Journal of Psychiatric Research 46 (2012) 1086e1092

Contents lists available at SciVerse ScienceDirect

Journal of Psychiatric Research

journal homepage: www.elsevier.com/locate/psychires

Intrinsic motivation and metacognition as predictors of learning potential in

patients with remitted schizophrenia
Cumhur Tas a, b, *, Elliot C. Brown a, b, Aysen Esen-Danaci c, Paul H. Lysaker d, Martin Brüne a, b
a
Research Department of Cognitive Neuropsychiatry and Psychiatric Preventative Medicine, Ruhr-University Bochum, Germany
b
International Graduate School of Neuroscience, Ruhr-University, Bochum, Germany
c
Department of Psychiatry, Celal Bayar University, Manisa, Turkey
d
Roudebush VA Medical Center, Indianapolis, IN, United States

a r t i c l e i n f o a b s t r a c t

Article history: Background: Previous research has suggested that neurocognitive functioning predicts best the potential
Received 19 January 2012 of patients with schizophrenia to acquire newly learned material, which, in turn may impact patients’
Received in revised form social functioning. Recent studies have also shown that intrinsic motivation and metacognitive abilities
24 April 2012
play a decisive role in social functioning in schizophrenia. Accordingly, the present study sought to
Accepted 27 April 2012
examine the relationship between intelligence, motivation, metacognition, and learning during
a cognitive remediation experimental training. We hypothesized that metacognition and intrinsic
Keywords:
motivation would have a strong relationship and independently predict learning potential.
Metacognition
Learning
Method: Thirty-two patients with schizophrenia who fulfilled the criteria of functional remission were
Motivation recruited. In a pre-training-post experimental design, patients’ learning potential was assessed using
Cognitive remediation previously defined cognitive remediation training for WCST. Intrinsic motivation was examined using
Schizophrenia Intrinsic Motivation Inventory for schizophrenia; mastery, a domain of metacognition, was measured
using the Metacognitive Assessment Scale.
Results: Metacognition significantly correlated with subdomains of intrinsic motivation. Patients with
higher intrinsic motivation and preserved metacognition improved more in the learning paradigm
compared to poorly motivated patients and patients with reduced metacognitive abilities. In particular,
“mastery” was determined as an independent predictor of learning potential.
Conclusions: Motivation and metacognition are important predictors of learning in schizophrenia.
Psychological interventions in schizophrenia may therefore consider incorporating techniques to stim-
ulate metacognitive and motivational abilities as well as developing individualized training programs.
Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction therapeutic techniques, there are a number of factors that have

Since Brenner et al. (1992) introduced the Integrated Psycho-
logical Therapy (IPT) as a modularized tool to improve cognition in
severe psychiatric disorders, psychosocial interventions for schizo-
phrenia have enjoyed a rapidly increasing popularity over the last 20
years. This development can largely be explained by two key find-
ings: one is based on evidence for the limited effects of psycho-
pharmacological treatment on social functioning and residual
symptoms in schizophrenia (Swartz et al., 2007), the other is the
proven efficacy of cognitive remediation and cognitive behavioral
therapy (Wykes et al., 2008, 2011). Despite the developments in new
* Corresponding author. International Graduate School of Neuroscience, LWL
University Hospital, Ruhr-University Bochum, Alexandrinenstrasse 1, D-44791
Bochum, Germany. Tel.: þ49 234 5077 1220; fax: þ49 234 5077 1119.
E-mail address: cumhur.tas@rub.de (C. Tas).
a direct influence on the efficacy of an intervention, including ther-
apeutic alliance, neurocognition, motivation and metacognition, the
individual roles of which still need to be further clarified, and may be
independent of the specificity of the intervention (Svensson and
Hansson, 1999).
Intervention studies of cognitive behavioral therapy in anxiety
disorders and cognitive remediation in post-stroke patients
(Hofmann and Smits, 2008; Rohling et al., 2009) have found higher
effect sizes from treatment when compared to similar interven-
tions in schizophrenia patients (Wykes et al., 2011). One possible
explanation for this is attributed to the neurocognitive impairment
in patients with schizophrenia (Wykes and Reeder, 2005), which
arguably has a negative impact on the potential to benefit from
training and practice (Kurtz et al., 2010), henceforth referred to as
“learning potential” (LP; Sternberg and Grigorenko, 2002). Previous
definitions of LP in schizophrenia largely focused on its link with

0022-3956/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jpsychires.2012.04.027
 C. Tas et al. / Journal of Psychiatric Research 46 (2012) 1086e1092
a person’s cognitive capacity, suggesting that LP is a crucial factor in
predicting performance after rehabilitation (Fiszdon et al., 2006).
LP has typically been measured through repeated administration of
specific cognitive tests before and after training of that particular
cognitive faculty (Lauchlan and Elliott, 2001). For instance, Kurtz
et al. (2010) evaluated LP using the California Verbal Learning
Test (CVLT; Delis et al., 2000) and the Wisconsin Card Sorting Task
(WCST; Heaton, 1981) with a pre-training-post design. They cate-
gorized a sample of schizophrenia patients into non-learners,
learners, and high achievers according to the participants’
improvement in performance after the CVLT and the WCST though
this failed to predict social functioning (Kurtz et al., 2009). Taken
together, it is plausible to predict that the LP of a neurocognitive
test has the potential to predict treatment outcome of a neuro-
cognitive remediation program as well as work skills training (Sergi
et al., 2005). A few studies have shown that LP can have predictive
power for treatment outcome in schizophrenia patients beyond
patients’ performance on neurocognitive measures (Watzke et al.,
2009; Rempfer et al., 2011). In addition, studies of LP in educa-
tional psychology suggest that motivation and metacognitive
abilities should be considered as mediating factors of learning (e.g.,
Flavell, 1979; McCabe, 2011).
Motivation has been defined as an “internal state, need, or desire
that serves to incite, direct, and maintain goal-oriented behaviors
which is believed to be implicated in all learned responses and can
either foster or hinder future actions” (Kleinginna and Kleinginna,
1981). Motivation has been conceptualized with the “self-deter-
mination theory” (SDT), which describes motives along
a continuum, with intrinsic motivation (IM) lying at one end and
extrinsic motivation (EM) at the other (Deci and Ryan, 1985). IM
refers to pursuing an action because it is inherently interesting or
enjoyable whereas EM refers to performing an action because it
leads to a distinct outcome (Ryan and Deci, 2000).
In general, motivational deficits in schizophrenia have previ-
ously been considered as part of a negative symptom syndrome
(Barch, 2005). Studies on motivation in schizophrenia have referred
to the self-determination theory of motivation (Choi et al., 2010;
Nakagami et al., 2008, 2010). For example, a recent study using
a standard and a motivational mathematical game reported that
schizophrenia patients with higher IM levels learned better than
those with lower IM levels (Choi and Medalia, 2010). In addition,
Nakagami et al. (2008) found that IM, as measured using
a composite index derived from the Quality of Life Scale for
Schizophrenia (Heinrichs et al., 1984), mediated the relationship
between neurocognition and social functioning. In view of these
findings, recent cognitive remediation strategies in schizophrenia
have included means to induce IM by educational approaches,
computer-based tasks, and virtual reality (Medalia and Thysen,
2008; Park et al., 2011).
Self-regulation refers to the capacity for altering behavior in
terms of SDT and also overlaps with features of “metacognition”. In
particular, “mastery” represents a metacognitive capacity to form
integrated narratives built around complex and coherent accounts of
how the mental states of both oneself and others change over time,
and then to use that kind of mentalistic information for purposeful
problem solving. In the case of mental illness, mastery refers to the
use of knowledge about oneself and other to effectively deal with
subjective distress and interpersonal challenges (Semerari et al.,
2003; Lysaker et al., 2005; Brüne et al., 2011). Metacognitive
mastery, for instance, concerns one’s ability to respond to unex-
pected conflicts, emergent symptoms or psychosocial setbacks on
the basis of their knowledge of themselves and others as beings with
unique sets of interrelated thoughts and feelings.
A substantial amount of research has demonstrated that meta-
cognitive deficits are related to the course of illness in schizophrenia,
1087
independent of symptomatology and neurocognition (Roncone et al.,
2002; reviewed in Brüne et al., 2011). Metacognitive impairments in
schizophrenia are associated with the capacity to monitor one’s own
thinking and behavior (Koren et al., 2006), to mentalize (Langdon
and Coltheart, 2001), and to form complex ideas of one’s own life
as a narrative spanning a lifetime (Berna et al., 2011). Recently,
Lysaker et al. (2010) provided substantial evidence that schizo-
phrenia patients struggle not only with how to cope with symptoms
and neurocognitive impairment, but also struggle to interpret and
make sense of the challenges that they face. This could profoundly
impact on variables such as IM. Although distinct but related, we
assert that metacognitive capacities also regulate IM among
schizophrenia patients.
The present study therefore sought to explore the influence of
metacognition and IM on LP in schizophrenia. Specifically, we
hypothesized that individual differences in metacognition and IM
would predict a significant amount of the variance to explain LP. To
minimize the confounding effects of symptomatology, we decided
to study this association in a sample of symptomatically remitted
patients with schizophrenia.
2. Methods
2.1. Participants
Fifty-two clinically stable individuals with schizophrenia were
recruited from the Psychosis Unit at Celal Bayar University. The
patients met DSM-IV criteria for schizophrenia as determined by
medical records and confirmed with the Structured Clinical Inter-
view for DSM-IV e Patient Edition (SCID; First et al., 1996). Only
patients meeting the criteria of being in remission according to the
Schizophrenia Working Group (Andreasen et al., 2005) were
included. Of the thirty-two patients meeting the criteria for
remission two patients refused participation. Thus, data from thirty
patients (53 percent men) were secured. The patients’ mean age
was 32.93  12.35 years; the number of years of education was
11.50  2.34, the mean duration of illness was 12.07  10.55 years,
and the number of hospital stays was 2  2. All patients received
second-generation antipsychotic medication. Chlorpromazine
equivalent dosages (CPZ) were calculated for all patients in order to
control for the medication effects (Rijcken et al., 2003). Accordingly,
mean dosages were 416.25  220.99 mg per day. All patients
provided written informed consent. The study was approved by the
local Institutional Review Board.
2.2. Clinical and neuropsychological measures
2.2.1. Clinical rating and remission criteria
The Positive and Negative Syndrome Scale (PANSS; Kay et al.,
1987) is a 30-item rating scale that was administered by clinically
trained research staff using a chart review and a semi-structured
interview. Accordingly, the average total PANSS score of the group
was 57.25  11.46 points. Symptomatic remission was assumed if the
following items of the Positive and Negative Syndrome Scale (PANSS;
Kay et al., 1987) were rated as moderate, mild, minimal or absent (i.e.
PANSS score for each item  3): delusions (PANSS-P1), unusual
thought content (PANSS-G9), hallucinations (PANSS-P3), conceptual
disorganization (PANSS-P2), mannerisms (PANSS-G5), blunted affect
(PANSS-N1), social withdrawal (PANSS-N4), and lack of spontaneity
(PANSS-N5). PANSS ratings were completed by a trained consultant
psychiatrist working in the outpatient department.
2.2.2. Estimated general intelligence
Age-corrected indices of the memory quotient (MQ) were
calculated over total WMS-WMS-III (Wechsler, 1997). The MQ is
1088 C. Tas et al. / Journal of Psychiatric Research 46 (2012) 1086e1092
highly correlated with standard measures of general intelligence
(Tulsky and Price, 2003) and was used for reasons of practicability
and time constraints of the experimental procedure.
2.3. Metacognition assessment
2.3.1. The Indiana Psychiatric Illness Interview (IPII)
The IPII (Lysaker et al., 2002) is a semi-structured interview
developed to assess how individuals understand their experience
with mental illness. A trained psychiatrist conducted the interview,
which typically lasted between 30 and 60 min. All original proce-
dures from the IPII were strictly followed, after the interview was
translated into Turkish (IPII; Lysaker et al., 2002). The interview is
conceptually divided into five sections. First, rapport is established
and participants are asked to tell the story of their lives, beginning
with their earliest memory. Second, participants are asked if they
think they have a mental illness and, if so, whether or not this
condition has affected different facets of their life. Third, partici-
pants are asked if and how their condition controls their life and,
alternately, how they control their condition. Fourth, they are asked
how their condition affects, and is affected by others. Finally,
participants are asked about their expectations for the future. The
IPII differs from other psychiatric interviews in that only minimal
content is introduced to the participant to comment on and thus
results in a self-report that can be analyzed in terms of the meta-
cognitive capacities that appear spontaneously.
2.3.2. The Metacognition Assessment Scale
The Metacognition Assessment Scale-Abbreviated (MAS-A) is
a rating scale designed to systematically identify the extent to
which persons can engage in increasingly complex and integrated
metacognitive abilities. The original MAS was developed to detect
growth within psychotherapy transcripts (Semerari et al., 2003)
and, in consultation with the authors, abbreviated and adapted for
the study of IPII transcripts (Lysaker et al., 2005). The MAS-A differs
from other more structured assessments of metacognition in that it
focuses on metacognitive functions that arise spontaneously rather
than cued, as in a task or referenced in a questionnaire. The MAS-A
contains four scales: “Understanding of one’s own mind”, or the
comprehension of one’s own mental states; “understanding of
others’ minds”, or the comprehension of other individuals’ mental
states; “decentration” describes the ability to see the world as
existing with others having independent motives; and “mastery”,
defined as the ability to work through one’s representations and
mental states, with a view to implement effective action strategies
in order to accomplish cognitive tasks or cope with problematic
mental states. The MAS-A asks the rater to indicate whether the
participant has successfully used or failed to use a function for each
task. For example, the rater must determine if the participant can
identify different emotions they feel and recognize that their
understanding of life events is subjective. The full presence of
a function is awarded a score of “1”, whereas a score of “0.5”is
awarded for the partial presence of a function. The highest score
obtainable for “understanding of one’s own mind” is “9”, for
“understanding of others’ minds”, a “7”, for “decentration” a “3”
and for “mastery” a “9”. For all scales higher scores indicates
a greater capacity to create or use increasingly more complex and
integrated representations of oneself or others. For example,
someone with a rating of “1” on mastery would not be able to
plausibly define a psychological problem, someone with a rating of
“4” on mastery could define a psychological plausible problem but
at most cope with it in an actively avoidant manner, while someone
with a rating of “6” on mastery could define a psychological plau-
sible problem and cope with it by altering their thought processes.
The evaluations were carried out by a psychiatrist who received
formal MAS training before the experiment.
2.4. Motivation assessment
2.4.1. Intrinsic motivation
Self-reported intrinsic motivation (IM) was measured using an
adapted version of the Intrinsic Motivation Inventory (IMI; Plant
and Ryan, 1985) for schizophrenia patients (Choi et al., 2010). The
IMI-SC consists of 21 items rated on a 7-point Likert scale with
responses ranging from “not at all true” to “very true”. It is designed
to assess a schizophrenia patients’ subjective experience of an
activity specifically in an experimental setting. The instrument has
five subscales: (1) perceived interest and enjoyment due to task
(interest/enjoyment), (2) perceived value and usefulness of the task
for the proposed test (value/usefulness), (3) efforts given to
successfully complete the task (effort), (4) perceived pressure and
tension while performing the task (pressure/tension), and (5)
perceived autonomy to perform the task (choice). Higher scores
indicate a greater IM for performing the task. The scale is highly
associated with constructs of motivation for health-related
behaviors, including perceived competency for attempting chal-
lenging tasks and autonomous treatment engagement (Choi et al.,
2010). In the present study, the IMI-SC was conducted after the
cognitive remediation training of WCST skills.
2.5. Learning potential (LP) assessment
The WCST is a neuropsychological test sensitive to impairments
in executive function. Participants sort cards that vary according to
an unspecified matching rule that changes periodically. The current
study utilized the number of correct responses meaning the right
matching rule between color, shape and number of the card pre-
sented as a goal of learning.
Participants received a 128-card computer version of the Wis-
consin Card Sorting Task (WCST; Heaton, 1981) with standard
instructions at baseline. Eight weeks later, a previously defined
structured training procedure, based on cognitive remediation
using errorless learning techniques, was administered as training
with computer (for details, see Kern et al., 1996). In summary, the
training procedures for the WCST used the techniques of errorless
learning. The WCST was chunked into its small components to
create a hierarchy of to-be-learned skills. The training was held in
three steps: stimulus feature identification, card matching, single
shift execution, double shift execution. A subject who repeatedly
gave the correct answer for 10 trials in each phase was considered
to have successfully completed the task. None of the subjects failed
to complete the training. Each phase of training consisted of a series
of trials that included self- and other-delivered cues, as well as
modeling, explicit instruction, and response feedback to facilitate
correct responding and to reduce the possibility of errors. Training
was conducted by a psychiatrist who received certified training in
applying cognitive remediation (C.T.). Training typically lasted
30e45 min. The training material was prepared using PowerPoint
for Windows and a freely available psychological experiment tool
(PsychoPy 1.71, Nottingham University, UK). After the training
period, within the same session, post-training data was collected
using the exact same procedure used at baseline.
2.6. Data analysis
Data was evaluated for normality and in no case was there
evidence that variables included in the study violated the
assumptions underlying the use of parametric statistical proce-
dures. Pearson correlations were calculated between estimated IQ,
 C. Tas et al. / Journal of Psychiatric Research 46 (2012) 1086e1092 1089

metacognition and motivation for exploring the relationship
between variables and identifying multicollinearity among inde-
pendent variables.
To define LP, we firstly conducted a categorical approach, in
which patients were categorized as “non-learners”, “learners” and
“high achievers” according to a previously established algorithm
(for details, see Schottke et al., 1993; Wiedl, 1999). For every pretest
score, a test score on a hypothetical parallel test was predicted in
a linear regression model and compared with the real post-test
score. We categorized a non-learner as one whose post-test score
was below the predicted confidence interval. Participants whose
scores were 1.5 standard deviations above the group mean gain
difference were classified as “high achievers”, whereas the
remaining group of patients (who scored in-between) was
considered as “learners”. Furthermore, to control for ceiling effects
participants whose pre-training number of correct responses was
above 86 were also classified as high achievers (Wiedl, 1999;
Vaskinn et al., 2009), a procedure that has been used in a similar
fashion in previous studies of LP (Fiszdon et al., 2006; Rempfer
et al., 2006). A between group ANOVA was conducted to identify
the differences in IQ, motivation and metacognition between
learning group categories.
A hierarchical regression model was performed to identify the
unique contribution of each variable in predicting the WCST
number of correct responses (dependent variable) after training. A
hypothetical model was constructed according to the ongoing
literature on motivation and metacognition and learning (Choi
et al., 2010; Lysaker et al., 2008). To control for the effects of
pretest performance and intelligence, we included the baseline
number of correct responses in the WCST in the first step of the
regression analysis, and added the estimated intelligence (MQ)
score in the second step. The aim of this regression analysis was to
observe the unique influence of intrinsic motivation and meta-
cognition on learning from training (i.e. LP) when controlling for
differences in baseline performance and intelligence.
3. Results
3.1. Data quality
Preliminary analyses were conducted to ensure that there was no
violation of the assumptions of normality, linearity and homoge-
neity of variances, homogeneity of regression slopes and reliability.
3.2. Correlation analyses
The correlations between intelligence, subdomains of IMI, and
metacognition (MAS-A scores) are summarized in Table 1.
Accordingly, interest and enjoyment was correlated only with
perceived usefulness and subjective value of the training, but not
with perceived effort, autonomy and pressure to complete the
training. All subdomains of metacognition were highly correlated
with the subcategories of IM, i.e. perceived value, usefulness and
effort to complete training, except interest and enjoyment.
3.3. Between group ANOVA
Table 2 summarizes the between group ANOVA results comparing
estimated intelligence, intrinsic motivation and metacognition
between non-learners, learners and high achievers. No differences
between learning categories were found in terms of medication (CPZ
equivalents (F(2,27 ¼ 1.701, p ¼ .202). Significant group differences
were found between high achievers and non-learners on all factors. In
addition to this, there were significant group differences between
learners and non-learners for the memory quotient, the perceived
usefulness/value and effort subscales of the IMI, understanding
others mind, decentration and mastery levels of metacognition.
Finally, differences in self-reflectivity between non-learners and
learners approached statistical significance (p ¼ .061).
3.4. Multiple regression analysis
Before the hierarchical regression model was performed, to
ensure no violation of multicollinearity, we decided to include the
total IMI score instead of subscale scores and mastery levels of
metacognition into the equation. Variance inflation factors were
checked for before the analyses. The results of regression model
including the pre-training WCST number of correct responses (Step
1), MQ (Step 2), total IMI (Step 3) and metacognitive mastery (Step
4) are presented in Table 3. In the last step we included meta-
cognitive mastery, firstly because metacognitive mastery has
specifically been found as a potential mediator social functioning in
previous studies (Lysaker et al., 2008), and secondly due to the
models of development of intrinsic motivation and learning (Ryan
and Deci, 2000; Flavell, 1979). Accordingly, pre-training perfor-
mance was entered in the first step, and significantly explained
22.5% of the variance. Including memory quotient scores in the
model showed that it explained a further 6.5% of the variance, with
intrinsic motivation accounting for another 8.5%, though non-
significantly. Metacognitive mastery was found as the only vari-
able that significantly explained 48% of the overall variance in the
last step. Previous performance and metacognitive mastery
uniquely predicted the post-training performance when variance
from other variables was controlled for. The memory quotient and
total intrinsic motivation failed to explain overall variance inde-
pendently in any of the steps, although they seem to have
a contributing role in the equation.

Table 1
Inter correlations among the variables.

1 2 3 4 5 6 7 8 9 10
1. MQ e
2. IMI-SC Interest/Enjoyment .21 e
3. IMI-SC Value/Usefulness .34 .78** e
4. IMI-SC Effort .03 .32 .38* e
5. IMI-SC Pressure/Tension .29 .07 .11 .07 e
6. IMI-SC Choice .13 .27 .19 .38* .03 e
7. MC-Self reflectivity .27 .29 .37* .32 .03 .14 e
8. MC-Understanding other minds .25 .26 .39* .49** .05 .33 .85** e
9. MC-Decentration .31 .29 .46* .45* .10 .10 .86** .74** e
10. MC-Mastery .50** .28 .46* .23 .27 .03 .80** .73** .79** e

Note: MQ: Memory quotient, Correlation values in italic type remained significant after multiple correlational analyses. IMI-SC: Intrinsic motivation inventory for Schizo-
phrenia MC: Metacognition.
*P < .05.
**P < .01.
1090 C. Tas et al. / Journal of Psychiatric Research 46 (2012) 1086e1092

Table 2
ANOVA between group analysis of learning categories.

Non-learners (n:6) learners (n:14) high achievers (n:10) F Sig.

Mean S.D. Mean S.D. Mean S.D.

MQ 71.83 15.03 90.00 15.11 89.40 11.20 3.98 .030
IMI-SC Interest/Enjoyment 32.67 9.00 36.93 7.00 42.50 6.04 3.83 .034
IMI-SC Value/Usefulness 30.17 10.89 41.00 9.07 42.70 5.27 4.63 .019
IMI-SC Effort 18.33 2.50 25.57 5.47 22.60 3.37 5.79 .008
IMI-SC Pressure/Tension 20.83 5.34 19.07 7.05 19.40 3.34 .20 .818
IMI-SC Choice 24.00 8.72 32.64 7.61 27.70 11.29 2.07 .145
MC-Self-reflectivity 3.17 1.03 5.32 2.24 6.25 1.57 5.24 .012
MC-Understanding other minds 2.92 .38 4.43 1.14 4.65 1.27 5.30 .011
MC-Decentration .50 .55 1.57 .92 1.85 .67 5.93 .007
MC-Mastery 3.50 1.05 5.68 1.91 6.50 1.45 6.47 .005

Note: MQ: Memory quotient, IMI-SC: Intrinsic Motivation Inventory for Schizophrenia. MC: Metacognition.

4. Discussion (Artaloytia et al., 2006). Nevertheless, a recent study found that IM

The present study sought to examine the relationship between
intrinsic motivation and metacognition, and its impact on learning
potential in remitted patients with schizophrenia. In line with
predictions, we found that intrinsic motivation was highly associ-
ated with metacognitive abilities. LP correlated positively with
metacognition, motivation and intelligence. Specifically, a regres-
sion analysis revealed that metacognitive mastery was the best
predictor of LP, whereas motivation and intelligence did not inde-
pendently contribute to explaining the variance.
Metacognitive deficits in schizophrenia have recently been
related to a large number of diverse processes that can malfunction
in a number of different ways (Lysaker et al., 2011a). Previous
studies reported the mediating role of mastery on neurocognition
and social functioning (Lysaker et al., 2008, 2011a). In addition,
metacognitive capacities were found as predictors of therapeutic
alliance (Davis et al., 2011). According to our results, metacognition
has a substantial influence on learning potential.
Self-determination theory (SDT), as mentioned earlier, proposes
a continuous pathway where the final step to stimulate intrinsic
motivation is acquired by regulating one’s own self independent
from extrinsic motivators (Ryan and Deci, 2000). In other words,
people who develop superior intrinsic motivation through self-
regulation will be more engaged in their social lives and therefore
be more willing to participate in cognitive and social tasks.
In clinical settings, attaining this most mature level of IM seems
largely unlikely for patients with schizophrenia, especially in light
of evidence that antipsychotic drug treatment further deteriorates
motivation, induces anhedonia and impairs reward processing
Table 3
Hierarchical regression analysis for predicting learning potential.
B t value
Step 1 Baseline WCST number of correct responses .70
Step 2 Baseline WCST number of correct responses .66
MQ .47
Step 3 Baseline WCST number of correct responses .66
MQ .33
IMI-SC Total score .31
Step 4 Baseline WCST number of correct responses .50
MQ .07
IMI-SC Total score .19
Mastery e Metacognition 5.31
Step 1: R2 ¼ .225, df ¼ 1, 28, p ¼ .008.
Step 2: R2 ¼ .290, df ¼ 1, 27, p ¼ .010; Sig. F Change: p ¼ .079.
Step 3: R2 ¼ .375, df ¼ 1, 26, p ¼ .006; Sig. F Change: p ¼ .088.
Step 4: R2 ¼ .479, df ¼ 1, 25, p ¼ .002; Sig. F Change: p ¼ .042.
Note: Dependent variable: Post training WCST number of correct responses. MQ:
Memory quotient. IMI-SC Total score: Intrinsic Motivation Inventory for Schizo-
phrenia Total score.
in schizophrenia is a dynamic process that changes over time
(Nakagami et al., 2010), and can actually be fostered by attempts to
improve perceived interest and enjoyment. According to our find-
ings, better metacognitive skills can subsequently improve learning
potential. Therefore, perceived interest and enjoyment can improve
IM, though this is subjective and independent of metacognition,
whereas perceived usefulness and value also improves IM, but
relies on metacognitive skills, particularly mastery. Our assumption
is in line with SDT, suggesting that when extrinsic motivation
becomes self-integrated, it can turn into an intrinsic motivator, and
this can be achieved by enhancing the personal relevance of a task
through conscious evaluation, i.e. metacognition.
In addition, with regards SDT, our findings suggest that patients
with poor motivation, low levels of metacognitive capacity, and
consequently poor LP, may be the ones who rely more on extrinsic
incentives, which is the diametrical opposite of IM (Ryan and Deci,
2000). Several studies of cognitive remediation in schizophrenia
have used monetary incentives to reduce dropout rates (e.g., Horan
et al., 2011). This may indirectly impact upon the efficacy of the
intervention, firstly because including patients who consider
extrinsic incentives as important can create a bias in the sampling
of patients toward a group of patients with poorer LP. Secondly,
extrinsic incentives may also mask the intrinsic value and subjec-
tively perceived usefulness of the training. This may further explain
why such studies report smaller effect sizes of the intervention
compared to community care-based studies (Brekke et al., 2009).
Furthermore, Koren et al. (2006) found that metacognitive judg-
ment plays a fundamental role in capacity to consent.
Wykes and Reeder (2005) have argued that the transfer from
a training task to real-life performance critically depends on
improving motivation and metacognition in patients with schizo-
phrenia. This may either be achieved by explicitly teaching meta-
Sig.
cognitive skills to promote active monitoring of one’s own
performance, or by implementing educational techniques such as
2.85 .012
2.70 .010 scaffolding and errorless learning, which promotes success and
1.57 .127 enjoyment. Moreover, Lysaker et al. (2011b) have supported this
2.83 .009 assumption, suggesting that metacognitive capacities can mean-
1.09 .288 ingfully be addressed in psychotherapy by cuing interventions
1.87 .072
2.16 .040
according to patient’s current maximal capacity for metacognitive
.24 .816 activity.
1.20 .243 Our results suggest considerable variability in LP of patients
2.23 .034 with schizophrenia, which arguably interacts with patients’ moti-
vation and metacognition. Consequently, this interdependence of
LP with motivation and metacognition may have substantial
implications for the efficacy of cognitive remediation training (CRT)
in schizophrenia. Raffard et al. (2009) have highlighted that LP can
be used as a tool in predicting treatment outcome after CRT. In sum,
the benefit of CRT will be less for patients with poorer LP. Therefore
 C. Tas et al. / Journal of Psychiatric Research 46 (2012) 1086e1092
a pre-treatment assessment of LP, in consideration of IM and
metacognition, could be used as a screening tool to select patients
who are more likely to benefit from CRT. Conversely, in previous
studies the inclusion of patients who lack LP, motivation and
metacognition may also explain the low reported effect sizes of CRT
(e.g., Wykes et al., 2011). For those patients, it is perhaps advisable
to tailor psychoeducational approaches, which in the first place
enhance IM as a window to improve LP in the second line.
This study has several limitations. Firstly, we did not take into
account the impact of other neurocognitive confounders like sus-
tained attention, which may have an influence on LP. Secondly, our
assessment of estimated IQ with the MQ from the WMS-III is less
common than other methods of estimated IQ evaluation. Finally, to
further elucidate the relationship between motivation and meta-
cognition it may be more reliable to perform a structured inter-
vention with multiple sessions, as compared to the experimental
design used here.
In conclusion, to our knowledge this is the first study demon-
strating the direct role of IM and metacognitive capacity on
learning potential in patients with schizophrenia. We present
a novel perspective on approaching predictors of learning, sug-
gesting that metacognitive capacity and motivation play crucial
role, but semi-independent roles in successfully accomplishing
cognitive remediation in schizophrenia. Future studies may address
the question if these factors are also involved in mediating the
effects of cognitive remediation on social functioning.
Role of funding sources
Drs Tas and Brown have been funded by a scholarship from
International Graduate School of Neuroscience, Ruhr University of
Bochum.
Contributors
Drs Tas and Brüne designed the study. Dr. Tas and Esen-Danaci
collected data for this study. Dr Tas conducted statistical analyses,
and wrote the first draft of the manuscript. Drs Tas, Brown, Lysaker
and Brüne contributed to the writing for the subsequent drafts of
manuscript. All authors have approved the final manuscript.
Conflicts of interest
None.
Acknowledgments
Drs Tas and Brüne would like to thank Zeynep Cubukcuoglu, MD
and Orkun Aydin MD, for their assistance in data collection.
References
Andreasen NC, Carpenter Jr WT, Kane JM, Lasser RA, Marder SR, Weinberger DR.
Remission in schizophrenia: proposed criteria and rationale for consensus.
American Journal of Psychiatry 2005;162:441e9.
Artaloytia JF, Arango C, Lahti A, Sanz J, Pascual A, Cubero P, et al. Negative signs and
symptoms secondary to antipsychotics: a double-blind, randomized trial of
a single dose of placebo, haloperidol, and risperidone in healthy volunteers.
American Journal of Psychiatry 2006;163:488e93.
Barch DM. The relationships among cognition, motivation, and emotion in
schizophrenia: how much and how little we know. Schizophrenia Bulletin
2005;31:875e81.
Berna F, Bennouna-Greene M, Potheegadoo J, Verry P, Conway MA, Danion JM.
Impaired ability to give a meaning to personally significant events in patients
with schizophrenia. Consciousness and Cognition 2011;20:703e11.
Brekke JS, Hoe M, Green MF. Neurocognitive change, functional change and service
intensity during community-based psychosocial rehabilitation for schizo-
phrenia. Psychological Medicine 2009;39:1637e47.
1091
Brenner HD, Hodel B, Genner R, Roder V, Corrigan PW. Biological and cognitive
vulnerability factors in schizophrenia: implications for treatment. British Jour-
nal of Psychiatry 1992;(Suppl. 18):154e63.
Brüne M, Dimaggio G, Lysaker HP. Metacognition and social functioning in
schizophrenia: evidence, mechanisms of influence and treatment implications.
Current Psychiatry Reviews 2011;7:239e47.
Choi J, Medalia A. Intrinsic motivation and learning in a schizophrenia spectrum
sample. Schizophrenia Research 2010;118:12e9.
Choi J, Mogami T, Medalia A. Intrinsic motivation inventory: an adapted measure for
schizophrenia research. Schizophrenia Bulletin 2010;36:966e76.
Davis LW, Eicher AC, Lysaker PH. Metacognition as a predictor of therapeutic alli-
ance over 26 weeks of psychotherapy in schizophrenia. Schizophrenia Research
2011;129:85e90.
Deci EL, Ryan RM. Intrinsic motivation and self-determination in human behavior.
New York: Plenum; 1985.
Delis DC, Kramer JH, Kaplan E, Ober BA. The California verbal learning test. 2nd ed.
San Antonio: The Psychological Corporation; 2000.
First MB, Gibbon M, Spitzer RL. User’s guide for the structured interview for DSM-IV
Axis I Disorders e Research Version (SCID-I, version 2.0). New York: Biometrics
Research; 1996.
Fiszdon JM, McClough JF, Silverstein SM, Bell MD, Jaramillo JR, Smith TE. Learning
potential as a predictor of readiness for psychosocial rehabilitation in schizo-
phrenia. Psychiatry Research 2006;143:159e66.
Flavell JH. Metacognition and cognitive monitoring: a new area of cognitive
developmental inquiry. American Psychologist 1979;34:906e11.
Heaton R. Wisconsin card sorting test manual. Florida: Psychological Assessment
Resources; 1981.
Heinrichs DW, Hanlon TE, Carpenter Jr WT. The Quality of Life Scale: an instrument
for rating the schizophrenic deficit syndrome. Schizophrenia Bulletin 1984;10:
388e98.
Hofmann SG, Smits JA. Cognitive-behavioral therapy for adult anxiety disorders:
a meta-analysis of randomized placebo-controlled trials. Journal of Clinical
Psychiatry 2008;69:621e32.
Horan WP, Kern RS, Tripp C, Hellemann G, Wynn JK, Bell M, et al. Efficacy and
specificity of social cognitive skills training for outpatients with psychotic
disorders. Journal of Psychiatric Research 2011;45:1113e22.
Kay S, Fiszbein A, Opler LA. The positive and negative syndrome scale (PANSS) for
schizophrenia. Schizophrenia Bulletin 1987;13:261e76.
Kern RS, Wallace CJ, Hellman SG, Womack LM, Green MF. A training procedure for
remediating WCST deficits in chronic psychotic patients: an adaptation of
errorless learning principles. Journal of Psychiatry Research 1996;30:283e94.
Kleinginna PR, Kleinginna AM. A categorized list of emotion definitions, with
suggestions for a consensual definition. Motivation and Emotion 1981;5:345e79.
Koren D, Seidman LJ, Goldsmith M, Harvey PD. Real-world cognitiveeand meta-
cognitiveedysfunction in schizophrenia: a new approach for measuring (and
remediating) more “right stuff”. Schizophrenia Bulletin 2006;32:310e26.
Kurtz MM, Jeffrey SB, Rose J. Elementary neurocognitive function, learning potential
and everyday life skills in schizophrenia: what is their relationship? Schizo-
phrenia Research 2010;116:280e8.
Kurtz MM, Seltzer JC, Fujimoto M, Shagan DS, Wexler BE. Predictors of change in life
skills in schizophrenia after cognitive remediation. Schizophrenia Research
2009;107:267e74.
Langdon R, Coltheart M. Visual perspective-taking and schizotypy: evidence for
a simulation-based account of mentalizing in normal adults. Cognition 2001;
82:1e26.
Lauchlan F, Elliott J. The psychological assessment of learning potential. British
Journal of Educational Psychology 2001;71:647e65.
Lysaker PH, Carcione A, Dimaggio G, Johannesen JK, Nicolo G, Procacci M, et al.
Metacognition amidst narratives of self and illness in schizophrenia: associa-
tions with neurocognition, symptoms, insight and quality of life. Acta Psy-
chiatrica Scandinavica 2005;112:64e71.
Lysaker PH, Clements CA, Plascak-Hallberg CD, Knipscheer SJ, Wright DE. Insight and
personal narratives of illness in schizophrenia. Psychiatry 2002;65:197e206.
Lysaker PH, Tsai J, Maulucci AM, Stanghellini G. Narrative accounts of illness in
schizophrenia: association of different forms of awareness with neurocognition
and social function over time. Consciousness and Cognition 2008;17:1143e51.
Lysaker PH, Shea AM, Buck KD, Dimaggio G, Nicolo G, Procacci M, et al. Meta-
cognition as a mediator of the effects of impairments in neurocognition on
social function in schizophrenia spectrum disorders. Acta Psychiatrica Scandi-
navica 2010;122:405e13.
Lysaker PH, Erickson M, Ringer J, Buck KD, Semerari A, Carcione A, et al. Meta-
cognition in schizophrenia: the relationship of mastery to coping, insight, self-
esteem, social anxiety, and various facets of neurocognition. British Journal of
Clinical Psychology 2011a;50:412e24.
Lysaker PH, Buck KD, Carcione A, Procacci M, Salvatore G, Nicolo G, et al. Addressing
metacognitive capacity for self reflection in the psychotherapy for schizo-
phrenia: a conceptual model of the key tasks and processes. Psychology and
Psychotherapy 2011b;84:58e69.
McCabe J. Metacognitive awareness of learning strategies in undergraduates.
Memory & Cognition 2011;39:462e76.
Medalia A, Thysen J. Insight into neurocognitive dysfunction in schizophrenia.
Schizophrenia Bulletin 2008;34:1221e30.
Nakagami E, Hoe M, Brekke JS. The prospective relationships among intrinsic
motivation, neurocognition, and psychosocial functioning in schizophrenia.
Schizophrenia Bulletin 2010;36:935e48.
1092 C. Tas et al. / Journal of Psychiatric Research 46 (2012) 1086e1092
Nakagami E, Xie B, Hoe M, Brekke JS. Intrinsic motivation, neurocognition and
psychosocial functioning in schizophrenia: testing mediator and moderator
effects. Schizophrenia Research 2008;105:95e104.
Park KM, Ku J, Choi SH, Jang HJ, Park JY, Kim SI, et al. A virtual reality application in
role-plays of social skills training for schizophrenia: a randomized, controlled
trial. Psychiatry Research 2011;189:166e72.
Plant RW, Ryan RM. Intrinsic motivation and the effects of selfconsciousness, self-
awareness, and ego-involvement: an investigation of internally-controlling
styles. Journal of Personality 1985;53:435e49.
Raffard S, Gely-Nargeot MC, Capdevielle D, Bayard S, Boulenger JP. Learning potential
and cognitive remediation in schizophrenia. L’Encéphale 2009;35:353e60.
Rempfer M, Hamera E, Brown C, Bothwell RJ. Learning proficiency on the Wisconsin
Card Sorting Test in people with serious mental illness: what are the cognitive
characteristics of good learners? Schizophrenia Research 2006;87:316e22.
Rempfer M, Brown C, Hamera E. Learning potential as a predictor of skill acquisition
in people with serious mental illness. Psychiatry Research 2011;185:293e5.
Rijcken CA, Monster TB, Brouwers JR, van den Berg LT. Chlorpromazine equivalents
versus defined daily doses: how to compare antipsychotic drug doses? Journal
of Clinical Psychopharmacology 2003;23:657e9.
Rohling ML, Faust ME, Beverly B, Demakis G. Effectiveness of cognitive rehabilitation
following acquired brain injury: a meta-analytic re-examination of Cicerone
et al.’s (2000, 2005) systematic reviews. Neuropsychology 2009;23:20e39.
Roncone R, Falloon IR, Mazza M, De Risio A, Pollice R, Necozione S, et al. Is theory of
mind in schizophrenia more strongly associated with clinical and social func-
tioning than with neurocognitive deficits? Psychopathology 2002;35:280e8.
Ryan RM, Deci EL. Intrinsic and extrinsic motivations: classic definitions and new
directions. Contemporary Educational Psychology 2000;25:54e67.
Schottke HH, Bartram M, Wiedl KH. Psychometric implications of learning potential
assessment: a typological approach in learning potential assessment: theoret-
ical, methodological, and practical issues. Netherlands: Swets and Zeitlinger
Publishers; 1993.
Semerari A, Carcione A, Dimaggio G, Falcone M, Nicolo G, Procacci M, et al.
Assessing problematic states in patients’ narratives: the grid of problematic
states. Psychotherapy Research 2003;13:337e53.
Sergi MJ, Kern RS, Mintz J, Green MF. Learning potential and the prediction of work
skill acquisition in schizophrenia. Schizophrenia Bulletin 2005;31:67e72.
Sternberg RJ, Grigorenko EL. Dynamic testing: the nature and measurement of
learning potential. Cambridge, UK: Cambridge: University Press; 2002.
Svensson B, Hansson L. Therapeutic alliance in cognitive therapy for schizophrenic
and other long-term mentally ill patients: development and relationship to
outcome in an in-patient treatment programme. Acta Psychiatrica Scandinavica
1999;99:281e7.
Swartz MS, Perkins DO, Stroup TS, Davis SM, Capuano G, Rosenheck RA, et al. Effects
of antipsychotic medications on psychosocial functioning in patients with
chronic schizophrenia: findings from the NIMH CATIE study. American Journal
of Psychiatry 2007;164:428e36.
Tulsky DS, Price LR. The joint WAIS-III and WMS-III factor structure: development
and cross-validation of a six-factor model of cognitive functioning. Psycholog-
ical Assessment 2003;15:149e62.
Vaskinn A, Sundet K, Friis S, Ueland T, Simonsen C, Birkenaes AB, et al. Using the
Wisconsin Card Sorting Test to assess learning potential in normal IQ schizo-
phrenia: does it have potential? Nordic Journal of Psychiatry 2009;14:1e7.
Watzke S, Brieger P, Wiedl KH. Prediction of vocational rehabilitation outcome in
schizophrenia: incremental prognostic validity of learning potential beyond
basic cognitive performance. Journal of Cognitive Education and Psychology
2009;8:52e62.
Wechsler D. Wechsler memory scale. 3rd ed. San Antonio, TX: The Psychological
Corporation; 1997.
Wiedl KH. Rehab rounds: cognitive modifiability as a measure of readiness for
rehabilitation. Psychiatric Services 1999;50:1411e9.
Wykes T, Reeder C. Cognitive remediation therapy for schizophrenia: theory and
practice. New York: Routledge; 2005.
Wykes T, Steel C, Everitt B, Tarrier N. Cognitive behavior therapy for schizophrenia:
effect sizes, clinical models, and methodological rigor. Schizophrenia Bulletin
2008;34:523e37.
Wykes T, Huddy V, Cellard C, McGurk SR, Czobor P. A meta-analysis of cognitive
remediation for schizophrenia: methodology and effect sizes. American Journal
of Psychiatry 2011;168:472e85.