Journal of Psychotherapy Integration

2016, Vol. 26, No. 2, 116 128

2016 American Psychological Association

1053-0479/16/$12.00 http://dx.doi.org/10.1037/int0000032

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

An Integrative Theory of Psychotherapy: Research and Practice

Seymour Epstein

Martha L. Epstein

University of Massachusetts Amherst

Park City, Utah

A dual-process personality theory and supporting research are presented. The dual
processes comprise an experiential system and a rational system. The experiential
system is an adaptive, associative learning system that humans share with other
higher-order animals. The rational system is a uniquely human, primarily verbal, reasoning
system. It is assumed that when humans developed language they did not abandon their
previous ways of adapting, they simply added language to their experiential system. The 2
systems are assumed to operate in parallel and are bidirectionally interactive. The validity
of these assumptions is supported by extensive research. Of particular relevance for
psychotherapy, the experiential system, which is compatible with evolutionary theory,
replaces the Freudian maladaptive unconscious system that is indefensible from an evolutionary perspective, as subhuman animals would then have only a single system that is
maladaptive. The aim of psychotherapy is to produce constructive changes in the experiential system. Changes in the rational system are useful only to the extent that they
contribute to constructive changes in the experiential system.
Keywords: experiential system, rational system, dual process, adaptive system,
maladaptive system

A basic assumption in cognitive-experiential

theory (CET) is that people operate with two
systems, an experiential system and a rational
system (Epstein, 2014). The experiential system
is primarily nonverbal and imagistic and learns
automatically from experience, normally doing
so outside of awareness. The rational system is
a uniquely human verbal reasoning system.
Both systems are adaptive, though in different
ways, and each has limitations. This article describes the conceptual foundation of CET, presents supporting research, and addresses its clinical implications.

A Cognitive-Experiential Theory of
Human Functioning
To understand the experiential system, it is useful to divide it into process and content. Process
consists of operating principles, while content
consists of what has been automatically learned.
Particularly important content includes implicit
needs and beliefs automatically learned from experience. Such implicit needs and beliefs influence
almost all aspects of functioning, including the
interpretation of events, automatic thoughts, feelings, conscious thinking, and behavioral tendencies.
Experiential SystemProcess

This article was published Online First April 21, 2016.

Seymour Epstein, Department of Psychology, University
of Massachusetts Amherst; Martha L. Epstein, Park City,
Utah.
Seymour Epstein is now retired.
This research was supported by National Institute of
Mental Health (NIMH) Research grant MH 01293 and
NIMH Research Scientist Award 5 K05 MH00363 to Seymour Epstein.
Correspondence concerning this article should be addressed to Seymour Epstein, 37 Bay Road, Amherst, MA
01002. E-mail: SEpstein3@comcast.net
116

There are three important ways humans and

other higher-order animals automatically learn
from experience: classical conditioning, instrumental/operant conditioning, and observational
learning (i.e., modeling; Epstein, 2014). The different learning methods comprise a single system
because they all serve a common purpose. That is,
regardless of how we learn, if we feel good or bad,
it reinforces an implicit need and/or belief. This is
fundamentally different from classic learning theory that narrowly links reinforcement to modify-

INTEGRATIVE PSYCHOTHERAPY

ing observable behaviors. This key difference

gives CET tremendous flexibility and integrative
power by providing the content for the construction of an implicit theory of reality, which directs
the interpretation of events, behavioral response
tendencies, and implicit thoughts and feelings.

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Experiential SystemContent
Depending on the fulfillment or frustration of
an implicit need, a person will have a related
implicit belief (Epstein, 2014). For example, if
the need for interpersonal relatedness is fulfilled, a person will tend to view the self as
love-worthy and others as caring. Alternatively,
if the need for relatedness is consistently frustrated, the person will tend to have the implicit
belief that the self is unlovable and others are
rejecting. Such implicit beliefs, coming from
the interpretation of life events, may or may not
coincide with a persons explicit beliefs.
Discrepancies between the two kinds of beliefs
can cause maladjustment. For example, assume
that John has worked hard all his life, as he believed that becoming famous and wealthy would
make him happy. However, instead of being
happy when he achieved his goal, he felt sad. An
explicit need may have been fulfilled, but not an
implicit need, such as the need to be loved.
A common fallacy is to equate implicit cognition with unconscious processing. Implicit cognitions are not equivalent to unconscious beliefs for
two reasons. First, if an unconscious belief is
made conscious it continues to operate as an implicit cognition that follows the same principles as
before. Second, the Freudian view of the unconscious considers the unconscious as operating by
primary process thinking (e.g., wish-fulfillment,
displacement, condensation, overdetermination,
association), whereas the experiential system has
been empirically shown to operate by associativelearning principles (Epstein, 2014).
Research Underpinnings of
Experiential Knowing
Operating Principles of the
Experiential System
The first author and his associates have conducted extensive research on the operating principles and attributes of the experiential system
(Epstein, 2014). Among other procedures, we

117

adapted procedures used by Tversky and Kahneman (1982) and other cognitive and social
cognitive psychologists to study heuristic information processing by using specially
constructed vignettes. Most research on heuristics has failed to distinguish between two kinds
of heuristic processing. Thus, some researchers
interpret all their results on heuristics as indicative of only a single system that includes deliberative processing. Alternatively, others attribute all their results on heuristics to two
separate systems, only one of which includes
deliberate processing. This latter approach has
been criticized by noting that the results supporting a different kind of information processing from deliberative processing could just as
well be attributed to cognitive shortcuts in a
single processing system (Kruglansky, Thompson, & Spiegel, 1999). Ways of establishing that
a different system of processing exists in addition to deliberate processing are to demonstrate
that they operate by different principles and
attributes and may even conflict with each other. This approach was used in our research, and
our results clearly support the existence of twosystems.
Irrational reactions to unfortunate arbitrary outcomes. People often react to arbitrary outcomes as if they or others are responsible for the outcomes. Thus, they reward the
proverbial bearer of good tidings and punish the
conveyer of bad news. Such irrational reactions
are readily explained by the associativelearning principle of the experiential system.
We investigated peoples reactions to arbitrary unfavorable outcomes by asking how they
would behave in situations described in several
vignettes, how most people would behave, how
the protagonist in the vignettes would behave,
and how a completely logical person would
behave (e.g., Epstein, Lipson, Holstein, & Huh,
1992). In one vignette that investigated the effect of near versus far misses, a woman missed
her flight due to a lengthy, unforeseen traffic
jam. In one group, a woman missed her flight by
a small margin; for another group, she missed
her flight by a much wider margin. Who felt
more foolish about having dawdled at home?
Tversky and Kahneman (1982) found that their
participants reported that the protagonist who
barely missed her flight was more upset, a finding we replicated. We also found that most
participants reported that a logical person would

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

118

EPSTEIN AND EPSTEIN

feel much less foolish than they expected the

protagonist to feel. The significance of the consequences of missing the flight also influenced
the protagonists behavior. When the outcomes
were very unfavorable, the protagonists were
said to feel more foolish for having dawdled
than when the outcomes were less unfavorable.
The full results of this study support the following hypotheses based on the CET model:
People are aware of two different modes of
thinking, one that determines mainly how
people normally behave, which, according
to CET, is with their experiential system,
and the other that determines how a logical
person would behave, which, according to
CET, is primarily by their rational system.
The more people are emotionally engaged
in a situation the more they tend to think
and behave according to their experiential
system.
Most peoples ratings for themselves were
between those of a logical person and
their ratings of most peoples behavior,
suggesting that most behavior is based on a
compromise between the two systems.
The experiential system can coopt the rational system, indicating an important
source of irrationality in humans, despite
their capacity for rational thinking.
The ratio-bias phenomenon. The ratiobias (RB) phenomenon is of particular interest
for dual-process theories, as it pits experiential
against rational processing in a situation in
which the two modes of processing are equally
accessible and in which both require negligible
cognitive effort (Kirkpatrick & Epstein, 1992).
Imagine if every time you blindly draw a red
jellybean from a bowl containing red and white
jellybeans that you win 2 dollars. You can
choose between two bowls that both offer a
10% probability of obtaining a red jellybean.
One bowl contains one red and nine white jellybeans. The other contains 10 red and 90 white
jellybeans. When people are asked how much
they would pay to draw from the bowl of their
choice, almost all said they had no preference
and would not pay a cent. Yet, when they are
placed in a situation in which they can actually
win a significant amount of money over trials,
most people parted with small sums of money to
draw from the bowl with more red jellybeans
(Kirkpatrick & Epstein, 1992). The preference
for the bowl with more red jellybeans, despite it

having no probability advantage, is referred to

as the RB phenomenon and can be attributed to
the experiential system.
Even more impressive are the results when a
bowl with more red jellybeans offers a lower
probability of obtaining one. In one study, a
probability-advantaged bowl always contained
1 in 10 red jellybeans, and a frequencyadvantaged bowl contained 59 red jellybeans
out of 100 jellybeans (Denes-Raj & Epstein,
1994). Most participants preferred the slightly
frequency-advantaged (e.g., 9/100 vs. 1/10)
bowl despite its probability disadvantage. Some
participants said they knew such behavior was
irrational, but that they had a feeling that they
had a better chance of getting a red jellybean
when there were more of them.
Additional RB studies are summarized elsewhere (Epstein, 2014), and the conclusions
from these RB studies are below:
People operate with two independent systems, experiential and rational.
The RB effect in low-probability conditions can be attributed to a frequency effect. This is consistent with the concrete
operation of the experiential system and
that single numbers are more concrete than
percentages (based on the relationship between two numbers).
Either system can dominate depending on
the situation and the person. Although most
participants make compromises between
the two systems by making nonoptimal,
slightly frequency-advantaged responses,
many consistently make optimal, probability-advantaged responses.
The two systems sometimes conflict with
each other. Some participants reported feeling torn between frequency-advantaged
and probability-advantaged choices. Clinically, we know that patients may indicate
that they know intellectually something
bad will not happen, but nonetheless believe it emotionally, which often determines their behavior.
The experiential system is more responsive
than the rational system to real situations
and to vividly imagined situations, whereas
the rational system is more responsive to
verbal information.
With increasing maturation from childhood
to adulthood, the rational system gains increasing dominance. The dominance of the

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

INTEGRATIVE PSYCHOTHERAPY

experiential system is clearly manifested

when conducting therapy with adolescents
before they have developed a more realistic
way of viewing the world.
Priming the rational system in children before they have formal knowledge of ratios
by asking them to give the reasons for their
responses interferes with their experiential
thinking. This indicates that processing in
the rational system can interfere with processing in the experiential system.
An increase in incentive produces an increase in optimal, rational responses in
some participants, but an increase in experiential processing, nonoptimal responses
in an approximately equal number of participants. This suggests that these two systems are always operational and the interaction between them depends on both the
persons predisposition (e.g., tendencies to
process rationally vs. experientially) and
the situation (e.g., importance, emotional
involvement, etc.).
The global-evaluation heuristic. The
global-evaluation heuristic refers to the tendency for people to evaluate others as generally
being either good or bad people, rather than
restricting their evaluations to specific behaviors (Epstein, 2014). As the experiential system
operates holistically, it follows that global evaluation is attributable to the experiential system,
and therefore, will be automatic, compelling,
and not easily overcome. This heuristic is particularly important because of its prevalence in
causing serious problems, as in stereotyping
others and in misattributing the motives and
emotions of others.
The first author and his associates investigated the global-evaluation heuristic (Epstein,
1994) by having participants respond to a vignette adapted from a study by Miller and Gunasegaram (1990). In the vignette, a rich benefactor tells three friends that if each throws a
coin that comes up heads, all three will be given
$100. The first two throw a heads, but Smith,
the third, throws a tails. Most participants reported that Smith would feel guilty and the
others would be angry at him. In an alternative
version in which we reduced the stakes to a
more modest sum, the ratings of guilt and anger
were reduced. When asked if the other two
would be willing to invite Smith on a gambling
vacation in Las Vegas, where they would share

119

wins and losses, most participants said that now

they would not invite him, because hes a
loser. These responses were made both from
the perspective of how the participants reported
they would react and how they reported most
people would react. When responding from the
perspective of how a completely logical person
would behave, most said a logical person would
recognize that the outcome of the coin tosses
was completely arbitrary, that they would not
resent Smith, and that they would invite him on
their gambling venture. This study indicates that
people are intuitively aware of two systems of
information processing that operate in a manner
consistent with the operating principles and attributes of the experiential and rational systems.
Conjunction problems. The Linda conjunction problem is probably the most researched vignette in the history of psychology
(Epstein, 2014). It has evoked great interest
because of its paradoxical results. Although the
solution to the Linda problem requires the application of one of the simplest rules of probability, almost everyone, including individuals
who are sophisticated about statistics, use their
intuition rather than logic when responding to it.
In the vignette, Linda is described as 31 years
old, single, outspoken, and very bright. In college she was a philosophy major who participated in antinuclear demonstrations and was
concerned about social justice. In our research,
we asked participants to rank-order the following alternatives: Linda is a feminist, Linda is a
bank teller, and Linda is a feminist and a bank
teller. Most ranked Linda as being a feminist
and a bank teller ahead of Linda just being a
bank teller. In doing so, they made what Tversky and Kahneman (1982) labeled a conjunction error (CE), because, according to probability theory, the occurrence of two events
cannot be more likely than the occurrence of
only one of them.
The usual explanation of the high rate of
conjunction errors (CEs) elicited by the Linda
problem is that people either do not know the
conjunction rule or they do not think of it in the
context of the Linda vignette. According to
Tversky and Kahneman (1982), they make such
errors because they respond by the representativeness heuristic in making judgments, according to which being both a bank teller and a
feminist is more representative or likely of Lindas personality than being just a bank teller.

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

120

EPSTEIN AND EPSTEIN

In several studies on conjunction problems,

including several on the Linda problem (Donovan & Epstein, 1997; Epstein, Denes-Raj, &
Pacini, 1995; Epstein, Donovan, & Denes-Raj,
1999), we demonstrated that the major reason
for the difficulty of the Linda problem is that it
provides a very unusual way of presenting a
conjunction problem. Consequently, people
tend not to view the Linda problem as a probability problem, even when told to do so.
The difficulty of the Linda problem cannot,
however, be fully explained by the misleading
manner in which it is presented in the form of a
personality problem, for even with full disclosure about the nature of the problem and the
request to treat it as a probability problem, most
participants make CEs. However, when responding to a conceptually equivalent problem
presented in a concrete, natural manner, as in a
problem in judging the likelihood of winning
one or two lotteries or one or two horse races,
almost no one makes a CE (Epstein et al.,
1995).
Making CEs in the Linda problem is so compelling that when presented with the correct
conjunction rule (i.e., two events are less likely
to occur than just one of them) among several
incorrect rules, most selected a wrong rule that
was consistent with making a CE. In other
words, they made the rule correspond to their
incorrect response, rather than using the correct
rule and then applying it appropriately to the
Linda problem. This demonstrates the highly
compelling nature of experiential processing
and its ability to coopt rational processing.
Priming intuitive knowledge by presenting
easy-to-solve, natural conjunction problems followed by unnatural ones, like the Linda problem, facilitates solving unnatural problems that
people are otherwise unable to solve.
Conclusions from the Linda studies are summarized below:
The difficulty of the Linda problem can be
explained by the operating rules and attributes of the experiential system, which is
the mode employed by most people when
responding to it. Thus, people automatically tend to process information associatively, concretely, holistically, and in a narrative way, rather than abstractly,
analytically, and rationally when responding to the Linda problem.

Processing in the experiential mode is intrinsically highly compelling and can override processing in the rational mode even
when the latter does not require more effort
or resources. Thus, many participants, despite knowing and thinking of the conjunction rule, simply find an experiential response more compelling than a rational
response.
The experiential system can contain learning that is not understood and cannot be
articulated by the rational system.
The rational system, especially when it is
evoked initially, can influence the experiential system.
Interactions Between Systems
and Among Needs
Interactions between the two processing
systems. An important research finding is that
the experiential and rational systems are both
simultaneously and sequentially interactive. Simultaneous interaction was demonstrated in the
compromises between the two systems in the
ratio-bias studies described earlier. Sequential
interaction was demonstrated in research that
shows that processing in the experiential mode
can compromise subsequent processing in the
rational mode (Chaiken & Maheswaran, 1994;
Denes-Raj, Epstein, & Cole, 1995; Epstein et
al., 1992). There is also considerable evidence
that priming the experiential system subliminally can influence subsequent responses in the
rational system (Bargh, 1989). Research on
conjunction problems in which presenting problems first in a natural way that taps the experiential system made participants more likely to
successfully solve abstract, out-of-context problems presented subsequently suggests that experiential processing may sometimes also improve rational processing.
Interactions among basic needs. A basic
assumption in CET is that behavior often represents a compromise among basic needs. This
process is considered to be particularly important, as it provides checks and balances against
the excessive investment in the fulfillment of a
particular need. It occurs because greatly fulfilling any need (e.g., the need for relationship) at
the expense of fulfilling other needs (e.g., the
need for autonomy) increases the strengths of
the other needs. To test this assumption about

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

INTEGRATIVE PSYCHOTHERAPY

interactions among needs we examined the

combined influence of the needs for selfenhancement and self-verification. Swann and
his associates previously demonstrated that the
needs for enhancement and verification often
operate sequentially, with the former preceding
the latter (e.g., Hixon & Swann, 1993; Swann,
1990). We demonstrated that the interaction
also occurs simultaneously in the form of compromises between needs. Our procedure consisted of varying the favorableness of evaluative
feedback and observing whether participants
had a preference for feedback that either
matched or was more favorable than their selfassessments (Epstein & Morling, 1995; Morling
& Epstein, 1997). In support of this hypothesis,
participants preferred feedback that was only
slightly more favorable than their self-assessments, consistent with a compromise between
fulfilling the needs for verification and selfenhancement.
Individual Differences in Experiential and
Rational Information Processing
Individual differences in the experiential
system. There are excellent intelligence tests
that measure the intelligence of the rational
system. Such tests are fairly good predictors of
academic performance and, to a somewhat
lesser extent, of performance in the real world,
including workplace performance, particularly
in situations that require complex operations
(see Gordon, 1997; Gottfredson, 1997; Hunter,
1983). However, as there was no test that measured experiential intelligence we constructed
one, which we named the Constructive Thinking Inventory (CTI; Epstein, 2001).
People respond to the CTI by reporting the
degree to which they have certain adaptive and
maladaptive automatic or spontaneous thoughts
or feelings. An example of an item is, I tend to
dwell more on pleasant than unpleasant incidents from the past. The CTI provides a Global
Constructive Thinking scale and six main
scales, all but one of which have subscales. The
six main scales are Emotional Coping, Behavioral Coping, Categorical Thinking, Esoteric
Thinking, Nave Optimism, and Personal Superstitious Thinking, and are predictive of a
wide variety of criteria related to success in
living. A review of the extensive literature supporting the CTIs validity is available in the CTI

121

manual (Epstein, 2001). Experiential intelligence has been found to be directly related to a
wide range of successful performance, including performance in the workplace, superior
achievement by managers, superior academic
performance, social competence, leadership
ability, ability to cope with stress, emotional
adjustment, physical wellbeing, and an absence
of drug and alcohol abuse.
It was found in several studies that scores on the
Global CTI scale were either unrelated or negligibly
related to IQ (see review in Epstein, 2001), supporting the view that the experiential and rational systems process information differently. Of particular
interest, constructive thinking and intellectual intelligence exhibit a nearly opposite course of development across the life span. Constructive thinking is at
its nadir in adolescence when intellectual intelligence
is near its peak and then gradually increases throughout most of the adult years when IQ is gradually
declining. Constructive thinking is less strongly related to academic achievement than intellectual intelligence. Nevertheless, it contributes significant
variance beyond that of intellectual intelligence to
performance in the classroom, as indicated by grades
received and by class rank (Epstein, 2001). Apparently, good constructive thinkers are able to behave
in ways that gain them recognition for and appreciation of their intellectual ability, whereas poor constructive thinkers are less able to do so.
A finding of particular importance is that there
is a general factor of experiential intelligence that
is measured by the CTI. As this factor includes a
wide variety of favorable nonintellective abilities
and attributes, it is suggestive of a general factor
of nonintellective ability comparable to the general factor of intellectual ability.
A measure of individual differences in rational and experiential thinking styles. The
Rational-Experiential Inventory (REI) was constructed to measure individual differences in the
extent to which people process information according to the two processing modes (Epstein,
2001). The REI includes two main scales that
measure the extent of processing information in
rational and in experiential thinking styles. Each
of the main scales has subscales of the degree to
which one uses a particular processing mode
and ones self-assessed ability to use it effectively. In studies with the REI (Epstein, 2001;
Epstein, Pacini, Denes-Raj, & Heier, 1996;
Norris & Epstein, 2011; Pacini & Epstein,
1999; Pacini, Muir, & Epstein, 1998), a rational

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

122

EPSTEIN AND EPSTEIN

thinking style was more strongly positively associated than an experiential thinking style with
intellectual performance, as well as with low
anxiety, low depression, low stress, low neuroticism, high self-esteem and high meaningfulness of life. An experiential thinking style was
more strongly associated than a rational thinking style with measures of creativity, empathy,
aesthetic judgment, intuitive ability, and establishing satisfactory interpersonal relationships.
It follows that no statement can be made about
the general superiority of either thinking style,
as each is superior in some important ways and
inferior in other important ways.
Individual differences in basic beliefs
about the self, others, and the impersonal
world. The content of beliefs in the experiential system is of particular importance for clinical practice. Peoples implicit beliefs about
themselves, others, and the impersonal world
are acquired automatically from emotionally
significant experiences, mostly in childhood.
Such beliefs are important determinants of feelings, behavior, adjustment, and overall quality
of life. Based on this consideration, the Basic
Beliefs Inventory (BBI; Catlin & Epstein, 1992)
was constructed. The BBI is a self-report questionnaire that includes a global measure of the
overall favorability of basic beliefs. It also includes scales for measuring the favorability of
more specific beliefs: self-esteem, loveworthiness, competence, meaningfulness of life,
views about relationships, and views about the
world.
A study with the BBI (Catlin & Epstein,
1992) tested the hypotheses that two important
sources of basic beliefs are extreme life events,
as in the loss of a loved one or experiencing a
transforming love relationship, and relationships with parents during early childhood. In
support of the hypothesis, both kinds of experiences were related to individual differences in
basic beliefs. People who reported favorable
relationships with their parents and those who
obtained high scores on overall favorableness of
extreme life events based on both positive and
negative events obtained higher scores than others on the BBI scales of self-esteem, meaningfulness of life, favorable views of others, and an
optimistic view about the world. Those who
reported unfavorable relationships with their
parents or who reported predominantly unfavorable extreme life events, such as the death of a

beloved pet or a significant failure in an important event, obtained less favorable scores than
others on all basic beliefs.
It was also found that the more recently an
extreme negative event occurred, the less favorable a persons current basic beliefs. Over time,
the effect of extremely unfavorable life events
became less unfavorable and sometimes even
favorable. Several participants reported that
they ultimately gained strength from having had
to cope with adversity (Epstein, 1979).
A General Cognitive-Experiential
Framework for Psychotherapy
Cognitive-Experiential Psychotherapys
(CEP) major contribution to psychotherapy is
that it provides an integrative framework for all
forms of psychotherapy. CEP is based on the
following:
1. The automatic, experiential learning in the
form of implicit needs and beliefs is of
particular importance and much more so
than the learning of behavioral responses.
2. According to CEP, psychopathology is the
result of implicit needs and beliefs that,
regardless of how they were acquired, are
currently maladaptive. Identifying maladaptive implicit needs and beliefs is important because it reveals the problems
that have to be resolved. However, this is
more easily said than done. Since a need
or belief may be implicit, the client will be
unaware of it and, therefore, unable to
report it. Alternatively, the client may
have no awareness of the maladaptive implicit need or belief, because he or she has
repressed it (i.e., has an automatic avoidance reaction to it). Lastly, one may be
aware of the maladaptive need or belief
and be able to articulate it, but avoid discussing it because it is too upsetting to do
so. A number of procedures can be used to
uncover implicit maladaptive needs and
beliefs. One procedure for identifying
such needs and beliefs is to look for maladaptive repetitive behavior patterns in the
real world or in the psychotherapy sessions. Another is to attend to automatic
thoughts. Still another is to infer from
negative emotions the implicit beliefs that
are their likely source.

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

INTEGRATIVE PSYCHOTHERAPY

3. Intellectual insight is important, as it provides a road map to what important

changes in the experiential system need to
be corrected, but insight alone is insufficient.
4. Successful therapy requires changes in the
experiential system. Therapists can use
the considerable information that is available about automatic, associative learning
to bring about these changes in the experiential system.
In most cases, therapy based on CEP begins
with instructions on the rudiments of the two
systems, and that the source of an emotion is an
underlying implicit or explicit cognition. Remediation has two components, diagnosis and
treatment. The diagnostic phase requires identification of the dysphoric emotion that is experienced (e.g., fear, anger, sadness, jealousy,
envy), the underlying belief or cognition, and
the unmet need that is the root cause. For example, anger (the dysphoric emotion) may result from a heightened sensitivity to believing
that one has been unjustly treated, which may
stem from an unfulfilled need for unconditional
love, likely stemming from ones treatment as a
child.
The treatment phase consists of fulfilling the
frustrated need in a constructive way. An important starting point is to establish a therapeutic relationship, so the client genuinely experiences being treated with unconditional positive
regard. It is also helpful for the client to learn to
identify and reframe maladaptive thinking in a
constructive manner. The first author taught a
college class in constructive thinking that
helped students become skilled at doing this.
Students were asked to keep a journal in which
they recorded the most negative experience of
each day for 30 days. In addition to recording
the relevant situation (e.g., a bad grade, a distressing conversation with a parent, a romantic
breakup) students were asked to note their accompanying emotions (e.g., anger, frustration,
sadness). Most importantly, they were required
to explore what implicit or explicit thought produced the emotion and to consider alternative
ways of thinking about the same situation that
would lead to more constructive feelings and
actions. Class time was used to discuss what
students had written and learned about themselves, as well as for other students to offer

123

additional constructive ways of reframing their

thinking. Anecdotally, most students reported
that this was a growth experience, as they had
not previously understood how thoughts influence feelings and that the way people frame a
situation can affect their well-being.
This process can also be used in a therapeutic
setting, with journal-keeping and discussion.
Writing or typing out entries in full is preferred
because it is more deliberate (than say talking,
texting, dictating, and/or using emoji shorthand)
and allows one to reexperience the situation and
allows more time to attend to accompanying
thoughts and emotions. The main difference
between the class and the therapeutic setting is
the presence of other students to offer reframing
options. In therapy, reframing options can be
explored by discussion between the client and
therapist. Of course, prompting clients to come
up with constructive reframing options on their
own is preferable. Yet, if clients struggle to
accomplish this, the therapist can offer possibilities. As clients replace their maladaptive
thoughts with more constructive ones, it is important for them to attend to the different feelings that result. As one does this repeatedly, the
positive feelings that result from having more
constructive thoughts should be self-reinforcing
and will therefore become increasingly automatic.
There may be times when clients are unable
to discern any thoughts that accompany their
emotions. In such cases, a therapist can work
with clients to infer their implicit thoughts from
their emotions. For example, sadness is likely to
be the result of feeling one has suffered a significant, irreplaceable loss. Likewise, anger often follows an implicit thought that one has
been treated badly and unfairly.
It is also helpful to provide and encourage
other constructive experiences (either in real
life, in therapy, or in fantasy) and to encourage
clients to apply their new way of thinking constructively to these experiences.
The final step is to discuss with the client
adaptive procedures for coping with the situation after therapy. Specifically, there should be
an emphasis on the importance of establishing
positive relationships and avoiding toxic ones.
If the client returns too soon to the same environment that created and reinforced feelings of
being unlovable, it may produce a relapse from
the improvement that had taken place.

124

EPSTEIN AND EPSTEIN

Since CEP is integrative, there are several

standard forms of psychotherapy that use treatment procedures that are relevant to CEP. The
first authors recent book (Epstein, 2014) discusses all of them from the perspective of CET.
In the present article, because of space limitations, we will only discuss client-centered therapy and psychosynthesis.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Client-Centered Therapy
An important agreement between Rogers
theory and CEP is that both assume that peoples beliefs are the most important sources of
their behavior and feelings. The two theories
also agree that discrepancies between beliefs
acquired from direct experience (e.g., if a child
is consistently rejected by their parents) and
how that experience is labeled by significant
others (e.g., the rejecting parents tell the child
how much they love the child) are an important
source of maladjustment.
Both also assume that people have a basic
need to maintain and enhance their conceptual
systems. However, for Rogers this is the only
basic need whereas for CEP there are six basic
needs, two of which are superordinate (behaving according to the hedonic principle and controlling arousal within homeostatic limits) and
four of which are subordinate (the needs for a
stable, coherent conceptual system; security; relatedness; and self-esteem). In addition, clientcentered therapy relies on only one procedure,
reflecting clients statements with acceptance
and positive regard. According to CEP, this
procedure is insufficient, and it is preferable
once a therapeutic relationship has been established to change from an initial reflective approach to an approach that is based on an equal
relationship between therapist and client, where
the therapist feels free to make clarifications and
interpretations, albeit in a warm, accepting, and
tentative manner. It should be noted that, according to CEP, an important reason for the
effectiveness of client-centered therapy may be
the experience of a warm and accepting relationship with a person who likes and respects
the client apart from the particular technique
that is used.
Psychosynthesis and The Use of Fantasy
If the aim of therapy is to produce changes in
the experiential system, the use of fantasy can

help to do this for several reasons. First, people

react to vividly imagined situations in a similar
way to how they react to real situations (Crisp,
& Turner, 2009; Dadds, Bovbjerg, Redd, &
Cutmore, 1997; Epstein & Pacini, 2001; Kosslyn, Ganis, & Thompson, 2001; Taylor, Pham,
Rivkin, & Armor, 1998). Second, the same principles of learning theory can be applied to representations in fantasy as to representations in
reality. Third, one can try out things in fantasy
that would be impractical or too risky (emotionally and/or physically) to do in reality. Fourth
and perhaps most importantin fantasy, one
can create and implement powerful metaphors
laden with symbolism that can reach deeply into
the experiential system, triggering change.
Psychosynthesis relies heavily on fantasy in
therapy. What follows is an in-depth look at a
single case of the use of fantasy in psychosynthesis, that of Alice Epsteinthe first authors
wifeand her use of fantasy in her successful
battle against terminal cancer (A. Epstein,
1989). When traditional medical treatments offered little hope, Alice was given 3 months to
live. She decided to explore alternative approaches to fight her cancer or at least to improve the quality of her life in her remaining
months. One of the approaches was psychotherapy. Alice chose a therapist who used a psychosynthesis approach and, unlike other therapists
we interviewed, said she would help Alice work
on becoming the person she wanted to be with
whatever time was left to her. That was exactly
what Alice wanted to hear.
The psychosynthesis approach relies heavily
on the use of imagery and fantasy. One of the
psychosynthesis tools Alice found most helpful
was the creation of subpersonalities. A subpersonality is a personified aspect of oneself. One
vividly imagines this subpersonality as a person, giving it a name and personality. It can be
aggressive, strong, empathic, helpful, or anything else one can think of. One of the tools of
psychosynthesis is to ask each subpersonality
what it really needs and wants. At some point,
when the subpersonality is ready, each subpersonality is taken up the mountain. Going up
the mountain is a metaphor for moving to a
higher state of perception and understanding.
By going up the mountain the subpersonality is
often transformed, happier, and more accepting
of and cooperative with the other subpersonalities. The final aim of psychosynthesis is to

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

INTEGRATIVE PSYCHOTHERAPY

create an integrated and constructive overall

personality.
Alice embraced identifying and exploring her
subpersonalities with great interest and enthusiasm. Her first subpersonality was Baby Alice.
Baby Alice is about 2 1/2 years old and sits in
a corner and cries. She knows that her mother is
always right. Her main wish is to have a perfect
mother, and this subpersonality will make significant compromises to believe in her mothers
perfection. Baby Alice even lies to her father to
cover up her mothers failings. Baby Alice says
she wants a perfect mother. What she really
wants at the experiential level is unconditional
love. Out of the five major subpersonalities,
Baby Alice was the one who developed the
most until she became a highly competent and
mature adult who was the leader of the subpersonalities and helped them finally become a
unified whole.
Alice quickly moved on to creating other
subpersonalities. Hope and hopelessness were
very important to her. She envisioned a crab
running aimlessly on the beach holding its front
legs up in the air in a helpless manner. He
belonged to a gray haired lady who spoke
sweetly to him and always called him dear.
His need is to be free, but she tells him thats
impossible because she needs him to cut the
threads on a tapestry she is making. When he
asks her what he can hope for, she laughs and
says he can hope to be the flying horse in the
tapestry.
The crab was the first one Alice took up the
mountain. He got excited when he reached a
ledge at the top of the mountain and he jumped
off. Alice expected to see his mangled remains
at the bottom of the hill. Instead there was a
beautiful Baltimore oriole that sang so sweetly
that she knew he was happy. He perched on her
shoulder, and he continued to do so on all
further trips up the mountain.
The next subpersonality was Amanda, who
was very strong and competent. Alice kept
chanting to herself when she felt depressed or
helpless, I am Amanda. I am strong. Kill the
cancer. Kill the cancer. Kill the cancer. I am
Amanda. I am strong.
Little One was the next subpersonality. She
is very angry and aggressive. She throws her
head back and laughs a lot. She is nasty, aggressive, devious and joyously happy about her
evil thoughts and deeds. She hates Baby Alice

125

and thinks she is stupid and deserving of contempt. Alice found Little One very interesting
and liked her. When Little One was taken up
the mountain, she became Bagera, a beautiful
tiger who uses her power only when justified.
She likes Baby Alice and lets her ride on her
back.
Mickey is the most complicated subpersonality. She hates and wants to kill Baby Alice.
Mickey is a very jealous little girl. She wants to
be admired by her playmates, but no one likes
her because she takes away their things. And
she is critical of everyone. In the valley below
the mountain, she keeps saying that the flowers
at home are more beautiful than the wild flowers. All the other subpersonalities taken up the
mountain had highly positive transformations.
In contrast, Mickey became Chicken Little,
who considered her life to be a sham and, if she
were found out the sky would fall. Alice
indicated that Mickey was the ghost in her
closet, a dark part of her that she hid while
presenting an opposite picture of herself to the
world.
Alice decided to take Mickey up the mountain again, but while still in the meadow she
became a giant chicken who came running after
Alice as if she wanted to kill her. Alice realized
this was the part of her that wanted her dead.
She did not know what to do or say, so she said,
I forgive you. With that, the giant chicken
turned into a huge vampire bat that came
straight at Alice. Alice wrestled her down and
stabbed her to death.
How does one explain the dire consequence
of providing forgiveness in this fantasy? Forgiveness implies that someone has done something wrong and, as the chicken sees it, the
chicken is not at fault. Instead, Alice is at fault
because of her false personality and the things
she is hiding. It is Alice who is not only at fault;
it is Alice who does not even deserve to live.
The next day in therapy when Alice told her
therapist about this fantasy, her therapist got
upset and told her never to kill anything again.
The therapist was sympathetic to Mickey and
asked Alice to try to love her. Alice responded
by saying, How can I love someone who is
trying to kill me? Right now I hate her.
Sometime later, Alice told Mickey she had to
go. A clear and powerful image suddenly came
to mind. She saw Mickey standing at a train
station carrying a small suitcase. She was all

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

126

EPSTEIN AND EPSTEIN

alone and Alice was walking away telling her

she could not come. Mickey sobbed and sobbed,
begging to come with Alice. It broke Alices
heart to see her that way. Suddenly Alice realized that it was Mickey who was in fear of her
life. Alice realized that Mickey contained all her
passion and she could not live without her.
Alice once more took Mickey and all the
other subpersonalities up the mountain.
Amanda changed into a flying horse and invited
all of them to ride on her back. They had a
wonderful time with Amanda catching the wind
and soaring into space. Alice now loved all her
subpersonalities and they were beginning to cooperate with each other.
The main source of Alices maladjustment
warrants discussion. She believed, implicitly
and explicitly, that unless you sacrifice fulfilling
your own needs for fulfilling other peoples
needs, no one would like you. Consequently,
Alice consistently subjugated her own needs.
This left her prone to feelings of rage and to
antisocial behavioral tendencies, which she
needed to suppress constantly. Moreover, the
conflict between wanting her own needs fulfilled but feeling she had to sacrifice her own
needs in order to be liked and loved pitted parts
of her personality against one another, resulting
in an overall lack of integration. The use of
fantasy effectively addressed all these conflicts.
Alices self-sacrificing attitude was overcome by the transformations that occurred on
her trips up the mountain. For example, Little
One evolved into Bagera, a beautiful and powerful tiger, that asserts her own needs but only
when appropriate. In addition, when the crab
went up the mountain, it broke free of its
controlling owner, leapt off the ledge, and was
changed into a beautiful, joyous bird. This is
likely a metaphor for Alice believing that if she
took a leap of faith and broke free of subjugating her needs to those of others that she
could have a rewarding life instead of having a
life without freedom and love.
Alices suppressed frustration and anger were
addressed in two important ways. First, they
were released by instilling in Mickey and Little
One the ability to express hostility freely. Second, and more importantly, by showing Alice
how to communicate her needs in a socially
acceptable way, Bagera helped Alice become
less angry and frustrated and to be more authentic, satisfied and fulfilled.

Lastly, Alices unintegrated personality was

improved by Baby Alice encouraging cooperation among the elevated forms of the subpersonalities at every opportunity. Alices image of
Amanda as a flying horse and taking the other
subpersonalities for a wonderful ride on her
back was likely only possible when Alices
fundamental unmet need was finally satisfied,
allowing the subpersonalities to be at peace and
work harmoniously with each other.
Summary and Conclusions
Cognitive-Experiential Therapy (CET) rests
mainly on two basic assumptions: (a) in some
fundamental respects, people are no different
from other higher-order animals in that they
share the same experiential system, and (b) humans did not give up their experiential systems
when they developed language and a rational
system. Thus, to understand people, it is necessary to recognize that most behavior is primarily
automatically determined by their experiential
system. The aim of therapy, therefore, should be
to produce changes in the experiential system.
Cognitive-Experiential Psychotherapy (CEP)
draws attention to the importance of the rules
that govern the operation of the experiential
system (e.g., reinforcement, reward, punishment, extinction, spontaneous recovery). These
are well known aspects of associative learning,
such as having good or bad feelings after behaving in a certain manner serves to reinforce
the behavior. That is, behavior followed by
good feelings tends to be repeated, while behavior followed by bad feelings tends to be avoided. Because this normally occurs outside of
awareness, it has obvious implications for psychotherapy. Namely, it indicates the importance
of identifying the unmet need that is the root
cause of the problem.
To produce change, it is insufficient to simply
provide intellectual knowledge; one must reach
a persons experiential system. To do so, it is
necessary to communicate with it via the language it understands, namely the language of
associative learning (e.g., of having new experiences with positive outcomes). This can be
done by providing corrective experiences and
by continuously practicing replacement of the
maladaptive implicit thinking with more constructive implicit thinking. Specifically, one has
to provide corrective experiences in everyday

INTEGRATIVE PSYCHOTHERAPY

life, in fantasy, and in therapy sessions. Relapse

can be avoided by correcting the root cause of
the maladaptive, automatic thinking with repeated corrective experiences coupled with constructive ways of viewing the experiences and
by discussing coping with the environment to
which the client expects to return.

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

References
Bargh, J. A. (1989). Conditional automaticity: Varieties of automatic influence in social perception
and cognition. In J. S. Uleman & J. A. Bargh
(Eds.), Unintended thought (pp. 351). New York,
NY: Guilford Press.
Catlin, G., & Epstein, S. (1992). Unforgettable experiences: The relation of life-events to basic beliefs
about the self and world. Social Cognition, 10,
189 209. http://dx.doi.org/10.1521/soco.1992.10
.2.189
Chaiken, S., & Maheswaran, D. (1994). Heuristic
processing can bias systematic processing: Effects
of source credibility, argument ambiguity, and task
importance on attitude judgment. Journal of Personality and Social Psychology, 66, 460 473.
http://dx.doi.org/10.1037/0022-3514.66.3.460
Crisp, R. J., & Turner, R. N. (2009). Can imagined
interactions produce positive perceptions?: Reducing prejudice through simulated social contact.
American Psychologist, 64, 231240. http://dx.doi
.org/10.1037/a0014718
Dadds, M. R., Bovbjerg, D. H., Redd, W. H., &
Cutmore, T. R. (1997). Imagery in human classical
conditioning. Psychological Bulletin, 122, 89
103. http://dx.doi.org/10.1037/0033-2909.122
.1.89
Denes-Raj, V., & Epstein, S. (1994). Conflict between intuitive and rational processing: When people behave against their better judgment. Journal
of Personality and Social Psychology, 66, 819
829. http://dx.doi.org/10.1037/0022-3514.66.5
.819
Denes-Raj, V., Epstein, S., & Cole, J. (1995). The
generality of the ratio-bias phenomenon. Personality and Social Psychology Bulletin, 21, 1083
1092. http://dx.doi.org/10.1177/014616729521
10009
Donovan, S., & Epstein, S. (1997). The difficulty of
the Linda conjunction problem can be attributed to
its simultaneous concrete and unnatural representation, and not to conversational implicature. Journal of Experimental Social Psychology, 33, 120.
http://dx.doi.org/10.1006/jesp.1996.1309
Epstein, A. (1989). Mind, fantasy, and healing: One
womans journey from conflict and illness to
wholeness and health. New York, NY: Delacorte
Press.

127

Epstein, S. (1979). The ecological study of emotions

in humans. In P. Pliner, K. R. Blankstein, & I. M.
Spigel (Eds.), Advances in the study of communication and affect, Vol. 5: Perception of emotions in
self and others (pp. 47 83). New York, NY: Plenum Press. http://dx.doi.org/10.1007/978-1-46843548-1_2
Epstein, S. (1994). Integration of the cognitive and
the psychodynamic unconscious. American Psychologist, 49, 709 724. http://dx.doi.org/10.1037/
0003-066X.49.8.709
Epstein, S. (2001). Manual for the Constructive
Thinking Inventory. Odessa, FL: Psychological
Assessments Resources.
Epstein, S. (2014). Cognitive-experiential theory: An
integrative theory of personality. New York, NY:
Oxford University Press. http://dx.doi.org/10
.1093/acprof:osobl/9780199927555.001.0001
Epstein, S., Denes-Raj, V., & Pacini, R. (1995). The
Linda problem revisited from the perspective of
cognitive-experiential self-theory. Personality and
Social Psychology Bulletin, 21, 1124 1138. http://
dx.doi.org/10.1177/01461672952111001
Epstein, S., Donovan, S., & Denes-Raj, V. (1999).
The missing link in the paradox of the Linda
problem: Beyond knowing and thinking of the
conjunction rule, the intrinsic appeal of heuristic
processing. Personality and Social Psychology
Bulletin, 25, 204 214. http://dx.doi.org/10.1177/
0146167299025002006
Epstein, S., Lipson, A., Holstein, C., & Huh, E.
(1992). Irrational reactions to negative outcomes:
Evidence for two conceptual systems. Journal of
Personality and Social Psychology, 62, 328 339.
http://dx.doi.org/10.1037/0022-3514.62.2.328
Epstein, S., & Morling, B. (1995). Is the self motivated to do more than enhance and verify itself? In
M. H. Kernis (Ed.), Efficacy, agency, and selfesteem (pp. 9 29). New York, NY: Plenum Press.
http://dx.doi.org/10.1007/978-1-4899-1280-0_2
Epstein, S., & Pacini, R. (2001). The influence of
visualization on intuitive and analytical information processing. Imagination, Cognition and Personality, 20, 195216. http://dx.doi.org/10.2190/
G4VG-AKQP-2Q91-JQHP
Epstein, S., Pacini, R., Denes-Raj, V., & Heier, H.
(1996). Individual differences in intuitiveexperiential and analytical-rational thinking styles.
Journal of Personality and Social Psychology, 71,
390 405. http://dx.doi.org/10.1037/0022-3514.71
.2.390
Gordon, R. A. (1997). Everyday life as an intelligence test. Intelligence, 24, 203320. http://dx.doi
.org/10.1016/S0160-2896(97)90017-9
Gottfredson, L. S. (1997). Why g matters: The complexity of everyday life. Intelligence, 24, 79 132.
http://dx.doi.org/10.1016/S0160-2896(97)90014-3

This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

128

EPSTEIN AND EPSTEIN

Hixon, J. G., & Swann, W. B., Jr. (1993). When does

introspection bear fruit? Self-reflection, selfinsight, and interpersonal choices. Journal of Personality and Social Psychology, 64, 35 43. http://
dx.doi.org/10.1037/0022-3514.64.1.35
Hunter, J. E. (1983). Overview of validity generalization for the U.S. Employment Service (USES
Test research report, No. 43). Washington, DC:
U.S. Department of Labor, Employment, and
Training Administration.
Kirkpatrick, L. A., & Epstein, S. (1992). Cognitiveexperiential self-theory and subjective probability:
Further evidence for two conceptual systems.
Journal of Personality and Social Psychology, 63,
534 544. http://dx.doi.org/10.1037/0022-3514.63
.4.534
Kosslyn, S. M., Ganis, G., & Thompson, W. L.
(2001). Neural foundations of imagery. Nature Reviews Neuroscience, 2, 635 642. http://dx.doi.org/
10.1038/35090055
Kruglansky, A. W., Thompson, E. P., & Spiegel, S.
(1999). Separate or equal? Bimodal notions of
persuasion and a single process uni-model. In S.
Chaiken & E. Trope (Eds.), Dual-process theories
in social psychology (pp. 293313). New York,
NY: Guilford Press.
Miller, D. T., & Gunasegaram, S. (1990). Temporal
order and the perceived mutability of events: Implications for blame assignment. Journal of Personality and Social Psychology, 59, 11111118.
http://dx.doi.org/10.1037/0022-3514.59.6.1111
Morling, B., & Epstein, S. (1997). Compromises
produced by the dialectic between self-verification
and self-enhancement. Journal of Personality and
Social Psychology, 73, 1268 1283. http://dx.doi
.org/10.1037/0022-3514.73.6.1268

Norris, P., & Epstein, S. (2011). An experiential

thinking style: Its facets and relations with objective and subjective criterion measures. Journal of
Personality, 79, 10431080. http://dx.doi.org/10
.1111/j.1467-6494.2011.00718.x
Pacini, R., & Epstein, S. (1999). The relation of
rational and experiential information processing
styles to personality, basic beliefs, and the ratiobias phenomenon. Journal of Personality and Social Psychology, 76, 972987. http://dx.doi.org/10
.1037/0022-3514.76.6.972
Pacini, R., Muir, F., & Epstein, S. (1998). Depressive
realism from the perspective of cognitiveexperiential self-theory. Journal of Personality
and Social Psychology, 74, 1056 1068. http://dx
.doi.org/10.1037/0022-3514.74.4.1056
Swann, W. B., Jr. (1990). To be known or to be
adored: The interplay of self-enhancement and
self-verification. In R. M. Sorrentino & E. T. Higgins (Eds.), Handbook of motivation and cognition: Foundations of social behavior (Vol. 2, pp.
408 448). New York, NY: Guilford Press.
Taylor, S. E., Pham, L. B., Rivkin, I. D., & Armor,
D. A. (1998). Harnessing the imagination. Mental
simulation, self-regulation, and coping. American
Psychologist, 53, 429 439. http://dx.doi.org/10
.1037/0003-066X.53.4.429
Tversky, A., & Kahneman, D. (1982). Judgments of
and by representativeness. In D. Kahneman, P.
Slovic, & A. Tversky (Eds.), Judgment under uncertainty: Heuristics and biases (pp. 84 98). New
York, NY: Cambridge University Press. http://dx
.doi.org/10.1017/CBO9780511809477.007
Received February 2, 2016
Accepted February 4, 2016