Deficit in decision-making in chronic, stable schizophrenia: from a reward and punishment perspective

Abstract

Objective: We compared patients with chronic schizophrenia and normal controls with respect to decision-making ability. Measures were implemented to control for the participants' intelligence levels as well as to ensure to use of a moderate sample size. The goal of this study was to confirm inconsistent results from previous studies which had stemmed from too small of a sample size, highly variable performance of normal controls, and not controlling for intelligence as a confounding factor.

Methods: Fifty-two chronic stable schizophrenic inpatients and 55 healthy controls participated in the study. We controlled for intelligence by including subjects with intelligence quotient's (IQ) between 80 and 120, examining any differences in decision-making performance between groups on the Iowa Gambling Task (IGT). We also addressed several issues relating to performance on the IGT, such as working memory and clinical symptoms.

Results: Schizophrenic patients were found to perform poorly on the IGT relative to normal controls (F(1,105)=17.73, p<0.001); however, more importantly, they also displayed the slow yet profitable shift from disadvantageous decks to advantageous decks over time. We also found that when compared with healthy controls, schizophrenic patients showed a poorer performance on the Wisconsin Card Sorting Test (WCST)(t=-5.48, p<0.001 for perseverative error) which was not related to their performance on the IGT.

Conclusion: Based on previous literature and the results of this study, impaired sensitivity to both reward and punishment might be a more plausible explanation for the poor performance on the IGT in the schizophrenic group. We speculated that this impairment seemed related more to the different responsiveness to the magnitude than to the frequency of punishment, and to the different interpretation of less informative verbal cues in the context of the reinforcing schedule.

Keywords: Decision-making; Gambling task; Punishment; Reward; Schizophrenia.

FIGURE 1

Chronological card choice in five blocks of 20 cards. There were significant main effects for group, for block, and for the group by block interaction. A follow-up independent t-test showed that controls performed significantly better than schizophrenic patients in last three blocks. Data are presented as mean±S.E.M. ^*p<0.01, ^**p<0.001.

FIGURE 2

Number of card choices selected from each deck during the 100-card task. Schizophrenic patients selected more frequently from deck B andless frequently from deck D than controls, while there was no difference in selection from decks A and C (after a Bonferroni correction). Regardless of groups, decks B and D were chosen more frequently than decks A and C. Data are presented as mean±SD. ^*p<0.01, ^**p<0.001.