, 246, Simple task, 121, 127, 193
247, 259, 266, 270, 271, 275, 286, 288 Simulation, 6
Segmentation, 285 Simultaneous bilinguals, 30, 60, 262, 271, 272
SEK, 14 Simultaneous processing hypothesis, 81
Selection by proficiency model, 121 Siniy, 5
Selective, 41, 214, 215 Situated environment, 6, 20
Selective attention, 192, 193 Situational, 43, 78
Self, 124, 160, 178, 191, 230–232 Skin conductance response (SCRs), 56, 57, 59,
Self-awareness, 240 60
Self-concept Slovenian–Italian, 123
Self-other, 232 Small samples, 130, 241, 244, 250
Self-paced reading, 87 SNARC. See Spatial-numerical association of
Self-perception, 247, 248 response codes (SNARC)
Index 309
SNC. See Special Nonverbal Composite Spatial prime, 8, 11

(SNC) Spatial priming paradigm, 11
Social Spatial reasoning, 278, 280
cognition, 236, 237, 239, 241, 249 Spatial thinking, 8
cognitive, 236, 237 Spatial working memory task, 124
context, 40, 53, 174, 238, 283 Spatiotemporal, 168
cultural, 43, 243, 270 Speaking, 3, 10, 29, 30, 79
factors, 232, 260 Special education, 273, 274
initiative, 260 Special Nonverbal Composite
integration, 201, 231 (SNC), 280
interaction, 28, 237, 239 Special theory, 97
medium, 28 Specific language impairment, 215, 216, 221
speech, 28, 35 Spectrum disorders, 239
Sociocultural, 43, 243 Spectrum of cognitions, 236
Sociocultural factors, 270 Speech
Socioeconomic status, 124, 130, 131, 194, 238 emergence, 282
Socioemotional, 238 production, 28, 167, 240
Sociolinguistic, 174, 246 Speech-related, 168
Sociolinguists, 177 Split visual presentation, 91, 98
Software, 88, 167 Spoken language, 40, 263
Somatosensory, 76, 172, 178 Spreading activation, 166, 171, 173
Somatotopy, 64, 68 Standardized, 15, 229, 245, 274, 275, 279,
Sound 280, 282, 283
Awareness, 285 Standard pragmatic model, 79, 80
Blending, 285 Standards for Educational and Psychological
Source, 7, 31, 76, 98, 177, 181, 236, 241 Testing, 274
South, 10, 14, 242 State-level, 283, 284
South Asians, 230 Statue of Liberty, 150
South Florida, 260, 262 Statue of Mao, 150
Space, 8, 12, 13, 15, 76, 91, 145, 158 Stimulus-locked, 92
Space-pitch, 9 Storage, 84, 86–89, 106, 107, 138, 147,
Space-time, 4, 6–13, 15–17, 19–21 149–151, 153, 159, 178–180
Spacing effect, 145 Store, 99, 106, 138, 139, 147, 152, 160, 161,
Spain, 40, 243, 275 163–165, 168, 169, 173, 174, 180
Spanglish, 270 Story Recall, 285
Spanish, 10, 17, 29, 30, 34, 40, 44, 53, 58, 60, Story sequencing, 279
66, 70, 90, 94, 139, 142, 146, 148, 152, Strokes, 198, 204–205
153, 164–166, 180, 213, 216, 217, 219, Strong bilingualism, 40
220, 235, 236, 238, 242, 243, 261–266, Stroop interference, 54, 55, 59
270, 273, 275–276, 284–286 Stroop task, 48, 54, 69, 126, 129, 130
Spanish–Catalan, 46, 53, 70, 127 Structural task, 141
Spanish-dominant, 94, 95 Subcortical, 97, 204, 244
Spanish–English, 30, 34, 39, 43, 44, 46–48, Submarinos Marinela, 148
51, 53, 58, 70, 93–95, 105, 123–126, Subtractive bilingualism, 237, 249, 272
142, 145, 150, 152, 153, 201, 212, 217, Successive, 40, 214, 241, 271
246, 260, 262–265, 284 Successive bilingual. See Sequential
Spanish–English-speaking, 284 Superior, 5, 46, 52, 65, 92, 97, 151, 195
Spanish-speakers, 61, 66, 196, 213, 264, 275, Supplementary motor area, 28, 127
286 Suppression, 79, 100, 126, 128, 129, 192, 193
Spanish-speaking, 35, 61, 176, 202, 217, 230, Suppression of language, 149
234, 242, 261, 270 Supramarginal gyrus, 28
Spanish subgroups, 270 Swearing, 29, 43
Spatial-numerical association of response Swear words, 43, 246
codes (SNARC), 12 Sweden, 219
310 Index
Swedish idioms, 104 The Broad Oral Language Cluster, 286

Swedish-speaking, 219 The Kid, 148
Switchers, 128 Thematic Apperception Test (TAT), 233, 235
Switching tasks, 128 Theoretical frameworks, 79–84, 138, 157–184
Syllabic, 218 Theory, 6–8, 21, 76, 77, 84, 97, 98, 106, 138,
Symbol copying, 279, 280 139, 158, 161, 169, 172, 199, 200, 240,
Symbol interdependency hypothesis, 21 244, 272, 275
Symptomology, 229, 230, 232, 240–246, 248 The parallel-processing model, 81–83, 103
Symptoms, 199, 201, 203, 228–230, 232, 238, Therapy, 41, 42
240–245, 247, 249, 250 Thought, 4, 7, 12, 20, 28, 35, 76, 119, 158,
Synonyms, 105, 138, 144 160, 174, 190, 228, 231, 233, 235, 237,
Syntactic 238, 240, 241
analysis, 83 processes, 228, 229
ellipses, 28 Three-dimensional, 15, 16
flexibility, 83 Three-dimensional space, 10, 12, 13, 15
processing, 82, 166, 182, 200 Three-stage model, 79
system, 162 Tics, 239
Syntactically ill-formed, 77 Time-moving perspective, 10, 16, 19
Syntagmatic-paradigmatic, 96 Time-pointing task, 14
Syntax, 162, 163, 165, 170, 182, 183, 190, Tip of the Tongue (TOT), 120, 190
204, 205, 243, 276, 286 Toddlers, 31, 286
System of systems, 158 Token ratio, 200
Tokens, 35, 174, 190
Topic, 41, 42, 60, 75, 77, 84, 130, 153, 158,
T 163, 167, 191, 246, 271
Taboo words, 42, 43, 45, 47, 51, 52, 54–59, Topographic, 14, 20
140, 141, 246 Top-to-bottom, 10, 12, 13, 18
Tagalog, 270 Toronto, 227
Taiwan, 13, 17, 18 Total reading time, 86, 94, 95, 104
Target domains, 7, 76 Tourette Syndrome, 239
Target language, 40, 102, 129, 196, 199 Tower of Hanoi, 128
Task, 5, 30, 44, 84, 119, 138, 159, 190, 216, Transfer appropriate processing, 138, 140,
239, 273 147, 149–153
Task switching, 128, 130, 193 Translation
mechanism, 121 direction, 141, 144, 145, 153
Teenagers, 260 equivalent, 90, 91, 104, 105, 139, 144, 145,
Temporal 152
cortex, 28, 204 task, 46, 52, 105, 140, 141
gyrus, 5 Transparency, 83, 101, 105, 107, 216, 221
lobe, 68, 213, 215 Transverse, 13, 21
relationship, 11 Traumatic experiences, 151
sequence, 12 Trilinguals, 29, 30, 121
Temporoparietal, 68 Turkish, 4, 52, 56, 58, 140, 276
regions, 204 Turkish–English, 45, 47, 51, 57, 59, 70, 140
Tenor, 77 TVIP. See Test de Vocabulario en Imágenes
Test de Vocabulario en Imágenes Peabody Peabody (TVIP)
(TVIP), 275, 276, 286 Twinkie, 148
Tests of Achievement, 275, 285 Type ratio, 200
Tests of Cognitive Ability, 275 Types, 28, 30, 32, 35, 41, 45–50, 53, 55–62,
Texas, 260, 265 69, 78, 89, 93, 96–99, 104, 106–108,
Thaayorre, 10 122, 123, 125, 140, 141, 145, 147,
Thai, 50, 55 152, 160–162, 167, 172, 173, 178,
Thai–English, 43, 51 179, 190, 191, 197, 198, 200, 204,
Index 311
212, 215, 217, 234, 248, 271, 274, Visual

282 field, 5, 98–100
Typologist, 180 module, 161, 179
Tzeltal, 14 processing, 51, 60, 168
structure, 160, 168, 179, 180
system, 160, 164, 167–168
U working memory, 168, 169, 194
Unanalyzed chunks, 89 Visual-conceptual interface, 182
Unbalanced bilinguals, 53, 67, 120, 121, 125, Visual-orthographic skills, 218
129 Visual Puzzles, 279
Uneducated bilinguals, 197 Visual–spatial skills, 275, 279, 280
UNIT-2. See Universal Nonverbal Intelligence Vocabulary
Test-Second Edition (UNIT-2) size, 271
United Kingdom, 230 test, 124
United States, 17, 18, 30, 196, 202, 213, 217, Vocabulary-richness, 200
227, 230, 260, 262–265 Voluntary, 167
Unitization Hypothesis, 81–82 Voluntary switching, 125
Universal, 4–5, 11, 76, 108, 216
Universal Nonverbal Intelligence Test-Second
Edition (UNIT-2), 277–279 W
Universal view, 4–5 Weak bilingualism, 40
University counseling center, 231 Weaker link hypothesis, 196, 197
Up/down, 8, 10, 13 Wechsler Individual Scale for Children-Fifth
Upper case, 89, 140, 144 Edition (WISC-V), 277, 279
USA, 43, 44, 262 Wechsler Nonverbal Scale of Ability (WNV),
US cultural identity, 263–265 277, 279
Wernicke’s area, 213
West, 10, 14
V Western-developed, 230
Valence, 6, 7, 41, 43, 44, 47, 51, 53, 54, Western English, 230
60–63, 66–69, 96, 97 White House, 78
Validity, 131, 132, 153, 233, 235, 236, 244, White matter, 28
249, 261, 264, 284, 287 Whole-word reading, 215
Vascular dementia, 198, 203, 205 WIDA. See World-Class Instructional Design
Vehicle, 76, 77, 84, 228, 229 and Assessment (WIDA)
Venetian, 130 WISC-V. See Wechsler Individual Scale for
Venetian–Italian, 130 Children-Fifth Edition (WISC-V)
Verbal Within-language, 100, 105, 138, 146, 147, 153
analogies, 238, 284, 285 Wolof–French, 243
fluency, 119–132, 192, 220, 221 Woodcock-Johnson III (WJ-III), 275, 285
fluency task, 122, 127, 220, 273 Woodcock-Johnson IV, Tests of Oral
memory, 213 Language (WJ-IV), 276, 285
processing, 191 Woodcock Muñoz Language Survey–Revised
production, 220 (WMLS-R), 276, 284, 285
thinking, 29 Word association task, 52, 140, 141
Verb+proposition, 92 Word-by-word, 76–78, 83, 87, 88, 90
Verb-proposition, 91 Word categories, 52, 56, 67, 139
Verbs, 85, 91–93, 96, 106, 108, 122, 183, 190 Word finding, 190, 199
Vertical Word-fragment completion task, 151, 152
metaphor, 12, 168, 169 Word-fragment identification task, 152
spatial, 7, 9–12, 19 Word frequency, 55, 143, 147, 148, 153, 200
Vietnam, 78 Word knowledge, 284
Vietnamese, 270, 276 Word-level, 78, 284, 286
312 Index
Word naming test, 197 systems, 16, 214, 215, 217, 218, 221
Word pairs, 22, 98, 100, 138, 144, 146, 191 tasks, 284
Word problems, 32–34 Written language, 182, 198, 211, 221
Word recognition task, 129
Word retrieval, 197, 198, 200, 284
Working memory (WM), 28, 124, 129, 130, Y
132, 168–171, 173, 191, 192, 194, 198, Years of education, 195, 217, 262
276, 277, 285 Young, 123–125, 129, 162, 175, 189, 193,
World-Class Instructional Design and 262–264, 271, 278, 279, 282, 286
Assessment (WIDA), 283 Younger adults, 190–192, 194
Writing Yupno, 14, 20

Psychology of Bilingualism The Cognitive and Emotional World of Bilinguals by Alfredo Ardila, Anna B. Cieślicka, Roberto R. Heredia, Mónica Roselli

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Psychology of Bilingualism The Cognitive and Emotional World of Bilinguals by Alfredo Ardila, Anna B. Cieślicka, Roberto R. Heredia, Mónica Roselli

Загружено:

Авторское право:

Доступные форматы

The Bilingual Mind and Brain Book Series 5

More information about this series at http://www.springer.com/series/13841

The Bilingual Mind and Brain Book Series

Library of Congress Control Number: 2017955224

© Springer International Publishing AG 2017

Printed on acid-free paper

This Springer imprint is published by Springer Nature

cal ability was learned. In Emotional Associations of Words in L1 and L2 in

Miami, FL, USA Alfredo Ardila

Roberto R. Heredia and Anna B. Cieślicka

Part I The Internal World of the Bilingual Person

Part II Bilingual Language Representation

Part III Bilingualism Across the Life-Span

Part IV Bilingual Cognitive and Personality Dimensions

Jeanette Altarriba University at Albany, State University of New York, Albany,

Anna B. Cieślicka received her Ph.D. from Adam Mickiewicz University in

Roberto R. Heredia is Regents Professor at Texas A&M International University

Mónica Roselli received her Ph.D. in biomedical sciences (neuropsychology)

Linguistic Relativity Hypothesis

Language is a communicative tool to convey information, such as to describe things,

© Springer International Publishing AG 2017 3

Conceptual Metaphor Theory

Conceptual Metaphors and Embodiment

Embodied cognition theories have different forms (Barsalou, 2008), such as

association, bodily experience (e.g., sitting erect) and perceptual characteristics

Linguistic Metaphors and Conceptual Metaphors

Linguistic metaphors are specific linguistic expressions, surface manifestations of

Linguistic Relativity on Conceptual Metaphors

A Case Study of Linguistic Relativity: Space–Time Metaphors

Space–Time Metaphors Across Languages

 oes Language Shape the Representation of Time: Empirical

One-Dimensional Timeline: Allocentric-Based

One-/Three-Dimensional Space–Time Metaphor

 xperience of Reading/Writing Directions and Space–Time

Even though the aforementioned evidence supported the influence of language on

Taiwan participants but a reversed pattern of results with US participants, language

In the previous sections, we introduced linguistic relativity, embodiment, and con-

space–time metaphors as a litmus test of the linguistic relativity hypothesis. In the

Role of Language in Time Representation

Time can be expressed in various ways, such as stationary as ego/object- moving

Role of Embodied Experience in Time Representation

Reading/writing direction is considered another kind of experience of language use

 rigin of Conceptual Metaphor: Preexisting or Acquired

from right to left in the movement conditions. In non-movement condition, singer’s

Linguistic Association or Embodiment

Fuhrman, O., & Boroditsky, L. (2010). Cross-cultural differences in mental representations of

What Is Inner Speech?

© Springer International Publishing AG 2017 27

Vygotsky (1934b/2012) systematized the concept of inner speech when he

Inner Speech in Bilinguals

tional) speech compared to early acquired languages. Significant differences in

Mental Calculation in Bilinguals

Calculation Abilities and Language

For some developmental cognitive psychologists, language is an essential part of

exact arithmetic was language specific, whereas approximate arithmetic depended

To further explore the specificity of language in bilinguals’ math learning,

 he Dynamic Process of the Inner Language of Mental

of the volunteers were college students who grew up in Spanish-speaking families

mental mathematics, particularly in cases in which education has been received in

Acknowledgement Our gratitude to Deven Christopher for her editorial support.

Johnson, J. R. (1994). Intrapersonal spoken language: An attribute of extrapersonal competency. In

Mónica Roselli, Idaly Vélez-Uribe, and Alfredo Ardila

© Springer International Publishing AG 2017 39

Emotional Associations in L1 and L2 Vocabulary

Miami, FL, USA Alfredo Ardila

Linguistic Relativity Hypothesis

Conceptual Metaphor Theory

Conceptual Metaphors and Embodiment

Linguistic Metaphors and Conceptual Metaphors

Linguistic Relativity on Conceptual Metaphors

A Case Study of Linguistic Relativity: Space–Time Metaphors

Space–Time Metaphors Across Languages

oes Language Shape the Representation of Time: Empirical

One-Dimensional Timeline: Allocentric-Based

One-/Three-Dimensional Space–Time Metaphor

xperience of Reading/Writing Directions and Space–Time

Role of Language in Time Representation

Role of Embodied Experience in Time Representation

rigin of Conceptual Metaphor: Preexisting or Acquired

Linguistic Association or Embodiment

What Is Inner Speech?

Inner Speech in Bilinguals

Mental Calculation in Bilinguals

Calculation Abilities and Language

he Dynamic Process of the Inner Language of Mental

Emotional Associations in L1 and L2 Vocabulary

Research Approaches for Studying Emotions in Bilinguals

The Emotion–Memory Effect: Recall Tasks

The Emotional Stroop and Other Emotional Interference Tasks

Autonomic Approaches: Skin Conductance Responses (SCRs)

lectrophysiological Approaches: Event-Related Potentials

euroimaging Approaches: Functional Magnetic Resonance

Conclusions and Future Research

Introduction: The Importance of Figurative Language

The Literal-First Model

The Figurative-First Model

The Parallel-Processing Model

The Graded Salience Model

Figurative Processing in L2 and Bilingual Language Users

Holistic Versus Compositional Storage and Retrieval

Literal or Figurative Salience in L2 Figurative Processing?

Neurological Bases of Bilingual Nonliteral Processing

L1 Effects in L2 Processing

Findings from Behavioral Measures

Numerous findings from work with behavioral tasks—measures of language-

Language-Specific Deficits in Switching Tasks

Fluency Effects Among Bilinguals

Age Effects Across Behavioral Measures

Effects of Early Versus Late Bilingualism

Effects of Aging Across the Lifespan

Insights from Neurocognitive Data

Switching and Executive Control

Support for a Bilingual Advantage in Executive Processing

Disputing the Bilingual Advantage in Executive Processing

Future Directions and Conclusions