AJSLP
Review Article
Effects of Age at Cochlear Implantation
on Learning and Cognition:
A Critical Assessment
Marc Marschark,a Louise Duchesne,b and David Pisonic
Purpose: Age at cochlear implantation frequently is assumed
to be a key predictor of pediatric implantation benefits, but
outcomes related to learning and cognition appear inconsistent.
This critical assessment examines relevant literature in an
effort to evaluate the impact of age at implantation in those
domains for individuals who received their devices as children.
Method: We examined 44 peer-reviewed articles from 2003
to 2018 considering age at implantation and conducted
statistical analyses regarding its impact on several domains,
including literacy, academic achievement, memory, and
theory of mind.
Results: Across 167 assessments in various experiments
and conditions, only 21% of the analyses related to age at
implantation yielded evidence in favor of earlier implantation,
providing greater benefits to academic achievement, learning,
or cognition compared to implantation later in childhood.
Among studies that considered cognitive processing (e.g.,
O
ver the past 20–30 years, cochlear implants (CIs)
and digital hearing aids have provided deaf chil-
dren with greater access to sound than ever
before. Since the first pediatric cochlear implantation in
1985, the primary focus of follow-up CI outcome studies
has been on spoken language, and there are now significant
bodies of literature concerning speech production (e.g., in-
telligibility, expressive vocabulary, narrative skills) and
speech recognition (e.g., phonemic awareness, speech recog-
nition, word repetition). The orientation toward research
a
National Technical Institute for the Deaf, Rochester Institute of
Technology, NY
b
Université du Québec à Trois-Rivières, Canada
c
DeVault Otologic Research Laboratory, Department of
Otolaryngology–Head and Neck Surgery, Indiana University School
of Medicine, Bloomington
Correspondence to Marc Marschark: marc.marschark@rit.edu
Editor-in-Chief: Julie Barkmeier-Kraemer
Editor: Susan Nittrouer
Received July 30, 2018
Revision received January 17, 2019
Accepted February 12, 2019
https://doi.org/10.1044/2019_AJSLP-18-0160
1318 American Journal of Speech-Language Pathology • Vol. 28 • 1318–1334 • August 2019 • Copyright © 2019 American Speech-Language-Hearing Association
executive function, memory, visual–spatial functioning),
over twice as many analyses indicated significant benefits
of earlier implantation when it was considered as a discrete
rather than a continuous variable.
Conclusion: Findings raise methodological, practical, and
theoretical questions concerning how “early” is defined in
studies concerning early cochlear implantation, the impact
of confounding factors, and the use of nonstandard outcome
measures. The present results and convergent findings
from other studies are discussed in terms of the larger
range of variables that need to be considered in evaluating
the benefits of cochlear implantation and question the
utility of considering age at implantation as a “gold standard”
with regard to evaluating long-term outcomes of the
procedure as a medical treatment/intervention for hearing loss.
Supplemental Material: https://doi.org/10.23641/asha.
8323625
on spoken language following early cochlear implantation,
frequently emphasizing the “early,” is understandable given
the strong everyday association between hearing, speech,
and language. Indeed, a primary motivation for parents’
opting for early cochlear implantation for their deaf child
is “because this would presumably provide the child with
the auditory information needed to more effectively develop
spoken language” (Christiansen & Leigh, 2002, p. 99; cf.
Duchesne, Sutton, & Bergeron, 2009; Niparko et al., 2010).
However, this article is not about early implantation1 and
spoken language development—or even language more
broadly—but about possible effects of age at implantation
on abilities that both underlie the acquisition of language
and are supported by it.
Beyond language per se, research concerning early
cochlear implantation has devoted relatively little attention
1
Age at implantation is used here so as to be consistent with the
relevant literature. In fact, age at activation is the important variable.
The two most often occur within 2–4 weeks of each other, but age at
activation is rarely reported (cf. Ching et al., 2014).
Disclosure: The authors have declared that no competing interests existed at the time
of publication.
to possible relations between cognitive processing and
factors associated with various degrees of hearing loss or
degrees of hearing restoration. That is, in addition to
supporting language development, providing deaf children
with greater access to environmental sounds, social com-
munication, and a wider variety of formal and informal
educational opportunities should have a significant impact
on their world knowledge, problem-solving skills, learning
strategies, and academic outcomes. This two-way street
thus clearly is dependent on the environmental/family con-
text of development (e.g., exposure to language, diversity
of conversational partners, parental income and educa-
tion, match of home and school language), but these are
issues rarely investigated with regard to the outcomes of
cochlear implantation (see Castellanos, Kronenberger, &
Pisoni, 2018; Holt, Beer, Kronenberger, & Pisoni, 2013).
For reasons that may not be entirely justified (an
interesting issue in itself, discussed later), age at implanta-
tion appears to have become the Holy Grail2 for clinicians
involved in pediatric cochlear implantation. In large mea-
sure, this focus in the associated literature stems from neu-
rophysiological research showing that development of the
auditory system is strongly influenced by the extent to
which auditory stimulation occurs within a critical period
after birth. It is only during this sensitive period—roughly
up to 3.5 years of age—that the neurological development
of the connections necessary for the efficient function of
the auditory system can occur. Auditory stimulation after
this period may result in those neurological connections
failing to develop or even be pruned, while existing con-
nections will not stabilize, all resulting in a subsequent
decrease in neural plasticity with age (e.g., Kral & Sharma,
2012). Simply put, longer periods of auditory deprivation
result in decreasing cortical plasticity over time with con-
comitant effects on the development of language and cog-
nition (Kral, Kronenberger, Pisoni, & O’Donoghue, 2016).
At the behavioral level, earlier access to speech supports not
only language development itself but also domains of cog-
nitive and social development in which deaf children fre-
quently lag behind their hearing peers (Knoors & Marschark,
2014). It therefore seems reasonable to suggest that the
earlier implantation occurs and the earlier a congenitally
deaf child is exposed to sound, the more “normal” (i.e.,
like typical, hearing children) will be neurophysiological
development underlying the auditory system and behavioral
development facilitated by it. Yet, as continuing research
indicates greater subtlety in developmental outcomes fol-
lowing early implantation, such as the distinctions of speech,
vocabulary, and language (see, e.g., Houston, Chen, Monroy,
& Castellanos, in press), assumptions about uniformity in
brain–behavior relations become more dubious, and the
Holy Grail of age at implantation as an omnipotent ex-
planatory variable becomes more questionable (Pisoni, 2016).
2
Although the “Holy Grail” metaphor is intended here primarily in
the sense of its denoting an object said to promise health, happiness,
and abundance, the allusion to a cup, dish, or other vessel holding
valuable (in this case, theoretical) contents also seems apropos.
Marschark et al.: Age at Implantation, Learning, and Cognition
With the focus of pediatric CI research on early speech
and language functioning, most relevant studies have involved
children of preschool and primary school age. This em-
phasis, in part, is a practical, methodological matter simply
because early pediatric implantation has not been widely
available long enough for the accumulation of a large em-
pirical literature involving long-term CI users (Ruffin,
Kronenberger, Colson, Henning, & Pisoni, 2013). In the
United States, for example, cochlear implantation for
children as young as 2 years of age was approved only in
2002, so the first cohorts of children who received CIs are
just now reaching the stages of secondary and postsecond-
ary education. At those ages, the language involved in
formal and informal learning (primarily basic interpersonal
communication skills or BICS) is relatively simple, at least
compared to the more complex language required for later
educational activities (involving cognitive–academic
language proficiencies or CALPs; Cummins, 1981). It thus
may well be that relations among cochlear implantation
variables, language, and other domains of development
found among children of preschool and primary school age
are not found among older children and young adults
(Marschark & Knoors, 2019). However, these are empirical
questions that we may not be ready to ask.
Evaluating Long-Term Outcomes
of Cochlear Implantation
Studies involving linguistic, cognitive, and academic
outcomes among secondary and postsecondary cohorts
of CI users now are underway. At least in the academic
domain, a number of such studies have indicated that CI
use is not as strongly related to academic outcomes as it
is in younger children. In fact, recent large-scale studies
with sample sizes of 500–980 deaf individuals from secondary
school (Crowe, Marschark, Dammeyer, & Lehane, 2017;
Marschark, Shaver, Nagle, & Newman, 2015) to postsec-
ondary settings and beyond (Convertino, Marschark,
Sapere, Sarchet, & Zupan, 2009; Dammeyer, Lehane, &
Marschark, 2017; Marschark, Spencer, et al., 2015) have
failed to find any long-term academic or cognitive advan-
tages for CI users over deaf peers who have never used
CIs (henceforth, nonusers). Such simple, dichotomous com-
parisons, however, may mask benefits linked to age at
implantation or duration of CI use (the latter, sometimes
referred to as hearing age, frequently unreported in published
studies).
In an effort to explore the extent of effects of age at
implantation on learning and cognition, we initially con-
ducted searches in diverse relevant journals and databases
for peer-reviewed research articles addressing academic
outcomes (e.g., reading, mathematics, standardized achieve-
ment scores) and various aspects of cognitive processing
(e.g., memory, problem solving, theory of mind). For the
present purposes, the search was limited to studies that
satisfied three criteria: (a) They explicitly addressed age at
implantation as a variable of interest, (b) data were analyzed
statistically, and (c) they were published in 2003 or later.
1319
That date was chosen because it appeared to be when rele-
vant outcome studies started utilizing rigorous methodologies
involving age at implantation (and, perhaps not coinciden-
tally, was the beginning of a series of influential studies
conducted by Geers and her colleagues; see below). The
result was the surprisingly small set of 44 empirical studies
considered here that met all three criteria. Some of those
studies included results from relatively small samples,
sometimes analyzed using parametric statistics better suited
to larger samples. Furthermore, even within the studies
considered here, there are difficulties for interpretation,
meta-analyses, or more fine-grained consideration of results
(e.g., findings described in text without corresponding
numerical/statistical information; IQ or duration of CI
use controlled in some analyses but not others). Although
such methodologies carry significant caveats, they appear
to represent the current “state of the art” in CI outcomes
research, and having been published in peer-reviewed journals,
they constitute the evidence frequently cited in support of
earlier cochlear implantation.
Within the 44 published studies considered in this
critical assessment, the majority of those reporting academic
outcomes focused on reading, sometimes only word read-
ing, most often as it related to various language measures.
The majority of studies reporting cognitive outcomes fo-
cused on executive function, often as it was related to other
cognitive processing. For convenience, the summaries be-
low (and in Supplemental Materials S1–S4) are organized
accordingly, with the first section including investigations
of literacy and academic achievement more broadly and
the second section including investigations of verbal and
nonverbal cognitive processes. When individual studies
examined academic and/or cognitive performance in more
than one domain, they are discussed in the one that was
reported of primary interest to the original investigators,
with only one or two exceptions. In addition, when the ini-
tial literature review had been completed, it appeared that
results of various studies differed depending on whether
age at implantation was considered a continuous variable
or a discrete variable. For that reason, the descriptions below
and more detailed information on each study included in
Supplemental Materials S1–S4 are organized first by category
(literacy/academic achievement or cognitive processing) and
then by whether age at implantation was treated as a con-
tinuous or discrete variable. Within those sections, studies
are organized primarily in chronological order, insofar as
scientific progress (and CI studies, in particular) builds on
previous work. The primary exception to that organization
is a handful of cases in which individual research groups
conducted multiple studies building on their own work.
Literacy and Academic Achievement: Age
at Implantation as a Continuous Variable
Geers and colleagues conducted a number of studies
involving children from an initial sample of 181 CI users
in the United States and Canada who were 8–9 years old
1320 American Journal of Speech-Language Pathology • Vol. 28 • 1318–1334 • August 2019
when tested between 1997 and 2000. The children had
received their implants within a restricted range, between
1;8 and 5;4 (years;months3; M = 3;4, SD = 0;10), and,
according to Geers, Brenner, and Davidson (2003), had
between 3;9 and 7;6 years;months of implant experience
(M = 5;6, SD = 0;9). Geers (2003) examined predictors
of reading ability in that sample, using the Reading
Recognition and Reading Comprehension subtests of the
Peabody Individual Achievement Test–Revised (PIAT)
and the Word Attack subtest of the Woodcock Reading
Mastery Test (WRMT). Neither age at implantation nor
duration of implant use was found to be significantly
related to reading outcomes. Geers (2004) examined the
extent to which age at implantation was related to speech,
language, and reading skills among 133 children from the
original sample. Age at implantation in the sample ranged
from 2 to 5 years (M = 3;6, SD = 0.9). Reading ability was
evaluated using the same tasks as the 2003 study, and
again, age at implantation was not significantly related to
performance on any of them.
Geers, Tobey, Moog, and Brenner (2008) retested
85 participants from the original Geers (2003) sample
when they were between 15 and 18 years of age, all of
whom were reported to have been implanted at essentially
the same age (M = 3.5 years, SD = 0.8). Rather than ex-
amining associations between age at implantation and indi-
vidual reading test scores, Geers et al. used weighted scores
composed of the total reading PIAT score, the WRMT
Word Attack score, and the total reading standard score
from the Test of Reading Comprehension. In contrast to
findings with the larger sample when the children were
younger, the weighted scores were significantly related to
age at implantation even when other factors were controlled.
The authors warned, however, that “implant age was
highly correlated with duration of implant use, which
accounted for a similar amount of variance when duration
replaced age at implant in the regression analysis” (Geers
et al., 2008, p. S29). Nevertheless, when considered separately,
duration of implant use was not significantly associated
with weighted reading scores.
Geers and Hayes (2011) retested 112 CI users from
the original sample when the children had reached 14.9–
18.5 years of age (M = 16.7, SD = 0.06). All had received
their CIs by the age of 5.4 years (M = 3.5). Reading ability
was assessed using the PIAT and the Test of Reading
Comprehension. Writing ability was assessed using a (picture-
based) spelling test and an essay-writing task. Rather
than considering age at implantation and duration of CI
use as separate predictors of reading and writing abilities,
Geers and Hayes evaluated the effects of age at im-
plantation by combining gender, duration of deafness
3
Various studies in this field provide ages either in the decimal format
common in most speech-language-audiology literature (e.g., 5.5 years)
or in the years;months format used in the psychology literature (e.g.,
5;6). The reporting format from the original studies has been retained
here. Detailed demographics and results, including all ages in decimal
format, are included in Supplemental Materials S1–S4.
(age at implantation to age of deafness onset), and age of
first hearing aid fitting in a multiple regression analysis.
Together, those variables accounted for only 8.6% of the
variance in composite literacy scores (based on factor anal-
yses); the authors concluded that “a shorter duration
between the onset of deafness (birth in most cases) and CI
surgery was associated with a higher level of literacy”
(p. 57S).
Stacey, Fortnum, Barton, and Summerfield (2006)
reported a study in which parents and teachers of a “re-
presentative sample” of 8,876 deaf children (in the United
Kingdom) initially were invited to complete surveys asking
about children’s educational achievements, quality of life,
spoken language skills, and auditory performance. Surveys
were returned by 468 parents and 383 teachers of CI users.
The sample of children with implants was divided into six
groups based on age at implantation (< 5 years and ≥ 5 years)
and length of implant use (< 2 years, ≥ 2 and < 4 years,
and ≥ 4 years). At the time of test, ages ranged from 3.3 to
20.6 years. There were five measures of educational achieve-
ment: (a) parents’ and (b) teachers’ ratings of children’s
academic abilities (in reading, writing, number, time, money,
and measurement) and (c) teachers’ ratings of children’s
key stage attainments (in reading, writing, mathematics,
and science), (d) reading age, and (e) participation/engage-
ment. Although the Stacey et al. results are not reported
in a way that allows distinguishing between effects of age
at implantation and duration of CI use, the investigators
reported that “beyond the domains of auditory perfor-
mance and speech perception, few statistically significant
advantages of implantation [i.e., with regard to educational
achievement and quality of life] were found unless children
had received implants before the age of 5 years and had used
implants for more than 4 years” (p. 177) and “it is hard,
nonetheless, to reconcile these results with the conclusion that
children with implants display literacy skills that are similar
to those displayed by their normal-hearing peers” (p. 170).
In contrast to most other studies involving older deaf
learners (Marschark & Knoors, 2019), Uziel et al. (2007)
reported that cochlear implantation was associated with
“satisfactory academic performance” in a group of 82 CI
users, 12–20 years old, who had received CIs 10 years earlier.
Uziel et al. reported that 37% of the sample had no aca-
demic delay, indicated by the fact that they were in age-
appropriate classrooms; the rest were reported to have
academic delays of 1–3 years. The investigators included
age at implantation in analyses of language measures, but
they apparently did not do so in analyses of their aca-
demic measures. Nevertheless, relations between age at
implantation, duration of CI use, and academic scores in
math, French, and foreign language can be computed using
information provided in their Table 1 (age at implantation,
age at test) and Table 3 (teacher-rated academic perfor-
mance), at least for students in junior high school (n = 33)
and high school (n = 12). According to results in those tables,
neither age at implantation nor duration of CI use was sig-
nificantly correlated with scores in any of the three academic
domains (see Supplemental Materials S1–S4).
Marschark et al.: Age at Implantation, Learning, and Cognition
Archbold et al. (2008) examined long-term effects of
pediatric cochlear implantation in a study of reading ability
among 105 children who had received their CIs between
16 and 83 months of age (M = 50 months). Reading was
assessed using the Edinburgh Reading Test 7 years after
the children had received their implants, when they were
between 8 and 13 years of age (M = 11;1). Seventy-seven
of the children also had been old enough to have the read-
ing skills evaluated previously, at 5 years after implanta-
tion. A “net reading age” was calculated as the difference
between the children’s chronological ages and their reading
ages. Archbold et al. found strong negative correlations
between age at implantation and net reading age at both 5
and 7 years postimplantation, correlations that remained
strong and significant when nonverbal intelligence was par-
tialed out. Archbold, Nikolopoulos, and O’Donoghue
(2006) further reported that, within the full sample, close
to 50% of the children who had been implanted between 1
and 3 years of age were within 1 reading year (±) of their
chronological age; slightly fewer showed a delay of more
than 1 year. Among children implanted between 4 and 5
years of age, 80% were at least 1 year delayed in reading
ability, and 100% of the children implanted between 6 and
7 years of age were at least 1 year delayed.
Ching, Day, and Cupples (2014) evaluated the impact
of age at implantation on word reading skills in the context
of a study of phonological awareness following cochlear
implantation. Their study included 165 five-year-olds who
were using CIs, 32% of whom had received them before
12 months of age. Word reading was evaluated using the Word
Identification and Word Attack subtests of the Woodcock-
Johnson III Diagnostic Reading Battery (WJ-III DRB).
Multiple regression analyses involved 62 of the 165 children
who had “full sets of scores” and used measures of phono-
logical processing, receptive and expressive language, and
receptive vocabulary as well as nonverbal cognitive ability
as predictor variables. Results indicated that earlier im-
plantation was a significant predictor of “word reading
ability” (p. S28), presumably WJ-III DRB Word Identification
scores.
Sarant, Harris, and Bennet (2015) examined the
potential impact of early unilateral and bilateral implanta-
tion on academic achievement in mathematics, reading,
and writing among 44 eight-year-olds, all of whom had
received CIs by the age of 3.5 years. Achievement was
assessed using the Wechsler Individual Achievement Test–
Second Edition. Although children with bilateral implants
scored higher than those with unilateral implants across
all three academic areas, age at implantation was signifi-
cantly related only to Wechsler Individual Achievement
Test–Second Edition scores in reading and not mathematics
or writing. That significant association was eliminated
when other factors were controlled through multiple regres-
sion analyses.
In summary, the eight studies described above included
15 correlations between performance and age at implan-
tation. Five of those analyses (33%) yielded significant (nega-
tive) coefficients indicating benefits of earlier implantation.
1321
If Geers and Hayes’ (2011) “duration of profound deaf-
ness” predicting weighted literacy scores is included,
19 analyses yielded nine (47%) significant effects indicating
benefits of earlier over later implantation. Reported correla-
tions between age at implantation and performance ranged
from −.24 to .12.
Literacy and Academic Achievement: Age
at Implantation as a Discrete Variable
Connor and Zwolan (2004) examined the reading
comprehension skills of 91 CI users in primary school. The
children had an average age of 10.98 years (SD = 2.67)
and had received their implants at an average age of 6.78 years
(SD = 3.06). Forty of the children had received their im-
plants at 5 years of age or earlier; 51 had received them later.
Connor and Zwolan found scores on the WRMT Passage
Comprehension subtest significantly correlated with age at
implantation, but not duration of implant use. Children in
the later-implanted group evidenced a greater disparity
between their reading achievement and that of hearing peers
compared to children in the earlier-implanted group, indi-
cating that the reading gap increased with age (and perhaps
the lessening of an effect of age at implantation). The in-
vestigators concluded that “early implantation provides
access to spoken English during the important early language
development phase, and the resulting stronger language
skills would tend to support stronger reading skills” (p. 520).
Importantly, Connor and Zwolan noted that, in addition
to having higher reading comprehension scores, children
who received their CIs at younger ages also had stronger
preimplant and postimplant vocabulary skills, which were
associated with better reading comprehension. Such find-
ings emphasize the importance of preimplant language abil-
ities for postimplant language and academic outcomes.
Preimplant language abilities also may be a confounding
factor with regard to age at implantation, but that variable
rarely is reported in the literature.
James, Rajput, Brinton, and Goswami (2007) were
not specifically investigating reading but examined syllable,
rhyme, and phoneme awareness in an early-implanted
group of nine children who received CIs between 2 and
3.6 years of age and a late-implanted group including 10 chil-
dren who received CIs between 7 and 10 years of age.
Both groups were compared to reading-matched hearing
groups. The deaf children, all of whom had been using
their CIs for at least 3 years, were tested twice, 12 months
apart, on the three spoken language assessments. Word
reading was evaluated using the Test of Word Reading
from the British Ability Scale. Age at the first testing ranged
from 7.6 to 9.1 years. Reading growth rates were similar
for early- and late-implanted groups. Word Reading scores
for the early-implanted group were within 1 SD of the
hearing mean at both testing times, whereas scores of the
late-implanted group were more than 1 SD below the hear-
ing mean at both times. The difference in standard scores
between the CI groups was significant at the first testing
1322 American Journal of Speech-Language Pathology • Vol. 28 • 1318–1334 • August 2019
(95.00 vs. 89.11) but not at the second (81.90 vs. 81.33). The
investigators concluded that the early-implanted group was
less delayed in word reading than the late-implanted group
relative to their age-matched hearing peers, but the differ-
ences clearly were not large, and the sample sizes were quite
small compared to other studies reviewed above.
In a subsequent study, Johnson and Goswami (2010)
specifically were interested in relations between phonological
awareness and reading ability in children with CIs. Their
study included 39 primary school CI users between the
ages of 5 and 15 years. Early- and late-implanted samples
were established by creating groups of 20 and 19 children
who had received their implants before and after 39 months
(M = 31 months, SD = 5.8, and M = 59 months, SD =
11.5), respectively. CI users with nonverbal IQs outside
the normal range were excluded from the “main analyses.”
Mean durations of implant use at the time of testing
were roughly equal: 79 months (SD = 27.5) for the early-
implanted group and 73 months (SD = 24.5) for the later-
implanted group. Reading comprehension was evaluated
using the Neale Analysis of Reading Ability–Revised
(NARA-R), which involves a series of short texts of in-
creasing complexity. Analyses indicated no significant
differences among the two CI groups and groups of hear-
ing aid users and hearing peers. The two CI groups also
did not differ significantly from each other in their stan-
dard scores or reading ages, although both were higher
than the hearing comparison group. When Johnson and
Goswami created “quotient scores” (i.e., reading age/
chronological age), children in the earlier-implanted group
obtained higher scores than those in the later-implanted
group.
DesJardin, Ambrose, and Eisenberg (2009) conducted
a longitudinal study involving 16 preschool and primary
school children who were between the ages of 2.7 and
6.3 years at the initial testing and were tested again 3 years
later (at 5.7–9.3 years), when they were in primary school.
Their mean age at implantation was 25.9 months (SD =
6.93), and their mean length of implant use was 26.1 months
(SD = 10.98) at the first testing and 63.4 months (SD =
10.68) at the follow-up test. Assessments of reading using
the WJ-III DRB indicated no significant difference in
reading scores between groups of eight earlier-implanted
(M = 1.74 years) and eight later-implanted (M = 2.58)
children at follow-up testing, the only point at which the
relationship was reported.
In a follow-up to the Uziel et al. (2007) study described
in the previous section, Venail, Vieu, Artieres, Mondain,
and Uziel (2010) investigated educational and employment
outcomes in a sample of 74 CI users (“without additional
disabilities,” p. 67) in a larger sample of 100 CI users. All
had received their implants before the age of 6 years (M =
44.1 months, SD = 1.4) and had used them for at least
4 years (M = 10.6 years, SD = 0.04). Parents and children
were interviewed by telephone concerning schooling,
employment status, age at acquiring literacy skills (to
determine any delays), and grade failures. Age at implan-
tation was considered by dividing the children according
to whether they had received their implants prior to the
age of 3 years or at 3 years of age or later. The only result
reported with regard to age at implantation and academic
achievement was that there was no association between age
at implantation and grade failure.
López-Higes, Gallego, Martín-Aragoneses, and Melle
(2015) examined reading comprehension abilities of 38 pri-
mary school children, 8–12 years of age, 19 of whom re-
ceived their CIs before 24 months of age and 19 of whom
received them after 24 months of age. The two groups were
approximately the same age at testing (9.6 and 9.9 years,
respectively) and thus differed in duration of implant use.
A Spanish sentence comprehension task (ECCO) was adapted
for use with primary school children. The ECCO-Prima
assessed grammatical comprehension using sentence–picture
pairs in which, for each picture, one sentence matched the
picture, one was a syntactic foil, and one was a lexical foil.
It also assessed “thematic role assignment” (depending
primarily on word order) in four types of Spanish sentence.
In addition, the investigators developed a test of readers’
awareness of Spanish nominal and verbal inflectional mor-
phology (Morphological Awareness test). The ECCO-Prima
yielded significant differences between the earlier- and
later-implanted groups on two of the four sentence types:
sentences that fit the canonical word order of Spanish and
those with only one proposition. On the Morphological
Awareness test, children who received their CIs earlier
outperformed those who received them later in their under-
standing of both nominal and verbal morphology. In addi-
tion to the differences in duration of CI use, however, the
investigators cautioned that more than twice as many chil-
dren in the early-implanted group had bilateral implants
compared to the late-implanted group, creating difficulty
for fully explaining their results (see Sarant et al., 2015).
Gallego, Martín-Aragoneses, López-Higes, and Pisón
(2016) evaluated reading comprehension abilities in the same
children as López-Higes et al. (2015) using an investigator-
developed Semantic and Syntactic Strategies Detection Test,
a cloze task that required providing the last word in sen-
tences from several alternatives that included syntactic and
semantic foils. The results showed that the early-implanted
group performed at a level essentially the same as the hear-
ing comparison group, whereas the late-implanted children
performed significantly worse than the hearing comparison
group overall. Considering four item types (short and long
sentences with frequent or infrequent target words), the
early- and late-implanted groups differed significantly with
short sentences but not the long sentences (see also Barajas,
González-Cuenca, & Carrero, 2016).
Colin, Ecalle, Truy, Lina-Granade, and Magnan
(2017) investigated the effects on deaf children’s literacy
skills of age at cochlear implantation and exposure to cued
speech, a supplement to spoken language, supporting it
with a set of sound-related handshapes in particular positions.
Ninety deaf children between the ages of 6.9 and 10.6 years
were divided into younger and older groups and, within
each, early-implanted versus late-implanted groups as well
as younger and older groups in terms of when they are exposed
Marschark et al.: Age at Implantation, Learning, and Cognition
to cued speech. Using a cloze sentence completion test similar
to that of Gallego et al. (2016), Colin et al. found signifi-
cant advantages for earlier implantation on sentence com-
prehension among both younger and older children. Colin
et al. indicated that the groups “were matched on other
variables that might also be predictors of outcome after im-
plant fitting and exposure to CS” (p. 64), but information
provided in their Table 1 indicates that children in the
earlier-implanted group had durations of implant use 30%–
60% longer than those in the later-implanted group. As in
several other studies described here, the contributions of
age at implantation and duration of CI use are thus unclear.
Reading ability and reading strategies used by deaf
children with and without CIs were explored by Domínguez,
Carrillo, González, and Alegria (2016). A group of 136 chil-
dren included seventy-seven 6- to 18-year-olds who had
received CIs between 8 months and 11 years of age. Chil-
dren in an early-implanted group had received their CIs at
an average age of 1.68 years (SD = 0.53); those in a late-
implanted group had received their CIs at an average age
of 5.41 years (SD = 2.57). Duration of implant use was
roughly comparable, 6.78 and 6.98 years, respectively.
Using a “readability test (READ)” to assess global reading
levels, Domínguez et al. found that the early-implanted
group did not differ significantly from a hearing compari-
son group, showing only a half-year delay; the late-implanted
group was significantly delayed relative to the hearing
group by an average of 1.25 years. However, the earlier-
and later-implanted groups did not differ significantly.
Finally, a study by Harris and Terlektsi (2010) inves-
tigated the impact of hearing aids and CIs on reading and
spelling abilities in a sample of eighty-six 12- to 16-year-olds.
That sample included 59 children with bilateral CIs and
another 27 who used hearing aids. The CI users were
divided into early-implanted (prior to 42 months) and later-
implanted (later than 42 months) groups. Reading abilities
were assessed using the Single Word Reading Test from
the British Ability Scale and the Edinburgh Reading Test.
Scores on the Single Word Reading Test indicated little
difference among the three groups. On the Edinburgh Test,
scores indicated smaller delays in reading ability from the
early-implanted group to the later-implanted group to the
hearing aid group. However, the early- and later-implanted
groups did not differ significantly from each other.
Nine of 20 comparisons (45%) involving earlier- and
later-implanted groups of children in the nine studies de-
scribed in this section indicated better performance by the
former group. If the Johnson and Goswami (2010) compari-
sons involving quotient scores rather than standard scores
are included, still only 10 of 21 comparisons (48%) indi-
cated significant benefits of earlier implantation.
Interim Summary
By way of an interim summary, 18 of the studies
described above included 40 assessments of the impact of
age at implantation on reading and other academic out-
comes. Only about half (19) of those assessments (48%) were
1323
statistically significant, a nonsignificant proportion according
to a sign test. Those results suggest that age at implantation is
not a strong predictor of literacy/academic benefits. As noted
earlier, however, two studies create some difficulty for in-
terpretation. Instead of considering age at implantation as
a separate variable, Geers and Hayes (2011) used it to
create a “duration of deafness” variable, which resulted
in significant findings. Johnson and Goswami (2010) failed
to find significant effects of age at implantation in stan-
dard reading scores (or reading age) on their NARA-R assess-
ment, but when they created “quotient scores,” both of
those effects were statistically significant. If those two stud-
ies are removed from calculations, 14 of 35 assessments
(40%) in 16 studies were statistically significant, still indicating
that age at implantation is not a significant predictor of
literacy/academic benefits, at least to the extent that it can
be determined from relevant studies in the literature.
It was suggested earlier that combining studies in
which age at implantation was considered a continuous
and discrete variable has the potential to obscure differences
resulting from the two statistical methodologies. Eight
studies described above in which age at implantation was
considered a continuous variable yielded nine significant
findings out of 19 assessments (47%), indicating benefits of
earlier ages at implantation or five of 15 (33%) if the Geers
and Hayes (2011) study is excluded. Only 10 correlation
coefficients were reported among those 19 assessments,
yielding a mean correlation coefficient of .25 (a mean
reduced to .14 if correlations involving the Archbold et al.,
2008, “net reading age” scores are eliminated). The 10 stud-
ies in which age at implantation was considered a discrete
variable yielded 10 of 21 significant findings (48%) or nine
of 20 (45%) if the Johnson and Goswami (2010) study is ex-
cluded. The average of 17 effect sizes that could be calculated
from data provided by authors was .33 (or .32 if NARA-R
standard scores rather than quotient scores are used for the
Johnson and Goswami study; see Supplemental Materials
S1–S4 for statistical details). Setting aside concerns about
several authors’ caveats about positive and negative find-
ings, possible confounding factors, and some unusual out-
come measures, these results, in aggregate, indicate that
when different conditions were taken into account, the asso-
ciation between earlier implantation and better academic
achievement is weak at best. However, at least in this
domain, treatment of age of implantation as a continuous
versus discrete variable does not appear to affect outcomes
to any great extent. We therefore now turn from academic
functioning to aspects of cognitive processing likely to
underlie it and just as likely to depend (implicitly or explicitly)
on audiologic and language factors.
Cognitive Processing: Age at Implantation
as a Continuous Variable
Executive Function
Executive function or executive functioning refers to a
group of higher level cognitive processes that coordinate
1324 American Journal of Speech-Language Pathology • Vol. 28 • 1318–1334 • August 2019
and control behaviors including attention, learning, social
interactions, and other internal and external behaviors (e.g.,
working memory, inhibition, visual–spatial processing,
processing speed). Research involving deaf children and
adults, both CI users and nonusers, has found that they
frequently demonstrate greater difficulties in some of these
executive function domains but not others relative to hear-
ing age-mates (e.g., Kronenberger, Beer, Castellanos, Pisoni,
& Miyamoto, 2014; Pisoni, Conway, Kronenberger, Henning,
& Anaya, 2010). For the present purposes, the issue is the
extent to which various executive functions are affected by
age at implantation.
Kronenberger, Pisoni, Henning, and Colson (2013)
investigated executive functions among 53 children, adoles-
cents, and young adults (M = 14.4 years, SD = 4.1) who
had received CIs prior to 7 years of age (Mage = 34.8 months,
SD = 19.3) and had used their CIs for at least 7 years
(M = 11.5 years, SD = 3.2). Three areas of executive func-
tioning were explored: short-term/working memory (seven
verbal and nonverbal tasks), fluency–speed (five verbal
and nonverbal tasks), and inhibition–concentration (five
measures across three verbal and nonverbal tasks; see
Supplemental Materials S1–S4). None of the performance
scores was found to be significantly related to age at im-
plantation; only performance on one of the inhibition–
concentration measures (a Stroop task) was related to
longer CI use. Kronenberger, Colson, Henning, and Pisoni
(2014) further examined the same three areas of executive
function in a sample of 64 CI users between the ages of 7.8
and 27.4 years (M = 15.0, SD = 4.9). The participants all
had received their CIs between 8.3 and 75.8 months of age
(M = 35.6, SD = 19.6) and had used them for 7.1–22.4 years
(M = 12.1, SD = 3.9). Tasks included five of the verbal
and nonverbal short-term/working memory used by
Kronenberger, Pisoni, Henning, et al. (2013), three of their
fluency–speed tasks, and the three measures of inhibition–
concentration. Neither age at implantation nor duration of
implant use was significantly related to any of the executive
function measures.
In a study involving preschool-aged children with
CIs, Beer et al. (2014) used the parent report Behavior Rating
Inventory of Executive Function to investigate three aspects of
executive function expected to be affected by delays in speech
and language: working memory, inhibition–concentration,
and organization–integration (i.e., executive–organizational–
integrative processes). The 24 children had received their
CIs between 10 and 36 months of age (M = 20.01, SD =
7.87) and had used them for 0.53–5.18 years (M = 2.73,
SD = 1.14). Age at implantation was unrelated to the
three executive function measures. Only the organization–
integration measure was significantly related to the duration
of CI use, as longer use was associated with fewer planning/
organization difficulties on the Behavior Rating Inventory
of Executive Function.
Marschark, Spencer, et al. (2015, Experiment 3)
examined the extent to which CI use and age at implantation
might affect eight dimensions of executive function among
deaf college students. Executive function was assessed using
the Learning, Executive, and Attention Functioning (LEAF)
Scale (Kronenberger, Beer, et al., 2014). Deaf participants
included 32 CI users who received their CIs at a mean age
of 6.9 years (SD = 5.1). CI users and nonusers did not differ
significantly on any of the eight LEAF scales, and none of
the correlations between age at implantation and LEAF
subscales approached significance.
Findings with regard to possible associations between
cognitive processes related to executive function and age at
implantation are easily summarized: In the 42 comparisons
made across four studies, all involving age at implantation
as a continuous variable, none yielded a significant effect.
Theory of Mind
Theory of mind is a central aspect of social cognition,
reflecting the understanding that individuals have wants,
beliefs, and preferences. Theory of mind has been found to
develop with increasing complexity consistently across
languages and cultures, but deaf children and youth typically
demonstrate delays in that domain. Both theory of mind
and delays in theory of mind most frequently are explained
in terms of a conversational account focusing on the amount
of social communication experienced by children both
directly (conversationally) and indirectly (via overhearing;
Courtin, 2000; Peterson & Siegal, 1995). Because CIs provide
deaf children with access to spoken language, one might
expect that CI users would demonstrate normal theory of
mind development or, at least, reduced delays. Bat-Chava
and Deignan (2001) and Most, Ingber, and Heled-Ariam
(2012), however, argued that CI users may not have suffi-
cient access to the subtleties of spoken language needed for
social interactions to provide the foundation for theory of
mind development (e.g., prosody, sarcasm).
Remmel and Peters (2009) examined relations between
language and theory of mind among 30 deaf children with
CIs, aged 3.1–12.0 years, who had received their implants
between 1.1 and 6.0 years of age (M = 2.9, SD = 1.4) and
had used them for 1.2–9.0 years. They used the five-item
version of Wellman and Liu’s (2004) theory of mind scale,
which taps diverse desires, diverse beliefs, knowledge
access, contents false belief, and real and apparent emotion.
Remmel and Peters also administered four other theory of
mind tasks: a hiding and finding (false belief ) game, a false
photograph task, a memory for complements task, and an
explanation of action task. Age at implantation was not
significantly correlated with performance on any of the five
theory of mind tasks. Longer duration of CI use was signif-
icantly related to better performance on only one of the
theory of mind scales: explanation of action.
Ketelaar, Rieffe, Wiefferink, and Frijns (2012) ob-
tained similar results in a study involving seventy-two 1- to
5-year-old CI users. All but one of the children were reported
to have received their implants between 6 and 39 months of
age and had used them for an average of 19.04 months
(SD = 12.67). Theory of mind was examined using an unex-
pected location false belief task, an intention-understanding
task that tapped children’s understanding of individuals’
Marschark et al.: Age at Implantation, Learning, and Cognition
(unsuccessful) intentions to manipulate physical objects,
and a common and uncommon desires task in which chil-
dren had to identify others’ food preferences that were the
same (common) or different (uncommon) from their own.
Neither age at implantation nor duration of CI use was
significantly related to scores in the desires task. (The
unexpected location false belief task could not be scored
because all CI users failed it, and correlations with the
intention-understanding task were not reported.)
Finally, Marschark, Edwards, Peterson, Crowe,
and Walton (2019) investigated theory of mind among
deaf and hearing college students, using advanced theory
of mind tasks tapping understanding of sarcasm, double
bluffs (i.e., telling the truth to someone who expects you to
lie), and second-order false belief tasks (“What does he
think that she believes?”). Their 94 deaf participants included
46 CI users who reported a mean age at implantation of
6.62 years (SD = 4.84). CI users’ performance on all three
theory of mind tasks was unrelated to age at implantation
or duration of CI use.
In short, there appears to be little evidence that CIs
provide sufficient hearing to support normal trajectories in
the development of theory of mind, nor is age at implanta-
tion associated with its development. Nine evaluations of
age at implantation were made across the three studies
described here, in which age was considered as a continuous
variable. Age at implantation was not significantly related
to theory of mind performance in any condition in any of
these studies.
Other Cognitive Processes
Willstedt-Svensson, Löfqvist, Almqvist, and Sahlén
(2004) examined language, novel word learning, and mem-
ory in a group of 15 deaf children, 5–11 years of age,
who had received CIs between 2 and 6 years of age (M =
4.03 years, SD = 0.09) and had used them for at least 26 months
(M = 4.09 years, SD = 0.11).4 Beyond the language mea-
sures, children were administered tests of novel word learning
and retention and a complex working memory task. In the
latter, children had to provide acceptable words to complete
each of three sentences and then four sentences and then
recall those words after each series (i.e., three words and
then four words). Age at implantation was not significantly
related to working memory performance. It was positively
correlated, however, with measures of both novel word
acquisition and retention, apparently indicating better
learning with “later” implantation (p. 511), but the authors
concluded that their results “seem to corroborate results
from earlier studies showing that implantation at a younger
age is important for the language development of children
with CIs” (p. 512). Neither working memory performance
nor word learning was related to duration of CI use.
4
Age at implantation and duration of CI use associated with the
Willstedt-Svensson et al. (2004) study are drawn from their Table 1,
which differ from statements in the text.
1325
 Fagan, Pisoni, Horn, and Dillon (2007) examined a
series of neuropsychological, memory, and reading mea-
sures with a group of twenty-six 6- to 14-year-olds who had
received CIs between 1 and 6 years of age (M = 2.5 years,
SD = 1.3). Their test battery included four sensorimotor
(hand/finger movement) tasks, a visual–spatial design copy-
ing task, forward and backward digit spans, and the
WRMT Word Attack subtest. Only backward digit span
(typically considered a test of working memory) was signifi-
cantly related (negatively) to age at implantation; none of
the performance scores was related to duration of implant
use. The investigators suggested that the lack of reliable
findings with regard to age at implantation and duration
of CI use may have been related to a relatively large vari-
ability in those variables.
Conway, Pisoni, Anaya, Karpicke, and Henning
(2011) tested 25 children who used CIs in a series of novel
tasks designed to tap implicit sequence learning (verbal
and nonverbal memory also were tested, but performance
was not examined with regard to age at implantation). The
children, aged 61–118 months (M = 90.1, SD = 9.9), had
received their implants between 10 and 39 months of age
(M = 21.2, SD = 8.3) and had used them for 36–98 months
(M = 68.9, SD = 19.4). The implicit sequence learning task
required the children to learn an artificial grammar created
using colored squares by reproducing sequences following
visual presentation (i.e., the Simon game). Following a
learning phase, similar sequences were presented from which
the investigators could determine children’s accuracy in
identifying “grammatical” from “ungrammatical” test
sequences. Conway et al. reported that, when chronological
age was controlled, age at implantation was marginally
related to implicit sequence learning scores (better performance
associated with earlier implantation) and duration of im-
plant use (better performance associated with longer use),
both ps = .058, but possible associations of sequence recall
with age of acquisition were not reported.
Relations between memory scores and age at implan-
tation were reported in a study by Kronenberger, Pisoni,
Harris, et al. (2013) in a 2-year study investigating the ex-
tent to which verbal working memory might predict speech
and language development in a sample of 66 CI users who
were 6.0–11.5 years of age (M = 7.6, SD = 1.4) at the first
testing and 8.1–16.1 years of age (M = 11.4, SD = 2.2) at
the final testing. All of the children had received CIs prior
to 8 years of age (M = 3.8, SD = 1.7). Memory tests in-
cluded forward and backward digit spans, and participants
were divided into clusters based on their forward span and
backward span performance. Performance on those tasks
was found to be unrelated to age at implantation (see also
Mikic et al., 2014).
Nicastri et al. (2014) investigated two higher level
cognitive skills in the verbal domain: metaphor comprehen-
sion and the ability to make inferences during conversa-
tions. Both tasks were drawn from the Italian Pragmatic
Language Skills MEDEA. The study included 31 CI users
between 6.3 and 14.6 years of age (M = 8.5, SD = 2.1).
Fifteen of them had received their CIs before 24 months
1326 American Journal of Speech-Language Pathology • Vol. 28 • 1318–1334 • August 2019
of age; the remainder received CIs between 24 and 48 months
of age. Inferencing scores on the Implicit Meaning Com-
prehension and Situations subtests indicated that the CI
users performed no differently than a comparison group
of age-matched hearing peers, whereas the CI users obtained
significantly lower scores on the Metaphors subtest (“verbal”
and “figured” metaphors) than their hearing peers. Among
the CI users, earlier implantation was associated with bet-
ter performance on both of the Metaphors tasks and on
two of three Implicit Meaning Comprehension items (not
individually described) but not the Situations subtest.
Mancini et al. (2015) examined referential communi-
cation skills in the same group of 31 CI users as Nicastri
et al. (2014). Referential communication ability was assessed
using the Color Game subtest of the Pragmatic Language
Skills MEDEA Battery. Most of the CI users (83.9%) were
reported to have “normal” pragmatic skills (i.e., within 1 SD
of appropriate norms), but neither age at implantation nor
duration of CI use was significantly related to performance.
Taken together, these two studies indicate that earlier im-
plantation was associated with verbal abilities that were
directly linked to interpersonal communication, a BICS
skill, but not the CALP skills needed for metaphor
comprehension.
AuBuchon, Pisoni, and Kronenberger (2015) investi-
gated verbal rehearsal speed (used to maintain verbal
information in working memory) and perceptual encoding
speed (the rapid search of semantic or long-term memory
for production of a verbal label) as they relate to executive
function following long-term CI use. Their sample included
55 children, adolescents, and young adults (Mage = 15.3 years,
SD = 4.9) who were CI users. The participants had re-
ceived their implants prior to 7 years of age (M age =
36.4 months, SD = 20) and had used them for at least
7 years (M = 12.3 years, SD = 3.9). Verbal rehearsal
speed was measured by examining the rate of overt articu-
lation in a set of meaningful sentences. Perceptual encoding
speed was measured by examining the rates at which indi-
viduals could name series of visually presented digits and
colors. (A composite working memory score was created
by combining scores from three verbal memory tasks, but
possible associations with CI use were not reported.) Ear-
lier cochlear implantation was associated with better perfor-
mance in the perceptual encoding task but not the verbal
rehearsal task.
Marschark, Spencer, et al. (2015) considered CI use
and age at implantation as variables of interest in their ex-
amination of relations of language and cognition among
college-age, deaf CI users and nonusers. Their Experiment
1 examined performance on three visual–spatial tasks: the
Spatial Relations and Pair Cancellation tasks drawn from
the WJ-III Tests of Cognitive Abilities and an embedded
figures task. Participants included 51 CI users who reported
receiving implants at a mean age of 6.4 years (SD = 4.8).
Age at implantation was significantly correlated only with
scores on the Spatial Relations task. Similar results were
obtained in comparisons of 12 CI users who received their
CIs before the age of 3 years and the 39 CI users who
received their CIs after the age of 3 years as well as when
25 participants who received CIs prior to the age of 5 years
were compared to the 26 who received them later. Marschark,
Spencer, et al.’s Experiment 2 used the Corsi blocks task
to evaluate nonverbal working memory. Seventy-five deaf
participants included 33 CI users who received their implants
at a mean age of 7.2 years (SD = 5.7). Age at implantation
was not related to either of the standard Corsi blocks
measures: the highest span reached or the total number
of trials. Experiment 3 involved participants who received
their CIs at a mean age of 6.9 years (SD = 5.1). Nonverbal,
logical reasoning skills were examined using the General
Ability Measure for Adults, which entails problem solving
involving abstract designs and shapes. Subtests include
Matching, Analogy, Construction, and Sequences tasks.
Correlational analyses indicated that age at implantation
was related only to General Ability Measure for Adults
Sequences scores. No significant differences were obtained
when small samples of individuals who received their CIs
at the age of 3 years or earlier were compared to larger
samples who had received them later.
De Hoog et al. (2016) examined short-term and
working memory in 39 children they tested ranging in age
from 5;3 to 10;1; all had received their implants prior to
4 years of age (range: 7–38 months). Verbal short-term
memory tasks required repeating strings of words, repeat-
ing sentences, forward digit span, and a nonword repetition
test. A backward digit span task was used to assess working
memory. Age at implantation was not significantly correlated
with performance on any of the tasks.
Verbal short-term memory in children with CIs also
was investigated by Talli, Okalidou, and Tsalighopoulos
(2018) using a digit span task and a nonword repetition
task. They also administered a rapid picture-naming task
(i.e., long-term memory retrieval) and Raven’s Progressive
Matrices (nonverbal intelligence). The 15 children were
between the ages of 4;7 and 8;6 (M = 79.80 months, SD =
14.80) at testing. They had received their implants at a
mean age of 31.33 months (SD = 10.25) and had used
them for an average of 48.60 months (SD = 14.19). Neither
age at implantation nor duration of CI use was significantly
related to any of the cognitive scores.
Klein, Walker, and Tomblin (2018) examined se-
quential learning in children with CIs. Their study included
20 children between the ages of 7.3 and 11.8 years (M =
9.4 years, SD = 1.5) who had received their implants between
the ages of 1.1 and 3.9 years (M = 1.9 years, SD = 0.9)
and had used them from 3.7 to 10.2 years. Using a visual–
spatial, implicit sequential learning task, Klein et al. found
the CI users were slower in responding than a comparison
group of hearing peers, although the two groups did not differ
in their degree of sequential learning. Age at implantation
was not significantly related to the CI children’s rate of
learning (duration of implant use was not examined).
Torkildsen, Arciuli, Haukedal, and Wie (2018) also
examined sequence learning in a group of 34 CI users aged
between 7;0 and 12;11 who had received their implants by
4 years of age and had used them for at least 3 years. Their
Marschark et al.: Age at Implantation, Learning, and Cognition
task involved children viewing a stream of visual stimuli
in which they had to implicitly learn a sequence of embed-
ded triplets. Age at implantation was not related to sequence
learning. (Relations of age at implantation with results
from three memory tasks and Raven’s Progressive Matrices
were not reported.)
Although studies described earlier that examined a
possible link between age at implantation and academic
achievement in mathematics failed to find such an associa-
tion, two studies have examined the possible impact of age
at implantation on numerical and nonnumerical estimation
abilities that underlie mathematical performance. Borgna
et al. (2018; see also Convertino, Borgna, Marschark, &
Durkin, 2014) examined estimation abilities of college-age
CI users and deaf nonusers using both numerical and
nonnumerical (i.e., real-world) tasks. Sixty-three participants
included 33 CI users who reported receiving their implants
between the ages of 1 and 16 years, with a mean of 6.03 years
(SD = 4.15). Age at implantation was unrelated to perfor-
mance on either task and also was unrelated to CI users’
broader mathematical ability as indicated by their mathe-
matics scores on the American College Test.
Considering the studies described in this section, when
age at implantation is treated as a continuous variable,
evaluations of benefits associated with early implantation
to executive functioning, theory of mind, and other com-
plex cognitive tasks offer a picture somewhat different than
that found with regard to literacy and academic achieve-
ment: The 19 studies, all with age at implantation considered
a continuous variable, included 97 analyses evaluating its
possible effects on performance. Only nine (9%) of those
analyses yielded significant results that suggested benefits
of earlier implantation.
Cognitive Processing: Age at Implantation
as a Discrete Variable
Theory of Mind
Most and Aviner (2009) examined possible links
between spoken language ability and theory of mind in a
study involving 10- to 17-year-olds. Twenty CI users were
divided into early and late implantation groups, with the
division at 6 years of age. The 10 early implantees were
aged 10;10–15;7 (M = 13;9, SD = 1;7), had received their
implants between the ages of 2;6 and 5;6 (M = 3;11, SD =
1;3), and had used them for 6;9–13;1. The 10 late implan-
tees were aged 10;0–17;6 (M = 15;1, SD = 2;7), had received
their implants between the ages of 6;2 and 16;2 (M = 11;4,
SD = 3;8), and had used them for 1;0–6;9. The children
were presented with single neutral sentences (“I am going
out now, and I’ll be back later”) varying in auditory (into-
nation contours) cues, visual (nonverbal) cues, or both
auditory and visual cues associated with happiness, sadness,
anger, and fear. None of the differences between the early
and late CI groups was significant.
Sundqvist, Lyxell, Jönsson, and Heimann (2014)
investigated theory of mind performance among 16 CI users,
1327
aged 4–9 years. Mean age at implantation was 6.5 years
(SD = 1.64). A median split divided the children into
groups receiving implants at a mean age of 17.6 months
(SD = 5.2) and 40.6 months (SD = 13.3); the two did not
differ in their length of CI use or on measures of language
or nonverbal intelligence. The theory of mind tasks used
by Sundqvist et al. included both an unexpected location
false belief task (the frequently used Sally-Anne task; Baron-
Cohen, Leslie, & Frith, 1985) and a second task in which
children had to recognize emotions and feelings experienced
by individuals in a series of stories depicting everyday
situations. Although, overall, the deaf children performed
significantly below a hearing comparison group on both
tasks, the earlier-implanted deaf children outperformed the
late-implanted children on the emotion recognition task
but not on the unexpected location task.
Only two theory of mind studies have considered age
at implantation as a discrete variable. One of them failed
to obtain significant results in three comparisons; the other
obtained significant results in one of two comparisons. As
shown in Supplemental Materials S1–S4, with the exception
of the Sundqvist et al. (2014) emotion recognition task, the
effect sizes were small, ranging from .22 to .30 (vs. .47).
Other Cognitive Processes
Colletti, Mandalà, Zoccante, Shannon, and Colletti
(2011) conducted a 10-year longitudinal study of 73 children
who received their CIs between 2 and 35 months of age.
The sample was divided into three groups based on age at
implantation: 19 of the children having received their CIs
between 2 and 11 months of age (M = 6.4, SD = 2.8), 21
having received them between 12 and 23 months of age (M =
19.3, SD = 3.8), and 33 having received them between 25
and 35 months of age (M = 30.1, SD = 5.9). Nonverbal
cognitive ability was assessed using the Griffiths Mental
Development Scale (Locomotor, Eye–Hand Coordination,
and Performance subscales) and the Leiter International
Performance Scale–Revised (Figure-Ground and Form Com-
pletion subtests to assess spatial attention and Sequential
Order and Matching/Repeated Pattern subtests to assess fluid
reasoning). At a 5-year follow-up testing, Colletti et al.
reported that the earliest-implanted group outperformed
the other groups only on the performance subscales, using
an overall mean score. At the 10-year follow-up, the earliest-
implanted group was reported to outperform the middle-
implanted group on sequential order and repeated pattern
subscales and to outperform the latest-implanted group on
Form Completion, Sequential Order, and Matching/Repeated
Pattern subscales. The investigators did not report the
statistical tests used for these comparisons, but the lack
of significant differences between the two later-implanted
groups, the inconsistent differences between them and the
earliest group, and variability with regard to which tests
yielded significant differences argue against concluding that
there was any generalized advantage for earlier implantation
demonstrated in this study. Nevertheless, the results raise
1328 American Journal of Speech-Language Pathology • Vol. 28 • 1318–1334 • August 2019
interesting questions concerning outcomes following im-
plantation during the first few months of life.
In their examination of world knowledge among deaf
and hearing college students, Convertino et al. (2014, Ex-
periment 2) engaged deaf and hearing participants in tasks
assessing their knowledge of history, professions of famous
people, and geography as well as tasks in which partici-
pants had to provide real-world magnitude information
(e.g., the length of a football field) and estimate magnitudes
of displays at laboratory stations (e.g., the number of
marbles that would fit in a box). Their 185 deaf participants
included 93 who used CIs and reported a mean age at im-
plantation of 8.29 years (SD = 6.02); the mean duration of
CI use was 13.10 years (SD = 5.25). Within the CI group,
25 students reported receiving their CIs by 3.5 years of age
(M = 2.56 years, SD = 0.75); the remaining 65 reported
receiving them at a mean age of 10.45 years (SD = 5.76).
Although no significant differences between the earlier-
and later-implanted groups were obtained on any of the
tasks, their Table 3 (p. 480) indicates that there was a
clear trend favoring greater knowledge/accuracy on the
part of individuals who received their implants later rather
than earlier.
The López-Higes et al. (2015) Spanish reading study
described earlier also reported on children’s performance
on several cognitive measures drawn from the Wechsler
Intelligence Scale for Children–Fourth Edition. Thirty-eight
8- to 12-year-olds, half of whom had received their CIs
before 24 months of age, were tested on backward digit
span and three visual–spatial perceptual reasoning tasks
(matrix reasoning, block design, and picture concepts); the
latter tasks were reported as a total perceptual reasoning
index. Although the investigators did not explicitly report
differences between the two groups on the cognitive tasks,
analyses of the data provided in their Table 5 (p. 140)
yielded only small effect sizes for both backward digit span
(.25) and perceptual reasoning (.26).
Considering the 25 assessments of the impact of age
at implantation on cognitive processing in the five studies
described above, only seven (28%) yielded significant dif-
ferences between earlier- and later-implanted samples that
indicated benefits for children who had received their CIs
earlier. López-Higes et al. (2015) did not report results in
that regard, but using data they provided, small effect sizes
suggested at least trends in that direction.
Another Interim Summary
The 24 studies described in the previous two sections
obtained significant results in only 16 of 127 assessments
(13%) examining the extent to which age at implantation
affects cognitive processing. Taken together, these results
suggest that, consistent with evidence from studies of literacy
and academic achievement, age at implantation is not a
robust predictor of later cognitive processing in the cognitive
domains most frequently evaluated in studies of children
with CIs. In contrast to results obtained from studies of
literacy and academic achievement, however, when age at
implantation is considered a continuous as opposed to a
discrete variable, the pattern of findings is somewhat differ-
ent. When considered a continuous variable, only 9% of
assessments indicated benefits of earlier implantation. When
age at implantation was considered a discrete variable,
23% of assessments indicated significant benefits of earlier
implantation, although as shown in Supplemental Mate-
rials S1–S4, with the exception of the Sundqvist et al. (2014)
emotion recognition task, the effect sizes are small, ranging
from .22 to .30 (vs. .47). In short, there appears to be little
consistent evidence that age at implantation is a strong pre-
dictor of cognitive growth.
Discussion
This critical assessment considered the extent to which
age at implantation is associated with various academic,
learning, and cognitive outcomes among CI users. Clearly,
there are numerous factors that contribute to such outcomes.
Nevertheless, we initially had questioned and sought to
evaluate two hypotheses based on widespread assumptions
common in the field: (a) Age at implantation is a major
factor in determining eventual outcomes among CI users,
and (b) “the earlier the better,” even if what is meant by
“early” varies widely (see Duchesne et al., 2009). In explor-
ing the evidence for these two assumptions, we considered
multiple experiments and conditions across 44 studies that
were conducted from 2003 to the present that considered
age at implantation as a relevant variable and included sta-
tistical consideration of that variable. Several of the studies
offered caveats from the authors with regard to positive
or negative findings, and several others conducted multiple
analyses of the same data considered in different ways
(sometimes providing positive and negative findings in the
same study). In all cases, we accepted the results presented
at face value.
Overall, the evidence presented here suggests that
the impact of age at implantation on various aspects of
learning and cognition is inconsistent at best. At a global
level, two emergent patterns of empirical results are worthy
of note. First, across the various tasks and groups in the
44 studies (not considering multiple analyses of the same
data), there was a total 167 assessments concerning the
effects of age at implantation. Of those 167 assessments,
only 35 (21%) yielded significant effects indicating benefits
for earlier implantation. There was no obvious pattern with
regard to age at testing, cutoff points when age at implan-
tation was treated as a dichotomous variable, or sample
sizes that yielded significant effects in that regard (see Sup-
plemental Materials S1–S4). Duration of implant use was
reported in several studies, but not enough data were avail-
able to be able to draw any firm conclusions beyond the
possibility that confounds with duration of CI use may be
problematic when considering the impact of age at implan-
tation on any aspect of development. Perhaps reflecting
that possibility, several studies found better performance
among later-implanted rather than earlier-implanted children.
Marschark et al.: Age at Implantation, Learning, and Cognition
A second emergent pattern of results is related to the
way in which studies dealt with age at implantation as a
variable. Examination of studies that investigated literacy
and academic achievement suggested that treating age at
implantation as a continuous or discrete variable had little
impact on their findings, as somewhat less than 50% of
relevant assessments yielded significant effects in either
case. Studies that had investigated various cognitive abili-
ties, in contrast, obtained over twice as many significant
findings when age at implantation was considered a dis-
crete variable rather than a continuous variable. The pat-
tern of findings found in the existing literature thus casts
doubt on assumptions that age at implantation, broadly
speaking, is a reliable determinant of either medium- or
long-term outcomes with regard to learning and cognition,
but methodological variability can cloud the issue.
The purpose of this critical assessment was not to
either accept or reject the null hypothesis that age at implan-
tation affects learning and cognition, and furthermore,
more carefully designed research will be needed before
strong conclusions can be drawn. We noted earlier, however,
the “widespread assumption” that earlier implantation has
broad, beneficial effects on the development of deaf children
for a variety of neurophysiological, linguistic, and social
reasons. Despite the lack of empirical support for that hy-
pothesis in the literature, it still makes intuitive sense.
There are variety of reasons why individual studies might
fail to demonstrate effects of earlier as opposed to later
implantation including inadequate sample sizes, insensitive
dependent measures, incomplete data collection, and inade-
quate control over potentially confounding variables. As
noted in several of the studies considered here, the latter in-
cludes child-related factors (e.g., preimplantation language
ability, nature and degree of hearing loss, unilateral vs.
bilateral implants, amount of benefit from the CI), family
factors (e.g., socioeconomic status, parental education and
sensitivity), and environmental factors (e.g., nature of early
intervention, educational settings, peer communication
partners). A close reading of the relevant literature, includ-
ing the studies discussed here, indicates that, for a variety
of practical reasons, most pediatric CI studies suffer from
one or more of these shortcomings.
At this juncture in the history of research concerning
cochlear implantation, such limitations might be expected.
A variety of key studies including programatic evaluation of
the impact of children’s use of hearing aids prior to im-
plantation or in addition to CIs, their use of sign language
prior to or in addition to spoken language, and the interac-
tive effects of language and cognition among CI users
remain to be conducted. Such issues need to be acknowledged
fully and addressed, with all appropriate caveats to find-
ings, rather than making general assumptions (or claims)
based on affirmative findings that occur only 50% of the
time—and sometimes even less than 50% of the time.
Importantly, the present findings do not minimize
the potential benefits of CI use for deaf learners. Rather,
they emphasize that the early benefits accruing to young
deaf children from early access to language via CIs—like
1329
those accruing to deaf children of deaf parents from early
access to sign language (Marschark & Knoors, 2019)—
vary widely and often are insufficient to fully support the
higher order academic and cognitive abilities required for
robust learning and cognition beyond the early school years
(e.g., Convertino et al., 2009; Crowe et al., 2017; Dammeyer
et al., 2017; Marschark, Shaver, et al., 2015). Indeed, it is
the fact that young CI users, like young native signers,
demonstrate enhanced linguistic and academic function-
ing that typically results in the reduction or removal of their
support services. Marschark and Knoors (2019) argued that
this is a serious mistake. They suggested that BICS that
serve young deaf and hearing children well at home and
at school lack much of the vocabulary, grammar, and prag-
matic sophistication needed for CALP in the real world.
That is, BICS may be sufficient for normal cognitive and
academic growth early on, but the absence of CALP disad-
vantages deaf children (and their parents and teachers), re-
gardless of the children’s primary mode of communication.
Marschark and Knoors therefore suggested that continua-
tion of support services for deaf native signers and CI users
beyond the primary school years might allow them to main-
tain early gains and build on them in order to meet the more
complex demands of secondary and postsecondary education.
The present discussion, like any consideration of
literature associated with cochlear implantation or other
interventions, is limited by the fact that the meaning of
“early,” as in “early implantation,” varies widely. Evaluation
of the extent to which age at implantation influences aca-
demic abilities is limited further by the paucity of studies
examining educational outcomes beyond reading. Evaluating
the impact of age at implantation on cognitive processing
also is hindered by the fact that assessment tools differ in
studies across and even within laboratories. One difficulty
in establishing common methodologies is the need for dif-
ferent kinds of cognitive processing tasks for children of
different ages. However, preferences for different tasks
among investigators with different academic backgrounds
constitute an additional challenge for studies involving
similar age groups. Statistical methods for determining the
impact of age at implantation also vary across and within
laboratories, with various (sometimes unique) composite
and other calculated performance scores that prevent
straightforward comparisons across studies and lack of
reported means, standard deviations, and correlation coef-
ficients that prevent calculation of effect sizes (see Supplemental
Materials S1–S4).
Even more concerning than these four methodological
issues is simply the relative lack of research on age at implan-
tation on medium- and long-term outcomes in almost any
area. The assumption that “earlier is better” may be one
causal factor (Why study it when the answer is obvious?),
but pediatric cochlear implantations have been conducted
for over 20 years in several countries, and the continuing,
relative lack of long-term research on learning and cognition
is surprising. We recognize that long-term follow-up studies
of children who have received CIs are both difficult and
expensive, particularly if the goal is to continue to follow all
1330 American Journal of Speech-Language Pathology • Vol. 28 • 1318–1334 • August 2019
implant recipients in the initial sample, even those who
have discontinued their CI use (and clinic visits). Without
consideration of complete, consecutive samples, only “suc-
cessful” CI users are evaluated, and it is difficult to draw
valid conclusions about the benefits of related variables
from any particular study (Beadle et al., 2005; Moberly,
Castellanos, & Mattingly, 2018).
At a practical level, there are at least three other
contributing factors to the lack of research with regard to
links between age at implantation and learning/cognition.
One is that many of the CI centers providing cochlear im-
plantation simply are not research focused. Closely related
is the fact that those centers that do conduct related re-
search frequently are funded only for immediate postopera-
tive or short-term outcome studies rather than having the
multiple-year funding that would be required for longer-
term evaluations. Furthermore, many, if not most, implan-
tation centers are unlikely to have interest in long-term
learning and cognitive outcomes. As Archbold (2006) noted,
“What the surgeons start, teachers have to finish.” For
that to happen—for teachers and parents to finish what the
surgeon began—it is essential that we understand the long-
term influences of early cochlear implantation on various
domains of cognition and learning beyond their earliest
trajectories. As the meaning of “early” in “early implanta-
tion” widens, so will the meaning of “long-term” with regard
to outcomes, and not just for individuals who received their
implants “relatively late by current standards.”
Conclusions
If future research is going to continue to focus on
age at implantation as a potentially important demo-
graphic factor in predictions (and claims) about outcomes,
this critical assessment suggests that a number of other
factors need to be considered as well. Those factors will not
all be medical ones or even audiologic/linguistic ones (e.g.,
duration of implant use, preimplant language abilities re-
gardless of modality). They also will need to include family
and other environmental factors well documented to sup-
port the language, cognitive, and academic achievement of
hearing and deaf children (e.g., socioeconomic status,
maternal education). Indeed, Holt and colleagues (Holt
et al., 2013; Holt, Beer, Kronenberger, Pisoni, & Lalonde,
2012) have demonstrated that the family environment
following pediatric cochlear implantation can have signifi-
cant effects on children’s cognitive processing and language
functioning. In essence, this reiterates the “two-way street”
alluded to earlier with regard to language and learning
having ever-widening impact over the course of development
in terms of both the auditory system and language and
cognitive growth.
At a theoretical level, even this admittedly narrow
consideration of the literature raises the question of how it
is that we have arrived at the assumptions that age at im-
plantation is a major factor in determining broad outcomes
for deaf children and that the earlier the implantation the
better, rather than considering these as hypotheses to be
rigorously tested. There are some domains in which these
assumptions do appear to be true, most notably, speech in-
telligibility and other early expressive language measures
(e.g., Castellanos, Pisoni, Kronenberger, & Beer, 2016;
Montag, AuBuchon, Pisoni, & Kronenberger, 2014). How-
ever, although early implantation is associated with use of
spoken language and use of spoken language is associated
with more positive academic outcomes, the studies reviewed
above indicate that transitivity does not necessarily hold.
Moreover, even within the linguistic domain, there are other
aspects of language development in which assumptions
about benefits attributable to early implantation appear
misplaced (Duchesne & Marschark, in press; Houston et al.,
in press; Niparko et al., 2010). An obvious example is the
fact that, although Niparko et al. (2010) frequently are
cited as demonstrating the importance of early implantation
on outcomes following pediatric cochlear implantation,
their Figure 1 (p. 1502) clearly shows a huge individual
variability in expressive and receptive language outcomes
among children receiving CIs roughly from 1 to 5 years of
age. In each of their three age groups, it is clear that some
children benefited little from implantation, if at all. The au-
thors noted, a priori, that in behavioral studies concerning
cochlear implantation, “variance in observed results is no-
toriously high” (Niparko et al., 2010, p. 1498), but the vari-
ability in their own results is not addressed anywhere in their
report.
There is no doubt that age at implantation carries
some enticing face validity and is far easier (and eco-
nomical) to measure than other variables such as the use of
different speech processors for varying lengths of time or
the preimplant language produced by and addressed to
deaf infants and toddlers, let alone the myriad other child-
related and environmental factors alluded to above. At the
same time, however, unqualified earlier-is-better assump-
tions based on beliefs about a rapidly fading brain plastic-
ity largely ignore the possibility of ongoing equilibration
between language development and cognitive development.
That is, earlier and later implantation may be asymptotic
with regard to various outcomes (for individuals, if not
more generally), and/or the relative importance of laying
down linguistic foundations for cognitive development or
laying down cognitive foundations for language develop-
ment may vary depending on individual children’s intrinsic
and extrinsic (environmental) characteristics. In any case,
just as it is a mistake to assume that young deaf children
who derive cognitive or academic advantages from early
access to language will maintain those advantages ad infi-
nitum, so it is a mistake to assume that any easy-to-quantify
demographic(s) will be the “Holy Implantation Grail” for
which so many are searching.
References
Archbold, S. (2006, March). Cochlear implants: Changing deaf
education? Paper presented at the Scottish Sensory Centre
Conference on Language and Deaf Education: Into the 21st
Century, Edinburgh, United Kingdom.
Marschark et al.: Age at Implantation, Learning, and Cognition
Archbold, S., Harris, M., O’Donoghue, G., Nikolopoulos, T., White,
A., & Richmond, H. L. (2008). Reading abilities after cochlear
implantation: The effect of age at implantation on outcomes
at 5 and 7 years after implantation. International Journal of
Pediatric Otorhinolaryngology, 72(10), 1471–1478. https://doi.org/
10.1016/j.ijporl.2008.06.016
Archbold, S., Nikolopoulos, T., & O’Donoghue, G. (2006). Reading
abilities after cochlear implantation: The effect of age of implan-
tation. Keynote presentation at the Conference on Cochlear Im-
plants: Educational Implications of the University of Nottingham,
Nottingham, United Kingdom.
AuBuchon, A. M., Pisoni, D. B., & Kronenberger, W. G. (2015).
Short-term and working memory impairments in early-implanted,
long-term cochlear implant users are independent of audibility
and speech production. Ear and Hearing, 36(6), 733–737.
https://doi.org/10.1097/AUD.0000000000000189
Barajas, C., González-Cuenca, A. M., & Carrero, F. (2016). Com-
prehension of texts by deaf elementary school students: The
role of grammatical understanding. Research in Developmental
Disabilities, 59, 8–23. https://doi.org/10.1016/j.ridd.2016.
07.005
Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985). Does the autistic
child have a “theory of mind”? Cognition, 21(1), 37–46. https://doi.
org/10.1016/0010-0277(85)90022-8
Bat-Chava, Y., & Deignan, E. (2001). Peer relationships in chil-
dren with cochlear implants. Journal of Deaf Studies and Deaf
Education, 6, 186–199. https://doi.org/10.1093/deafed/6.3.186
Beadle, E. A., McKinley, D. J., Nikolopoulos, T. P., Brough, J.,
O’Donoghue, G. M., & Archbold, S. M. (2005). Long-term
functional outcomes and academic-occupational status in im-
planted children after 10 to 14 years of cochlear implant use.
Otology & Neurotology, 26(6), 1152–1160. https://doi.org/
10.1097/01.mao.0000180483.16619.8f
Beer, J., Kronenberger, W. G., Castellanos, I., Colson, B. G.,
Henning, S. C., & Pisoni, D. B. (2014). Executive functioning
skills in preschool-age children with cochlear implants. Journal
of Speech, Language, and Hearing Research, 57(4), 1521–1534.
https://doi.org/10.1044/2014_JSLHR-H-13-0054
Borgna, G., Convertino, C., Walton, D., Marschark, M., Crowe, K.,
& Trussell, J. (2018). Numerical and real-world estimation
abilities of deaf and hearing college students. Deafness and
Education International, 20, 59–79. https://doi.org/10.1080/
14643154.2018.1437238
Castellanos, I., Kronenberger, W. G., & Pisoni, D. B. (2018).
Psychosocial outcomes in long-term cochlear implant users.
Ear and Hearing, 39(3), 527–539. https://doi.org/10.1097/AUD.
0000000000000504
Castellanos, I., Pisoni, D. B., Kronenberger, W. G., & Beer, J.
(2016). Early expressive language skills predict long-term
neurocognitive outcomes in cochlear implant users: Evidence
from the MacArthur–Bates Communicative Development
Inventories. American Journal of Speech-Language Pathology,
25(3), 381–392. https://doi.org/10.1044/2016_AJSLP-15-0023
Ching, T. Y., Day, J., & Cupples, L. (2014). Phonological aware-
ness and early reading skills in children with cochlear implants.
Cochlear Implants International, 15(Suppl. 1), S27–S29. https://doi.
org/10.1179/1467010014Z.000000000172
Christiansen, J., & Leigh, I. (2002). Cochlear implants in children:
Ethics and choices. Washington, DC: Gallaudet University
Press.
Colin, S., Ecalle, J., Truy, E., Lina-Granade, G., & Magnan, A.
(2017). Effect of age at cochlear implantation and at exposure
to cued speech on literacy skills in deaf children. Research in
Developmental Disabilities, 71, 61–69. https://doi.org/10.1016/
j.ridd.2017.09.014
1331
Colletti, L., Mandalà, M., Zoccante, L., Shannon, R. V., & Colletti,
V. (2011). Infants versus older children fitted with cochlear
implants: Performance over 10 years. International Journal of
Pediatric Otorhinolaryngology, 75(4), 504–509. https://doi.org/
10.1016/j.ijporl.2011.01.005
Connor, C., & Zwolan, T. A. (2004). Examining multiple sources
of influence on the reading comprehension skills of children
who use cochlear implants. Journal of Speech, Language, and
Hearing Research, 47, 509–526. https://doi.org/10.1044/1092-
4388(2004/040)
Convertino, C. M., Borgna, G., Marschark, M., & Durkin, A. (2014).
Word and world knowledge among deaf students with and
without cochlear implants. Journal of Deaf Studies and Deaf
Education, 19, 471–483. https://doi.org/10.1093/deafed/enu024
Convertino, C. M., Marschark, M., Sapere, P., Sarchet, T., &
Zupan, M. (2009). Predicting academic success among deaf
college students. Journal of Deaf Studies and Deaf Education,
14, 324–343. https://doi.org/10.1093/deafed/enp005
Conway, C. M., Pisoni, D. B., Anaya, E. M., Karpicke, J., &
Henning, S. C. (2011). Implicit sequence learning in deaf chil-
dren with cochlear implants. Developmental Science, 14, 69–82.
https://doi.org/10.1111/j.1467-7687.2010.00960.x
Courtin, C. (2000). The impact of sign language on the cognitive
development of deaf children: The case of theories of mind.
Journal of Deaf Studies and Deaf Education, 5, 266–276. https://
doi.org/10.1093/deafed/5.3.266
Crowe, K., Marschark, M., Dammeyer, J., & Lehane, C. (2017).
Achievement, language, and technology use among college-
bound deaf learners. Journal of Deaf Studies and Deaf Education,
22, 393–401. https://doi.org/10.1093/deafed/enx029
Cummins, J. (1981). Age on arrival and immigrant second lan-
guage learning in Canada: A reassessment. Applied Linguistics,
2, l32–l49. https://doi.org/10.1093/applin/II.2.132
Dammeyer, J., Lehane, C., & Marschark, M. (2017). Use of tech-
nological aids and interpretation services among children and
adults with hearing loss. International Journal of Audiology,
56, 740–748. https://doi.org/10.1080/14992027.2017.1325970
De Hoog, B. E., Langereis, M. C., van Weerdenburg, M., Keuning, J.,
Knoors, H., & Verhoeven, L. (2016). Auditory and verbal
memory predictors of spoken language skills in children with
cochlear implants. Research and Developmental Disabilities,
57, 112–124. https://doi.org/10.1016/j.ridd.2016.06.019
DesJardin, J. L., Ambrose, S. E., & Eisenberg, L. S. (2009). Liter-
acy skills in children with cochlear implants: The importance
of early oral language and joint storybook reading. Journal of
Deaf Studies and Deaf Education, 14, 22–43. https://doi.org/
10.1093/deafed/enn011
Domínguez, A.-B., Carrillo, M.-S., González, V., & Alegria, J.
(2016). How do deaf children with and without cochlear implants
manage to read sentences: The key word strategy. Journal of
Deaf Studies and Deaf Education, 21(2016), 280–292. https://
doi.org/10.1093/deafed/enw026
Duchesne, L. & Marschark, M. (in press). Effects of age at co-
chlear implantation on vocabulary and grammar: A review of the
evidence. American Journal of Speech-Language Pathology.
Duchesne, L., Sutton, A., & Bergeron, F. (2009). Language achieve-
ment in children who received cochlear implants between 1
and 2 years of age: Group trends and individual patterns. Journal
of Deaf Studies and Deaf Education, 14, 465–485. https://doi.
org/10.1093/deafed/enp010
Fagan, M. K., Pisoni, D. B., Horn, D. L., & Dillon, C. M. (2007).
Neuropsychological correlates of vocabulary, reading, and
working memory in deaf children with cochlear implants.
Journal of Deaf Studies and Deaf Education, 12, 461–471.
https://doi.org/10.1093/deafed/enm023
1332 American Journal of Speech-Language Pathology • Vol. 28 • 1318–1334 • August 2019
Gallego, C., Martín-Aragoneses, M. T., López-Higes, R., & Pisón, G.
(2016). Semantic and syntactic reading comprehension strate-
gies used by deaf children with early and late cochlear implanta-
tion. Developmental Disabilities, 49, 153–170. https://doi.org/
10.1016/j.ridd.2015.11.020
Geers, A. E. (2003). Predictors of reading skill development in
children with early cochlear implantation. Ear and Hearing, 24,
59S–68S. https://doi.org/10.1097/01.AUD.0000051690.43989.5D
Geers, A. E. (2004). Speech, language, and reading skills after
early cochlear implantation. Archives of Otolaryngology—Head
& Neck Surgery, 130(5), 634–638. https://doi.org/10.1001/
archotol.130.5.634
Geers, A. E., Brenner, C., & Davidson, L. (2003). Factors associ-
ated with development of speech perception skills in children
implanted by age five. Ear and Hearing, 24(1), 24S–35S.
https://doi.org/10.1097/01.AUD.0000051687.99218.0F
Geers, A. E., & Hayes, H. (2011). Reading, writing, and phono-
logical processing skills of adolescents with 10 or more years of
cochlear implant experience. Ear and Hearing, 32, 49S–59S.
https://doi.org/10.1097/AUD.0b013e3181fa41fa
Geers, A. E., Tobey, E., Moog, J., & Brenner, C. (2008). Long-
term outcomes of cochlear implantation in the preschool years:
From elementary grades to high school. International Journal
of Audiology, 47(Suppl. 2), S21–S30. https://doi.org/10.1080/
14992020802339167
Harris, M., & Terlektsi, E. (2010). Reading and spelling abilities
of deaf adolescents with cochlear implants and hearing aids.
Journal of Deaf Studies and Deaf Education, 16, 24–34. https://
doi.org/10.1093/deafed/enq031
Holt, R. F., Beer, J., Kronenberger, W. G., & Pisoni, D. B. (2013).
Developmental effects of family environment on outcomes in
pediatric cochlear implant recipients. Otology & Neurotology,
34(3), 388–395. https://doi.org/10.1097/MAO.0b013e318277a0af
Holt, R. F., Beer, J., Kronenberger, W. G., Pisoni, D. B., & Lalonde,
K. (2012). Contribution of family environment to pediatric co-
chlear implant users’ speech and language outcomes: Some
preliminary findings. Journal of Speech, Language, and Hearing
Research, 55(3), 848–864. https://doi.org/10.1044/1092-4388
(2011/11-0143)
Houston, D., Chen, E., Monroy, C., & Castellanos, I. (in press).
How early auditory experience affects children’s ability to
learn spoken words. In M. Marschark & H. Knoors (Eds.),
The Oxford handbook of deaf studies in learning and cognition.
New York, NY: Oxford University Press.
James, D., Rajput, K., Brinton, J., & Goswami, U. (2007). Phono-
logical awareness, vocabulary, and word reading in children
who use cochlear implants: Does age of implantation explain
individual variability in performance outcomes and growth?
Journal of Deaf Studies and Deaf Education, 13(1), 117–137.
https://doi.org/10.1093/deafed/enm042
Johnson, C., & Goswami, U. (2010). Phonological awareness, vocab-
ulary, and reading in deaf children with cochlear implants.
Journal of Speech, Language, and Hearing Research, 53, 237–261.
https://doi.org/10.1044/1092-4388(2009/08-0139)
Ketelaar, L., Rieffe, C., Wiefferink, C., & Frijns, J. (2012). Does
hearing lead to understanding? Theory of mind in toddlers and
preschoolers with cochlear implants. Journal of Pediatric Psy-
chology, 37, 1041–1050. https://doi.org/10.1093/jpepsy/jss086
Klein, K. E., Walker, E. A., & Tomblin, J. B. (2018). Nonverbal
visual sequential learning in children with cochlear implants:
Preliminary findings. Ear and Hearing. https://doi.org/10.1097/
AUD.0000000000000564
Knoors, H., & Marschark, M. (2014). Teaching deaf learners: Psycho-
logical and developmental foundations. New York, NY: Oxford
University Press.
Kral, A., Kronenberger, W. G., Pisoni, D. B., & O’Donoghue, G. M.
(2016). Neurocognitive factors in sensory restoration of early
deafness: A connectome model. The Lancet Neurology, 15(6),
610–621. https://doi.org/10.1016/S1474-4422(16)00034-X
Kral, A., & Sharma, A. (2012). Developmental neuroplasticity after
cochlear implantation. Trends in Neurosciences, 35(2), 111–122.
https://doi.org/10.1016/j.tins.2011.09.004
Kronenberger, W. G., Beer, J., Castellanos, I., Pisoni, D. B., &
Miyamoto, R. T. (2014). Neurocognitive risk in children with
cochlear implants. JAMA Otolaryngology—Head & Neck
Surgery, 140, 608–615. https://doi.org/10.1001/jamaoto.2014.
757
Kronenberger, W. G., Colson, B. G., Henning, S. C., & Pisoni, D. B.
(2014). Executive functioning and speech-language skills fol-
lowing long-term use of cochlear implants. Journal of Deaf
Studies and Deaf Education, 19(4), 456–470. https://doi.org/
10.1093/deafed/enu011
Kronenberger, W. G., Pisoni, D. B., Harris, M. S., Hoen, H. M.,
Xu, H., & Miyamoto, R. T. (2013). Profiles of verbal working
memory growth predict speech and language development in
children with cochlear implants. Journal of Speech, Language,
and Hearing Research, 56, 805–825. https://doi.org/10.1044/
1092-4388(2012/11-0356)
Kronenberger, W. G., Pisoni, D. B., Henning, S. C., & Colson, B. G.
(2013). Executive functioning skills in long-term users of cochlear
implants: A case control study. Journal of Pediatric Psychology,
38, 902–914. https://doi.org/10.1093/jpepsy/jst034
López-Higes, R., Gallego, C., Martín-Aragoneses, M. T., & Melle, N.
(2015). Morpho-syntactic reading comprehension in children
with early and late cochlear implants. Journal of Deaf Studies
and Deaf Education, 20, 136–146. https://doi.org/10.1093/
deafed/env004
Mancini, P., Dincer D’Alessandro, H., Guerzoni, L., Cuda, D.,
Ruoppolo, G., Musacchio, A., . . . Nicastri, M. (2015). Ade-
quate formal language performance in unilateral cochlear
implanted children: Is it indicative of complete recovery in all
linguistic domains? Insights from referential communication.
International Journal of Pediatric Otorhinolaryngology, 79,
598–604. https://doi.org/10.1016/j.ijporl.2015.02.003
Marschark, M., Edwards, L., Peterson, C., Crowe, K., & Walton,
D. (2019). Understanding theory of mind in deaf and hearing
college students. Journal of Deaf Studies and Deaf Education,
24, 104–118. https://doi.org/10.1093/deafed/eny039
Marschark, M., & Knoors, H. (2019). Sleuthing the 93% solution
in deaf education. In H. Knoors & M. Marschark (Eds.),
Evidence-based practice in deaf education (pp. 1–32). New York,
NY: Oxford University Press.
Marschark, M., Shaver, D. M., Nagle, K., & Newman, L. (2015).
Predicting the academic achievement of deaf and hard-of-hearing
students from individual, household, communication, and edu-
cational factors. Exceptional Children, 8, 350–369. https://doi.
org/10.1177/0014402914563700
Marschark, M., Spencer, L., Durkin, A., Borgna, G., Convertino, C.,
Machmer, E., Kronenberger, W. G., & Trani, A. (2015). Under-
standing language, hearing status, and visual–spatial skills.
Journal of Deaf Studies and Deaf Education, 20, 310–330.
https://doi.org/10.1093/deafed/env025
Mikic, B., Miric, D., Nicolic-Mikic, M., Ostojic, S., & Asanovic, M.
(2014). Age at implantation and auditory memory in cochlear
implanted children, Cochlear Implants International, 15(Suppl. 1),
S33–S35. https://doi.org/10.1179/1467010014Z.000000000191
Moberly, A. C., Castellanos, I., & Mattingly, J. K. (2018). Neuro-
cognitive factors contributing to cochlear implant candidacy.
Otology & Neurotology, 39(10), e1010–e1018. https://doi.org/
10.1097/MAO.0000000000002052
Marschark et al.: Age at Implantation, Learning, and Cognition
Montag, J. L., AuBuchon, A. M., Pisoni, D. B., & Kronenberger,
W. G. (2014). Speech intelligibility in deaf children after
long-term cochlear implant use. Journal of Speech, Language,
and Hearing Research, 57(6), 2332–2343. https://doi.org/
10.1044/2014_JSLHR-H-14-0190
Most, T., & Aviner, C. (2009). Auditory, visual, and auditory–visual
perception of emotions by individuals with cochlear implants,
hearing aids, and normal hearing. Journal of Deaf Studies and
Deaf Education, 14, 449–464. https://doi.org/10.1093/deafed/enp007
Most, T., Ingber, S., & Heled-Ariam, E. (2012). Social competence,
sense of loneliness, and speech intelligibility of young children
with hearing loss in individual inclusion and group inclusion.
Journal of Deaf Studies and Deaf Education, 17, 259–272.
https://doi.org/10.1093/deafed/enr049
Nicastri, M., Filipo, R., Ruoppolo, G., Viccaro, M., Dincer, H.,
Guerzoni, L., . . . Mancini, P. (2014). Inferences and metaphoric
comprehension in unilaterally implanted children with adequate
formal oral language performance. International Journal of
Pediatric Otorhinolaryngology, 78(5), 821–827. https://doi.
org/10.1016/j.ijporl.2014.02.022
Niparko, J. K., Tobey, E. A., Thal, D. J., Eisenberg, L. S., Wang,
N.-Y., Quittner, A. L., & Fink, N. E. (2010). Spoken language
development in children following cochlear implantation. Journal
of the American Medical Association, 303(15), 1498–1506. https://
doi.org/10.1001/jama.2010.451
Peterson, C. C., & Siegal, M. (1995). Deafness, conversation and
theory of mind. The Journal of Child Psychology and Psychiatry,
36, 459–474. https://doi.org/10.1111/j.1469-7610.1995.tb01303
Pisoni, D. B. (2016, May). Some neuromyths about cochlear implants:
Searching for the Holy Grail (invited presentation). Cleveland
State University, OH.
Pisoni, D. B., Conway, C. M., Kronenberger, W., Henning, S., &
Anaya, E. (2010). Executive function, cognitive control and
sequence learning in deaf children with cochlear implants. In
M. Marschark & P. E. Spencer (Eds.), The Oxford handbook
of deaf studies, language, and education (Vol. 2, pp. 439–457).
New York, NY: Oxford University Press.
Remmel, E., & Peters, K. (2009). Theory of mind and language in
children with cochlear implants. Journal of Deaf Studies and Deaf
Education, 14, 218–236. https://doi.org/10.1093/deafed/enn036
Ruffin, C. V., Kronenberger, W. G., Colson, B. G., Henning, S. C.,
& Pisoni, D. B. (2013). Long-term speech and language out-
comes in prelingually deaf children, adolescents and young adults
who received cochlear implants in childhood. Audiology and
Neurotology, 18(5), 289–296. https://doi.org/10.1159/000353405
Sarant, J. Z., Harris, D. C., & Bennet, L. A. (2015). Academic
outcomes for school-aged children with severe–profound hearing
loss and early unilateral and bilateral cochlear implants. Journal
of Speech, Language, and Hearing Research, 58, 1017–1032.
https://doi.org/10.1044/2015_JSLHR-H-14-0075
Stacey, P. C., Fortnum, H. M., Barton, G. R., & Summerfield, A. Q.
(2006). Hearing-impaired children in the United Kingdom. I:
Auditory performance, communication skills, educational achieve-
ments, quality of life, and cochlear implantation. Ear and Hearing,
27, 161–186. https://doi.org/10.1097/01.aud.0000202353.37567.b4
Sundqvist, A., Lyxell, B., Jönsson, R., & Heimann, M. (2014).
Understanding minds: Early cochlear implantation and the de-
velopment of theory of mind in children with profound hearing
impairment. International Journal of Pediatric Otorhinolaryn-
gology, 78, 538–544. https://doi.org/10.1016/j.ijporl.2013.12.039
Talli, I., Okalidou, A., & Tsalighopoulos, M. (2018). The relation
between short-term memory and vocabulary skills in Greek
children with cochlear implants: The role of hearing experi-
ence. First Language, 38, 359–381. https://doi.org/10.1177/
0142723717749073
1333
Torkildsen, J. V. K., Arciuli, J., Haukedal, C. L., & Wie, O. B.
(2018). Does a lack of auditory experience affect sequential
learning? Cognition, 170, 123–129. https://doi.org/10.1016/
j.cognition.2017.09.017
Uziel, A. S., Sillon, M., Vieu, A., Artieres, F., Piron, J. P., Daures,
J. P., & Mondain, M. (2007). Ten-year follow-up of a consecu-
tive series of children with multichannel cochlear implants.
Otology & Neurotology, 28(5), 615–628. https://doi.org/10.1097/
01.mao.0000281802.59444.02
Venail, F., Vieu, A., Artieres, F., Mondain, M., & Uziel, A. (2010).
Educational and employment achievements in prelingually
1334 American Journal of Speech-Language Pathology • Vol. 28 • 1318–1334 • August 2019
deaf children who receive cochlear implants. Archives of Oto-
laryngology—Head & Neck Surgery, 136, 366–372. https://doi.
org/10.1001/archoto.2010.91
Wellman, H. M., & Liu, D. (2004). Scaling of theory of mind
tasks. Child Development, 75, 523–541. https://doi.org/10.1111/
j.1467-8624.2004.00691.x
Willstedt-Svensson, U., Löfqvist, A., Almqvist, B., & Sahlén, B. (2004).
Is age at implant the only factor that counts? The influence of
working memory on lexical and grammatical development in
children with cochlear implants. International Journal of Audi-
ology, 43, 506–515. https://doi.org/10.1080/14992020400050065
Copyright of American Journal of Speech-Language Pathology is the property of American
Speech-Language-Hearing Association and its content may not be copied or emailed to
multiple sites or posted to a listserv without the copyright holder's express written permission.
However, users may print, download, or email articles for individual use.