The relationship between cognitive insight and cognitive performance among individuals with at-risk mental state for developing psychosis

SCHRES-07271; No of Pages 6 Schizophrenia Research xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Schizophrenia Research journal homepage: www.elsevier.com/locate/schres

The relationship between cognitive insight and cognitive performance among individuals with at-risk mental state for developing psychosis Noriyuki Ohmuro a,⁎, Masahiro Katsura a, Chika Obara a, Tatsuo Kikuchi b, Yumiko Hamaie a, Atsushi Sakuma a, Kunio Iizuka a, Fumiaki Ito a, Hiroo Matsuoka a,b,c, Kazunori Matsumoto a,b,c a b c

Department of Psychiatry, Tohoku University Hospital, 1-1 Seiryo-machi, Aoba-ku, Sendai, Miyagi, Japan Department of Psychiatry, Tohoku University Graduate School of Medicine, 2-1 Seiryo-machi, Aoba-ku, Sendai, Miyagi, Japan Department of Preventive Psychiatry, Tohoku University Graduate School of Medicine, 2-1 Seiryo-machi, Aoba-ku, Sendai, Miyagi, Japan

a r t i c l e

i n f o

Article history: Received 31 January 2017 Received in revised form 13 April 2017 Accepted 16 April 2017 Available online xxxx Keywords: Cognitive insight Cognitive function Wisconsin card sorting test Brief assessment of cognition in schizophrenia At-risk mental state Ultra-high risk

a b s t r a c t Impairments in cognitive insight—the capacity to appraise and modify one's own distorted beliefs—are believed to be associated with the formation of psychosis. Nevertheless, the association between cognitive insight and cognitive function among people with at-risk mental state (ARMS) for developing psychotic illness has not been made clear. In this study, we used the Beck Cognitive Insight Scale (BCIS) to assess cognitive insight and the Brief Assessment of Cognition in Schizophrenia (BACS) and the Wisconsin Card Sorting Test (WCST) to assess cognitive functions. Fifty subjects with ARMS and 29 healthy volunteers were recruited as participants. The scores for the two groups on the BCIS, BACS, and WCST were compared and Spearman's rank correlations between the domains of the BCIS and cognitive performance were examined in each group. No significant differences were found in BCIS scores between these groups, whereas all of the cognitive function scores were poorer in the participants with ARMS. In the ARMS group, higher self-certainty on the BCIS was significantly correlated with lower performance in the mean number of categories achieved (ρ = −0.31, P = 0.03) and perseverative errors of the Nelson type (ρ = 0.29, P = 0.04) on the WCST. This indicates that excessively high self-certainty might be linked with weaknesses in cognitive flexibility or set-shifting ability in people with ARMS. © 2017 Elsevier B.V. All rights reserved.

1. Introduction For decades, numerous researchers have explored the factors contributing to the onset of psychosis. For example, insight into ones' own illness, that is, “clinical insight,” has been believed to be one of the main factors that influences the differences between individuals with fully psychotic illness and those with subthreshold symptoms (Lappin et al., 2007). In this line, the “cognitive model” is suggested to be a promising model for explaining the cognitive mechanisms relating to the onset of psychosis and maintenance of positive symptoms (Garety et al., 2001) and has been supported by accumulating neurobiological findings (Garety et al., 2007). This model underscores the role of biased processes of reasoning and appraisal of anomalous experiences or thoughts. In the same vein, Beck et al. (2004) proposed the concept of “cognitive insight.” This was defined as the ability in individuals with psychosis to re-evaluate their own anomalous experiences and correct distorted beliefs and misinterpretations regarded as associated

⁎ Corresponding author at: Department of Psychiatry, Tohoku University Hospital, 1-1 Seiryo-machi, Aoba-ku, Sendai, Miyagi 980-8574, Japan. E-mail address: [email protected] (N. Ohmuro).

with the development and maintenance of psychotic symptoms. Given the proposal that the critical determinant of psychosis may be appraisal of anomalous experiences (Garety et al., 2001, Lappin et al., 2007, Morrison and Baker, 2000), cognitive insight should also be one of the critical factors that influences differentiation between psychosis and non-psychosis. To measure an individual's capacity for cognitive insight, Beck et al. (2004) developed the Beck Cognitive Insight Scale (BCIS). The BCIS is a 15-item self-report questionnaire that comprises two principle components—“self-reflectiveness,” which includes 9 items assessing the ability to re-evaluate unusual experiences and correct erroneous judgments, and “self-certainty,” which comprises 6 items representing one's tendency to be overconfident about one's own judgment. Previous studies have corroborated the reliability and validity of the BCIS (e.g., Pedrelli et al., 2004; Uchida et al., 2009) and showed that individuals with psychotic disorders exhibit lower cognitive insight (Engh et al., 2007; Martin et al., 2010; Warman et al., 2007). Furthermore, cognitive insight has been found to be associated with positive and negative symptoms (Bora et al., 2007; Pedrelli et al., 2004; Tranulis et al., 2008); neurocognitive dysfunctions (Nair et al., 2014, for a review); and variations in brain structure, including hippocampal volume (Buchy et al., 2010) and volume of the frontal, parietal, and temporal cortices (Buchy et al., 2016).

http://dx.doi.org/10.1016/j.schres.2017.04.031 0920-9964/© 2017 Elsevier B.V. All rights reserved.

Please cite this article as: Ohmuro, N., et al., The relationship between cognitive insight and cognitive performance among individuals with at-risk mental state for developing ps..., Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.04.031

2

N. Ohmuro et al. / Schizophrenia Research xxx (2017) xxx–xxx

Although cognitive insight has been hypothesized to influence the onset of psychosis and maintenance of positive symptoms, its nature in individuals with a high risk for developing psychosis remains unclear. So far, only two studies have investigated this topic. One study (Uchida et al., 2014) found impairments in cognitive insight—namely, excessively high self-certainty—and a relationship between high self-certainty and attenuated delusional symptoms in individuals with the at-risk mental state (ARMS; Yung et al., 2004). In the other study, however, individuals with ARMS showed intermediate self-certainty scores between healthy controls (HCs) and individuals with schizophrenia, who showed significantly higher self-certainty scores than did HCs (Kimhy et al., 2014). Furthermore, Individuals with ARMS with markedly severe unusual thought content showed significant associations between high self-certainty and high suspiciousness. Thus, excessively high self-certainty might underlie the severity of attenuated psychotic symptoms in ARMS. Recently, the relationship between cognitive insight and cognitive functions has drawn the interest of some researchers. One recent meta-analysis of seven studies in patients with psychotic disorders revealed that a composite index of cognitive insight was positively associated with total cognition, memory, and executive function (Nair et al., 2014). On the other hand, different relationship between sub-components of cognitive insight, that is, self-certainty and self-reflectiveness, and cognitive functions were also observed; higher self-certainty was significantly related to poorer performance on total cognition, IQ, and memory; however, there was no significant correlation between self-reflectiveness and performance in any neurocognitive domain. However, as was noted by the authors of the study, there were several limitations in this meta-analysis; the pooled sample of this meta-analysis was demographically and clinically heterogeneous; categorization of cognitive tests was somewhat crude; relatively recent and limited literature might increase publication bias. In fact, significant correlation between self-certainty and executive function (Cooke et al., 2010; Gilleen et al., 2011; Orfei et al., 2010) and those between self-reflectiveness and working memory (Orfei et al., 2010) and verbal memory (Buchy et al., 2010) were observed in individual studies, even though the meta-analysis failed to find such associations. Therefore, the relationship between cognitive insight and cognitive function does not appear to have been conclusively determined and further study is necessary. To date, however, there has been no study on the relationship between cognitive insight and neurocognitive function among individuals with ARMS, that showed neurocognitive dysfunctions similar to full psychosis, including verbal memory, visual memory, processing speed, general intelligence, and executive function (Fusar-Poli et al., 2012; Bora et al., 2014). Thus, in the current study, we investigated the correlations between cognitive insight and cognitive performance in individuals with ARMS and compared them with correlations in healthy controls. We hypothesized that similar to full psychosis, cognitive functions in subjects with ARMS correlate with their poorer cognitive insight, particularly with self-certainty.

a history of psychotic disorders. Most participants with ARMS in this study were also involved in another study that we had previously published; more details on the recruitment criteria used in the present study are described therein (Ohmuro et al., 2015). The individuals in the HC group were university students recruited through intramural advertisements. A declaration by the participants in a brief interview confirmed that they did not have any history of psychiatric illness. The present study received approval from the local ethics committees. Informed consent was provided by all the participants. 2.2. Measures 2.2.1. Clinical assessments In the ARMS group, psychopathology was evaluated with the CAARMS-J and the Positive and Negative Syndrome Scale (Kay et al., 1987), the severity of subjective depressive symptoms with the Beck Depression Inventory-II (Beck et al., 1996), and Global function with the Global Assessment of Functioning (American Psychiatric Association, 1994). Estimated premorbid IQ was evaluated using the Japanese version of the National Adult Reading Test (JART, Matsuoka et al., 2006; original National Adult Reading Test by Nelson, 1982) for both groups. 2.2.2. Assessment of cognitive insight The BCIS (Beck et al., 2004) is a self-report questionnaire containing 15 items, each rated from 0 (do not agree at all) to 3 (agree completely). Two component scores are calculated from these items: self-reflectiveness and self-certainty. Furthermore, a composite index indicating overall cognitive insight is calculated by subtracting the self-certainty score from that of self-reflectiveness, with higher scores indicating higher cognitive insight. We administered the Japanese version of the BCIS, which has confirmed reliability and validity (Uchida et al., 2009), to both groups. 2.2.3. Assessment of cognitive function 2.2.3.1. The brief assessment of cognition in schizophrenia (BACS). The Brief Assessment of Cognition in Schizophrenia (BACS) (Keefe et al., 2004), Japanese version (Kaneda et al., 2007), was administered by trained psychologists to assess subjects' overall and various specific types of neurocognitive performance. The BACS comprises six subtests: verbal memory, working memory, motor speed, verbal fluency, attention and processing speed, and executive function. Furthermore, a composite zscore was provided based on the results of the HC group, following the method of Keefe et al. (2004) because this method, as compared to the way of using the existing normative data set, enabled us to control for the scores of participants using covariates such as premorbid IQ. For the BACS, higher scores indicate better performance. All subjects were assessed with the BACS within two weeks of being assessed with the BCIS.

2. Methods 2.1. Participants Fifty subjects with ARMS and 29 HCs who were Japanese speakers and aged between 14 and 35 participated in this project. The participants with ARMS were users of the SAFE Clinic, an expert clinical setting for individuals with ARMS (Mizuno et al., 2009; Katsura et al., 2014). The data shown in this article are baseline data collected from individuals with ARMS who provided their informed consent on participation in this project. The Comprehensive Assessment of At-Risk Mental States-Japanese version (CAARMS-J; Miyakoshi et al., 2009) was used to confirm whether the participants fulfilled the ARMS eligibility criteria. The participants had to meet at least one of the criteria for ARMS established by the PACE Clinic (Yung et al., 2004) and not have

2.2.3.2. The Wisconsin card sorting test (WCST). The Wisconsin Card Sorting Test (WCST; Heaton, 1981) has been frequently used as a neuropsychological test for assessing certain types of executive function, in particular set-shifting ability (Greve et al., 2005; Nyhus and Barceló, 2009). We adopted this test to complement the BACS, because executive function as evaluated with the BACS is measured as the ability to mentally manipulate balls into indicated goals; as such, it lacks accurate assessment of deficits in other types of executive function, such as difficulty in set-shifting or a tendency to perseverate. In this study, we administered the modified computerized version of the WCST (i.e., the Keio version: Kashima et al., 1987) to both groups within two weeks of their being assessed with the BCIS. In this test, four stimulus cards illustrating one to four colored-geometric figures are presented in the upper part of the screen, while 48 response cards are shown in order

Please cite this article as: Ohmuro, N., et al., The relationship between cognitive insight and cognitive performance among individuals with at-risk mental state for developing ps..., Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.04.031

N. Ohmuro et al. / Schizophrenia Research xxx (2017) xxx–xxx

in the lower part. Participants are asked to match the presented response card to the four stimulus cards based on the perceived rule of categorization (color, form, or number). The participants were asked to identify the rule for sorting the cards based on the computer's responses of “correct” or “incorrect” to their matches. Once the participants had learned the sorting rule and made six consecutive correct responses, the sorting rule changed without warning. The outcomes were the numbers of categories achieved (CAs) and perseverative errors of the Nelson type (PENs). The number of PENs is the number of incorrect responses based on placing a card in the same category as an adjacent incorrect response, and reflects participants' tendency to perseverate; a higher number indicates a poorer result. The opposite is true for CAs—that is, a higher score indicates a better result. 2.3. Statistical analysis Fisher's exact test was used for comparison of sex ratios in the ARMS and HC groups. t-tests were used to compare the groups on certain demographic characteristics as well as the BCIS and BACS results. Further, for the BCIS and BACS results, analyses of covariance were performed to confirm the differences while controlling for covariates (i.e., age, years of education, estimated IQ, and dose of antipsychotic medication). Mann–Whitney U tests were conducted to determine differences in years of education and the results of the WCST between the groups, as neither variable was expected to show a normal distribution. Spearman's rank correlations were examined for the associations of the two indices (self-reflectiveness and self-certainty) of the BCIS with the composite z-score of the BACS and the two indices (numbers of CAs and PENs) of the WCST in each group. Subsequently, to exploratorily investigate the association of cognitive insight and neurocognitive subdomains, additional Spearman's rank correlational analyses were conducted without correcting for the increased probability of type I error. All of the statistical analyses were performed using SPSS Statistics 17.0 for Windows (SPSS Inc., Chicago, IL). Two-tailed tests were adopted and the significance level was set as 0.05.

3

Table 2 Fulfilled criteria, clinical indices, and medication in the ARMS group. ARMS (n = 50) Fulfilled ultra-high risk criteria Attenuated psychotic symptoms Brief limited intermittent psychotic symptoms State and trait factors Attenuated psychotic symptoms plus state and trait factors Attenuated psychotic symptoms plus brief limited intermittent psychotic symptoms Clinical variables PANSS positive, M (SD) PANSS negative, M (SD) PANSS general, M (SD) BDI-II, M (SD) GAF, M (SD) Medications Antipsychotics, n (%) Atypical antipsychotics, n (%) CP-equivalent dose (mg) (SD) Antidepressants, n (%) Benzodiazepines, n (%) Mood stabilizers, n (%) Anticholinergics, n (%)

39 (78%) 2 (4%) 1 (2%) 7 (14%) 1 (2%)

13.0 (3.5) 12.9 (5.0) 32.8 (7.4) 29.6 (12.7) 48.2 (8.3) 16 (32%) 14 (28%) 65.0 (112.6) 15 (30%) 28 (56%) 5 (10%) 5 (10%)

ARMS: at-risk mental state; PANSS: Positive and Negative Syndrome Scale; BDI-II: Beck Depression Inventory-II; GAF: Global Assessment of Functioning; CP: chlorpromazine.

While positive, negative, and general symptoms were milder compared to patients with psychotic disorders (e.g., Ohmuro et al., 2015), they showed moderate-to-severe depressive symptoms and serious functional impairment. Sixteen participants (32%) in the ARMS group were taking antipsychotic medication; the majority were prescribed atypical antipsychotics. 3.3. Profile of cognitive insight

3. Results

Table 1 summarizes the demographic data. The difference in age between the two samples was marginally significant, and the gender ratio did not show significant difference between two groups. The years of education and estimated IQ of the HC group were higher than those of the ARMS group.

Table 3 contains the BCIS scores. There were no significant differences in the mean scores of self-reflectiveness and self-certainty between the ARMS and HC groups. This did not change even after controlling for age, years of education, dosage of antipsychotics, and premorbid IQ. The composite index of the BCIS was lower in the ARMS group at a trend-level significance. This was significant even after controlling for age (P = 0.02), years of education (P = 0.02), and dosage of antipsychotics (P = 0.02).

3.2. Clinical characteristics

3.4. Cognitive profiles

Table 2 contains the fulfilled ultra-high risk criteria, clinical variables, and information on medication in the ARMS group. All but 3 participants with ARMS fulfilled the criteria of attenuated psychotic symptoms. The participants with ARMS were followed up by the psychiatrists in charge and seven (14%) individuals from the ARMS group transitioned to psychosis during the follow-up. The average follow-up duration was 31.0 months (SD = 21.9, median 28.0).

3.4.1. BACS results Table 3 contains the results of the BACS. The result of composite zscore of the ARMS group was significantly worse than that of the HC group. Further, the significance on these differences did not change even when controlling for age, years of education, dosage of medication, and premorbid IQ. Similarly, the result of z-score on each individual subtest in the ARMS group was significantly worse than that of the HC.

3.1. Demographic data

Table 1 Demographic data. ARMS (n = 50)

HC (n = 29)

Statistic

P

Males, n (%)

20 (40.0)

13 (44.8)

Fisher's exact test

0.81

Age at testing, M (SD) Years of education, M (SD) Estimated IQ, M (SD)

20.0 (4.1) 12.1 (2.3) 101.5 (10.6)

21.2 (1.0) 14.4 (0.8) 111.5 (6.4)

t = −1.98 U = 293.0 t = −5.23

0.053 b0.001 b0.001

Cohen's d

0.36 1.21 1.08

ARMS: at-risk mental state; HC: healthy control. Estimated IQ was evaluated by the Japanese version of the National Adult Reading Test (JART).

Please cite this article as: Ohmuro, N., et al., The relationship between cognitive insight and cognitive performance among individuals with at-risk mental state for developing ps..., Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.04.031

4

N. Ohmuro et al. / Schizophrenia Research xxx (2017) xxx–xxx

Table 3 Results of the BCIS, BACS, and WCST. ARMS (n = 50)

HC (n = 29)

BCIS

Mean (SD)

Mean (SD)

Self-reflectiveness Self-certainty Composite index

11.6 (4.1) 5.5 (2.6) 6.1 (4.7)

12.9 (2.9) 4.8 (2.3) 8.1 (4.0)

BACS

Raw score M (SD) Z-score M (SD)

Raw score M (SD) Z-score M (SD)

Verbal memory

43.6 (13.0) –2.51 (2.11) 18.6 (5.0) –2.21 (1.95) 69.9 (14.2) –0.77 (1.25) 40.8 (11.1) –1.25 (1.21) 62.6 (15.4) –1.35 (1.23) 17.5 (2.5) –0.95 (1.39) −2.84 (1.89)

59.1 (6.3) 0 (1) 24.3 (2.6) 0 (1) 78.7 (11.6) 0 (1) 52.2 (9.3) 0 (1) 79.6 (12.8) 0 (1) 19.2 (1.8) 0 (1) 0 (1)

Working memory Motor speed Verbal fluency Attention and processing speed Executive function Composite z-score WCST

Mean (SD)

Mean (SD)

CA PEN

3.6 (1.9) 4.9 (5.3)

4.8 (1.5) 2.3 (2.5)

Statistic

P

Cohen's d

t = −1.49 t = 1.24 t = −1.96

0.14 0.22 0.054

0.35 0.28 0.45

U = 448.0 U = 481.0

t = −7.12

b0.001

1.40

t = −6.63

b0.001

1.32

t = −2.82

0.006

0.66

t = −4.66

b0.001

1.10

t = −5.01

b0.001

1.17

t = −3.50

0.001

0.75

t = −8.67

b0.001

1.75

0.004 0.01

0.68 0.58

BCIS: Beck Cognitive Insight Scale; the composite index of the BCIS is the result of subtraction of the self-certainty score from the self-reflectiveness score; BACS: Brief Assessment of Cognition in Schizophrenia; WCST: Wisconsin Card Sorting Test; CA: the number of categories achieved; PEN: the number of perseverative errors of Nelson type; ARMS: at-risk mental state; HC: healthy control.

3.4.2. WCST results Table 3 contains the results of the WCST. For both the numbers of CAs and PENs, the ARMS group showed worse performance than did the HC group.

3.5. Correlations between cognitive insight and cognitive performance Table 4 summarizes the results of the Spearman's rank correlations between the indices of cognitive insight and cognitive assessment. In the ARMS group, higher self-certainty significantly correlated with lower performance in the numbers of CAs (Spearman's ρ = − 0.31, P = 0.03) and PENs (ρ = 0.29, P = 0.04) in the WCST. However, there were no correlations between self-reflectiveness and the Table 4 Spearman's rank correlation coefficients between the scores of the indices of the BCIS and the results of the BACS and WCST. ARMS (n = 50) BCIS BACS CS VM WM MS VF AP EF WCST CA PEN

HC (n = 29)

SR

SC

SR

SC

–0.09 –0.15 0.08 –0.14 –0.03 –0.11 –0.09

−0.07 0.02 −0.06 −0.25 −0.08 −0.11 0.06

0.03 0.14 −0.03 −0.19 −0.03 0.03 −0.08

–0.19 –0.10 –0.31 0.20 –0.02 –0.15 –0.39⁎

–0.04 –0.06

−0.31⁎ 0.29⁎

0.22 −0.32

–0.06 0.00

BCIS: Beck Cognitive Insight Scale; the composite index of the BCIS is the result of subtraction of the self-certainty score from the self-reflectiveness score; SR: self-reflectiveness; SC: self-certainty; CI: composite index; BACS: Brief Assessment of Cognition in Schizophrenia; VM: verbal memory; WM: working memory; MS: motor speed; VF: verbal fluency; AP: attention and processing speed; EF: executive function; CS: composite score; WCST: Wisconsin Card Sorting Test; CA: number of categories achieved; PEN: number of perseverative errors of Nelson type; ARMS: at-risk mental state; HC: healthy control. ⁎ P b 0.05.

WCST indices. In the HC group, on the other hand, there were no correlations between any indices of the BCIS and WCST. We observed no significant correlations between the indices of cognitive insight and the composite z-score of the BACS in either the ARMS or HC groups. Regarding an additional analysis that exploratorily investigated the correlation between cognitive insight and subdomains of the BACS, we could not find any significant correlations in the ARMS group. There was, however, a significant correlation between lower executive function as assessed with the BACS and higher self-certainty in the HC group. 4. Discussion To our knowledge, this is the first study to examine the association of cognitive insight and neurocognitive function among individuals with ARMS. The results showed that higher self-certainty in the ARMS group was significantly correlated with worse performance in terms of the numbers of CAs and PENs in the WCST; it was not, however, associated with any other domain of neurocognitive function. The poorer performance on the WCST indicates that participants had difficulty in shifting to the correct sorting rule and tended to perseverate (Greve et al., 1998). This implies that the excessively high self-certainty observed in individuals with the ARMS might be based on weaknesses in executive functions, including cognitive flexibility or set-shifting ability. Although the exact pathophysiology underlying ARMS remains unexplained, several studies have suggested that frontal dysfunctions might contribute (Allen et al., 2011; Fusar-Poli et al., 2010). This hypothesis is compatible with the present findings, because there is some evidence that poor performance on numbers of CAs or PENs (Barceló and Knight, 2002; Nyhus and Barceló, 2009) and excessive self-certainty (Buchy and Lepage, 2015; Buchy et al., 2016) are associated with dysfunctional or structural abnormalities in the frontal cortex. Thus, a common biological mechanism might underlie excessive self-certainty and dysfunctional set-shifting in subjects with ARMS. Our finding of an association of higher self-certainty with poorer performance on the WCST is consistent with that of a study investigating patients with schizophrenia (Orfei et al., 2010). This implies that the association of overconfidence in one's own judgments with impaired

Please cite this article as: Ohmuro, N., et al., The relationship between cognitive insight and cognitive performance among individuals with at-risk mental state for developing ps..., Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.04.031

N. Ohmuro et al. / Schizophrenia Research xxx (2017) xxx–xxx

set-shifting could exist in both individuals with ARMS and individuals with schizophrenia, and that a common putative mechanism contributes to this. According to this finding, we may speculate there is a possibility that a biological aberration common to both conditions may be responsible for excessively high self-confidence as well as preservation. On the other hand, other studies investigating patients with psychotic disorders (Azzam et al., 2009; Bruno et al., 2012; Cooke et al., 2010) found no such association. Although the reason for the inconsistency among studies is unclear, it might be attributable to methodological differences, including a lack of assessment of perseverative errors (Azzam et al., 2009) or a small sample size (Bruno et al., 2012). Unlike the findings of a meta-analysis of psychosis research, which showed an association of excessive self-certainty with neurocognitive deficits (Nair et al., 2014), no correlation between cognitive insight and neurocognition as evaluated with the BACS in the ARMS group was found. Although many researchers reported neurocognitive dysfunctions in ARMS group, the dysfunctions in this group are generally more variable and/or more subtle than those in the full psychosis group, and therefore the association of excessive self-certainty with neurocognitive dysfunctions should be more variable and/or more subtle in the ARMS group as well. In this case, the small sample size in the present study limited the ability to detect such an association. Because it is suggested that a subtle or higher-order dysfunction could emerge initially in the high-risk state of psychosis (Ohmuro et al., 2016), more complex cognitive abilities—namely, cognitive flexibility or set-shifting as assessed with the WCST—might play a more pivotal role in cognitive insight in the earliest stages of psychosis. This study includes several limitations. First, our results were underpowered due to the small sample size. For example, the difference between the HC and ARMS groups on the 2 BCIS score was not significant and only a trend-level difference on the composite index was observed. The small sample size may be particularly problematic in research on ARMS, given that differences between the HC and ARMS groups were generally more subtle and there is greater clinical, cognitive, etc. variability within the ARMS group. Second, our results might have been affected by various confounding factors caused by the correlational design. Further, the fact that the design was cross-sectional precludes any conclusions on the causality between poor cognitive insight and cognitive impairment. Third, medications might have influenced the cognitive performance of some of the subjects with ARMS. Fourth, we adopted non-parametric analysis for comparison of the WCST results between the ARMS and HC groups. Therefore, it did not let us adjust the effect of the covariates on the WCST and the possibility that covariates such as premorbid IQ would affect the WCST results could not be ruled out. Fifth, we only focused on CA and PEN and did not investigate other WCST indices such as the difficulty of maintaining set, which could hamper further discussion on cognitive flexibility in detail among the current sample. Finally, as the estimated IQ in the healthy subjects was better than the average of the population norm, the cognitive deficits in the ARMS group might have been overestimated. However, we have controlled for the effect of covariates of estimated IQ on cognition using the data of the present HC group instead of using the norm of BACS scores of a Japanese healthy sample (Kaneda et al., 2013), which did not have data on estimated IQ. Future studies should include a larger number of subjects and investigate relationships between cognitive insight and cognitive performance in individuals with ARMS using a longitudinal design. In summary, this study revealed an association between high selfcertainty and dysfunctional set-shifting ability in subjects with ARMS. These findings imply that a common pathophysiology might underlie dysfunctions in cognitive insight and specific executive function in ARMS, which in turn might underlie attenuated psychotic symptoms and the risk of developing psychosis.

5

Role of funding source This work was supported by JSPS KAKENHI Grant Numbers 22390219, 23791307, 25860984, and 16K10240. Contributors NO and KM designed the study and wrote the manuscript. HM and KM contributed to managing the project. NO, KM, CO, TK, FI, and KM recruited and clinically evaluated the participants. YH administered cognitive assessment. NO, MK, and CO managed the data. NO analyzed the data and MK, CO, TK, YH, AS, KI, FI and KM assisted NO with analysis and interpretation of the data. They also approved the final manuscript. Conflicts of interest All authors declare no conflicts of interest for the work presented here. Acknowledgement We thank Emi Sunakawa, Shiori Sato, Mayumi Saito, Aya Takahashi, Tomohiro Uchida, and Rie Koshimichi for their help with the management and preparation of the data.

References Allen, P., Seal, M.L., Valli, I., Fusar-Poli, P., Perlini, C., Day, F., Wood, S.J., Williams, S.C., McGuire, P.K., 2011. Altered prefrontal and hippocampal function during verbal encoding and recognition in people with prodromal symptoms of psychosis. Schizophr. Bull. 37 (4), 746–756. American Psychiatric Association, 1994. Diagnostic and Statistical Manual for Mental Disorders. fourth ed. American Psychiatric Association, Washington, DC. Azzam, H., Sayyah, H., Eissa, A., 2009. Clinical and neurocognitive correlates of insight. Curr. Psychiatr. Ther. 16 (1), 102–111. Barceló, F., Knight, R.T., 2002. Both random and perseverative errors underlie WCST deficits in prefrontal patients. Neuropsychologia 40 (3), 349–356. Beck, A.T., Steer, R.A., Brown, B.K., 1996. Beck Depression Inventory Manual. second ed. Psychological Corporation, San Antonio, TX. Beck, A.T., Baruch, E., Balter, J.M., Steer, R.A., Warman, D.M., 2004. A new instrument for measuring insight: the Beck cognitive insight scale. Schizophr. Res. 68 (2–3), 319–329. Bora, E., Erkan, A., Kayahan, B., Veznedaroglu, B., 2007. Cognitive insight and acute psychosis in schizophrenia. Psychiatry Clin. Neurosci. 61 (6), 634–639. Bora, E., Lin, A., Wood, S.J., Yung, A.R., McGorry, P.D., Pantelis, C., 2014. Cognitive deficits in youth with familial and clinical high risk to psychosis: a systematic review and metaanalysis. Acta Psychiatr. Scand. 130 (1), 1–15. Bruno, N., Sachs, N., Demily, C., Franck, N., Pacherie, E., 2012. Delusions and metacognition in patients with schizophrenia. Cogn. Neuropsychiatry 17 (1), 1–18. Buchy, L., Lepage, M., 2015. Modeling the neuroanatomical and neurocognitive mechanisms of cognitive insight in non-clinical subjects. Cogn. Ther. Res. 39 (4), 415–423. Buchy, L., Czechowska, Y., Chochol, C., Malla, A., Joober, R., Pruessner, J., Lepage, M., 2010. Toward a model of cognitive insight in first-episode psychosis: verbal memory and hippocampal structure. Schizophr. Bull. 36 (5), 1040–1049. Buchy, L., Barbato, M., MacMaster, F.P., Bray, S., Clark, D., Deighton, S., Addington, J., 2016. Cognitive insight is associated with cortical thickness in first-episode psychosis. Schizophr. Res. 172 (1–3), 16–22. Cooke, M.A., Peters, E.R., Fannon, D., Aasen, I., Kuipers, E., Kumari, V., 2010. Cognitive insight in psychosis: the relationship between self-certainty and self-reflection dimensions and neuropsychological measures. Psychiatry Res. 178 (2), 284–289. Engh, J.A., Friis, S., Birkenaes, A.B., Jonsdottir, H., Ringen, P.A., Ruud, T., Sundet, K.S., Opjordsmoen, S., Andreassen, O.A., 2007. Measuring cognitive insight in schizophrenia and bipolar disorder: a comparative study. BMC Psychiatry 7, 71. Fusar-Poli, P., Howes, O.D., Allen, P., Broome, M., Valli, I., Asselin, M.C., Grasby, P.M., McGuire, P.K., 2010. Abnormal frontostriatal interactions in people with prodromal signs of psychosis: a multimodal imaging study. Arch. Gen. Psychiatry 67 (7), 683–691. Fusar-Poli, P., Deste, G., Smieskova, R., Barlati, S., Yung, A.R., Howes, O., Stieglitz, R.D., Vita, A., McGuire, P., Borgwardt, S., 2012. Cognitive functioning in prodromal psychosis: a meta-analysis. Arch. Gen. Psychiatry 69 (6), 562–571. Garety, P.A., Kuipers, E., Fowler, D., Freeman, D., Bebbington, P.E., 2001. A cognitive model of the positive symptoms of psychosis. Psychol. Med. 31 (2), 189–195. Garety, P.A., Bebbington, P., Fowler, D., Freeman, D., Kuipers, E., 2007. Implications for neurobiological research of cognitive models of psychosis: a theoretical paper. Psychol. Med. 37 (10), 1377–1391. Gilleen, J., Greenwood, K., David, A.S., 2011. Domains of awareness in schizophrenia. Schizophr. Bull. 37 (1), 61–72. Greve, K.W., Ingram, F., Bianchini, K.J., 1998. Latent structure of the Wisconsin card sorting test in a clinical sample. Arch. Clin. Neuropsychol. 13 (7), 597–609. Greve, K.W., Stickle, T.R., Love, J.M., Bianchini, K.J., Stanford, M.S., 2005. Latent structure of the Wisconsin card sorting test: a confirmatory factor analytic study. Arch. Clin. Neuropsychol. 20 (3), 355–364. Heaton, R.K., 1981. The Wisconsin Card Sorting Test (Manual). Psychological Assessment Resources, Odessa, Fl. Kaneda, Y., Sumiyoshi, T., Keefe, R., Ishimoto, Y., Numata, S., Ohmori, T., 2007. Brief assessment of cognition in schizophrenia: validation of the Japanese version. Psychiatry Clin. Neurosci. 61 (6), 602–609. Kaneda, Y., Sumiyoshi, T., Nakagome, K., 2013. Evaluation of cognitive functions in a normal population in Japan using the brief assessment of cognition in schizophrenia Japanese version (BACS-J) (in Japanese). Seishin Igaku 55 (2), 167–175.

Please cite this article as: Ohmuro, N., et al., The relationship between cognitive insight and cognitive performance among individuals with at-risk mental state for developing ps..., Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.04.031

6

N. Ohmuro et al. / Schizophrenia Research xxx (2017) xxx–xxx

Kashima, H., Honda, T., Kato, M., Sakura, K., Yokoyama, N., Murakami, M., Shigemori, K., Muramatsu, T., Saito, H., Ooe, Y., Mimura, M., Asai, M., Hosaki, H., 1987. Neuropsychological investigation on chronic schizophrenia-aspect of its frontal functions. In: Takahashi, R., Flor-Henry, P., Gruzelier, J., Niwa, S. (Eds.), Cerebral Dynamics. Laterality and Psychopathology. Elsevier, Amsterdam, pp. 337–345. Katsura, M., Ohmuro, N., Obara, C., Kikuchi, T., Ito, F., Miyakoshi, T., Matsuoka, H., Matsumoto, K., 2014. A naturalistic longitudinal study of at-risk mental state with a 2.4-year follow-up at a specialized clinic setting in Japan. Schizophr. Res. 158 (1–3), 32–38. Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr. Bull. 13 (2), 261–276. Keefe, R.S., Goldberg, T.E., Harvey, P.D., Gold, J.M., Poe, M.P., Coughenour, L., 2004. The brief assessment of cognition in schizophrenia: reliability, sensitivity, and comparison with a standard neurocognitive battery. Schizophr. Res. 68 (2–3), 283–297. Kimhy, D., Jobson-Ahmed, L., Ben-David, S., Ramadhar, L., Malaspina, D., Corcoran, C.M., 2014. Cognitive insight in individuals at clinical high risk for psychosis. Early Interv. Psychiatry 8 (2), 130–137. Lappin, J.M., Morgan, K.D., Valmaggia, L.R., Broome, M.R., Woolley, J.B., Johns, L.C., Tabraham, P., Bramon, E., McGuire, P.K., 2007. Insight in individuals with an at risk mental state. Schizophr. Res. 90 (1–3), 238–244. Martin, J.M., Warman, D.M., Lysaker, P.H., 2010. Cognitive insight in non-psychiatric individuals and individuals with psychosis: an examination using the Beck cognitive insight scale. Schizophr. Res. 121 (1–3), 39–45. Matsuoka, K., Uno, M., Kasai, K., Koyama, K., Kim, Y., 2006. Estimation of premorbid IQ in individuals with Alzheimer's disease using Japanese ideographic script (Kanji) compound words: Japanese version of national adult reading test. Psychiatry Clin. Neurosci. 60 (3), 332–339. Miyakoshi, T., Matsumoto, K., Ito, F., Ohmuro, N., Matsuoka, H., 2009. Application of the comprehensive assessment of at-risk mental states (CAARMS) to the Japanese population: reliability and validity of the Japanese version of the CAARMS. Early Interv. Psychiatry 3, 123–130. Mizuno, M., Suzuki, M., Matsumoto, K., Murakami, M., Takeshi, K., Miyakoshi, T., Ito, F., Yamazawa, R., Kobayashi, H., Nemoto, T., Kurachi, M., 2009. Clinical practice and research activities for early psychiatric intervention at Japanese leading centres. Early Interv. Psychiatry 3, 5–9. Morrison, A.P., Baker, C.A., 2000. Intrusive thoughts and auditory hallucinations: a comparative study of intrusions in psychosis. Behav. Res. Ther. 38 (11), 1097–1106.

Nair, A., Palmer, E.C., Aleman, A., David, A.S., 2014. Relationship between cognition, clinical and cognitive insight in psychotic disorders: a review and meta-analysis. Schizophr. Res. 152 (1), 191–200. Nelson, H.E., 1982. National Adult Reading Test. NFER-Nelson, Windsor, UK. Nyhus, E., Barceló, F., 2009. The Wisconsin card sorting test and the cognitive assessment of prefrontal executive functions: a critical update. Brain Cogn. 71 (3), 437–451. Ohmuro, N., Matsumoto, K., Katsura, M., Obara, C., Kikuchi, T., Hamaie, Y., Sakuma, A., Iizuka, K., Ito, F., Matsuoka, H., 2015. The association between cognitive deficits and depressive symptoms in at-risk mental state: a comparison with first-episode psychosis. Schizophr. Res. 162 (1–3), 67–73. Ohmuro, N., Katsura, M., Obara, C., Kikuchi, T., Sakuma, A., Iizuka, K., Hamaie, Y., Ito, F., Matsuoka, H., Matsumoto, K., 2016. Deficits of cognitive theory of mind and its relationship with functioning in individuals with an at-risk mental state and first-episode psychosis. Psychiatry Res. 243, 318–325. Orfei, M.D., Spoletini, I., Banfi, G., Caltagirone, C., Spalletta, G., 2010. Neuropsychological correlates of cognitive insight in schizophrenia. Psychiatry Res. 178 (1), 51–56. Pedrelli, P., McQuaid, J.R., Granholm, E., Patterson, T.L., McClure, F., Beck, A.T., Jeste, D.V., 2004. Measuring cognitive insight in middle-aged and older patients with psychotic disorders. Schizophr. Res. 71 (2–3), 297–305. Tranulis, C., Lepage, M., Malla, A., 2008. Insight in first episode psychosis: who is measuring what? Early Interv. Psychiatry 2 (1), 34–41. Uchida, T., Matsumoto, K., Kikuchi, A., Miyakoshi, T., Ito, F., Ueno, T., Matsuoka, H., 2009. Psychometric properties of the Japanese version of the Beck cognitive insight scale: relation of cognitive insight to clinical insight. Psychiatry Clin. Neurosci. 63 (3), 291–297. Uchida, T., Matsumoto, K., Ito, F., Ohmuro, N., Miyakoshi, T., Ueno, T., Matsuoka, H., 2014. Relationship between cognitive insight and attenuated delusional symptoms in individuals with at-risk mental state. Psychiatry Res. 217 (1–2), 20–24. Warman, D.M., Lysaker, P.H., Martin, J.M., 2007. Cognitive insight and psychotic disorder: the impact of active delusions. Schizophr. Res. 90 (1–3), 325–333. Yung, A.R., Phillips, L.J., Yuen, H.P., McGorry, P.D., 2004. Risk factors for psychosis in an ultra high-risk group: psychopathology and clinical features. Schizophr. Res. 67 (2– 3), 131–142.

Please cite this article as: Ohmuro, N., et al., The relationship between cognitive insight and cognitive performance among individuals with at-risk mental state for developing ps..., Schizophr. Res. (2017), http://dx.doi.org/10.1016/j.schres.2017.04.031