- Email: [email protected]
Neurocognitive measures of prefrontal cortical dysfunction in schizophrenia
Schizophrenia Research 68 (2004) 65 – 73 www.elsevier.com/locate/schres
Neurocognitive measures of prefrontal cortical dysfunction in schizophrenia Laura M. Ritter a,b, James H. Meador-Woodruff b,c, Gregory W. Dalack a,b,* a
Psychiatry Service, VA Ann Arbor Health Care System, Ann Arbor, MI 48105, USA b Department of Psychiatry, University of Michigan, Ann Arbor, MI 48109, USA c Mental Health Research Institute, University of Michigan, Ann Arbor, MI 48109, USA Received 4 September 2002; received in revised form 18 February 2003; accepted 26 February 2003
Abstract Frontal lobe dysfunction in individuals with schizophrenia has frequently been detected using both neuroimaging and neuropsychological testing. Results from previous studies vary in the findings of regional specificity vs. generalized frontal cortical dysfunction. We sought to examine potential regional differences in frontal cortical functioning among patients with schizophrenia vs. a comparison group using two different neurocognitive tasks: the Gambling Task (GT) and the Wisconsin Card Sorting Test (WCST). In general, the GT is thought to reflect function of the ventromedial prefrontal cortex (VMPFC), while the WCST reflects function of the dorsolateral prefrontal cortex (DLPFC). Twenty individuals with schizophrenia or schizoaffective disorder and 15 nonpsychiatrically ill comparison subjects underwent an assessment battery consisting of the GT, WCST, and positive and negative symptom ratings. Patients with schizophrenia performed worse on the GT with respect to total monetary gain ( p = 0.05) and total monetary loss ( p < 0.05). They also preferred disadvantaged vs. advantaged cards ( p < 0.04). Surprisingly, WCST performance was poor in both groups and was not significantly different between groups. These findings are at some variance with those in the previously reported literature, but nonetheless support the idea that prefrontal cortical areas mediating different cognitive tasks may be distinguished by specific neurocognitive assessments. D 2003 Elsevier B.V. All rights reserved. Keywords: Gambling Task; Ventromedial prefrontal cortex; Wisconsin Card Sorting Test; Dorsolateral prefrontal cortex; Schizophrenia
1. Introduction Abnormal cortical function, particularly in the area of the frontal lobes, has long been recognized to be one important component of a broad spectrum of impairments seen in schizophrenia. Indeed, memory, * Corresponding author. Department of Psychiatry, Ann Arbor VA Medical Center, 2215 Fuller Road, 116A, Ann Arbor, MI 48105, USA. Tel.: +1-734-761-7926; fax: +1-734-769-7410. E-mail address: [email protected] (G.W. Dalack). 0920-9964/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0920-9964(03)00086-0
attention, and executive function are attributes of cognition that have been reported to be affected by frontal lobe dysfunction (Tollefson, 1996). These cognitive abnormalities have been assessed by neurocognitive tasks, with varying specificity for frontal cortex areas. Neuropsychological tests such as Wisconsin Card Sorting Test (WCST, Heaton, 1981), Trail Making B (Russell et al., 1970), and Wechsler Adult Intelligence Scale-Revised (WAIS-R) Digit Symbol subtest (Matarazzo, 1972) have been used to identify impairments in set shifting, judgment, attention, con-
66
L.M. Ritter et al. / Schizophrenia Research 68 (2004) 65–73
centration, mental sequencing, planning and anticipation, and concept formation among individuals with schizophrenia (Gold and Harvey, 1993; Goldberg et al., 1988, 1990, 1993). Of the different neurocognitive tasks used in identifying cortical dysfunction in schizophrenia, the WCST has received the most attention. This task is considered a somewhat more specific measure of dorsolateral prefrontal cortex (DLPFC) function. The WCST requires subjects to sort a deck of cards on the basis of a series of unknown categories. To determine the category by which to sort the cards, the individual must use feedback that accompanies each sort (i.e. ‘‘correct’’ or ‘‘wrong’’). As the test progresses and the subject sorts cards correctly, the card-sorting category changes. Executive functions are employed in determining the correct sorting category and adapting when the category changes. Performance on the WCST has been correlated to brain activity in the DLPFC region in neuroimaging studies in both comparison and schizophrenic subjects, and is typically impaired in individuals with schizophrenia (Berman et al., 1986, 1995; Marenco et al., 1993; Weinberger et al., 1986). For example, while performing the WCST, the DLPFC of comparison subjects is physiologically activated above baseline, but it is not activated to the same extent in subjects with schizophrenia (Berman et al., 1992; Weinberger et al., 1986, 1988). Further, individuals with schizophrenia show deficits in performance outcomes of the WCST (i.e. fewer categories completed, more total errors, and more perseverative errors) (Basso et al., 1998; Berman et al., 1997; Koren et al., 1998), and poor performance is typically correlated with negative symptom ratings (Basso et al., 1998; Berman et al., 1997). While some reports suggest that prefrontal cortical dysfunction in schizophrenia is regionally specific (Goldberg et al., 1987; Wilder et al., 1998), other data support generalized bilateral dysfunction (Blanchard and Neale, 1994; Braff et al., 1991; Chelune et al., 1979; Goldberg et al., 1990). The use of tasks with putative regional specificity, such as the Gambling Task (GT) (Bechara et al., 1994), that have been applied to other psychiatric populations (Abbruzzese et al., 1995; Cavedini et al., 2002b) could be useful in addressing the question of regionally specific deficits of the prefrontal cortical functioning in schizophrenia.
The GT was developed as an assessment tool of decision-making process of patients suffering from ventromedial prefrontal cortex (VMPFC) lesions. These patients appear to have intact intellectual function, but demonstrate impairments in personal and social decision making where integration of negative consequences and learning from previous actions are required. The GT is an attempt to mimic real-life decision making in which the patient is presented with options, receives feedback from decisions made, and has the opportunity to incorporate that information into subsequent decisions. Numerous studies have demonstrated the specificity of this task for bilateral VMPFC dysfunction in patients with lesions in this region from stroke or tumor resection (Bechara et al., 1994, 1998, 2000). The GT has been applied to some psychiatric populations to assess VMPFC function. These studies have included subjects with obsessive – compulsive disorder (OCD) (Cavedini et al., 2002b), antisocial personality disorder (Mazas et al., 2000), substance use disorders (alcohol, stimulants, and opiates) (Bartzokis et al., 2000; Bechara et al., 2001; Grant et al., 2000), and pathological gambling (Cavedini et al., 2002a). The studies focusing on decision making in individuals with substance use disorders suggest that active substance use was associated with VMPFC dysfunction. Individuals with antisocial personality disorder also demonstrated impaired performance on the GT. In the OCD and pathological gambling groups, patients with active symptoms (untreated or not responsive to treatment) demonstrated impairment on the task, indicative again of VMPFC impairment. Interestingly, OCD patients who responded to treatment performed like normal controls, suggesting that, for some diagnoses, abnormal GT performance may be a state-dependent phenomenon. Assessment of VMPFC dysfunction in schizophrenia has been limited. To our knowledge, only one study has previously studied the GT in individuals with this disorder. Wilder et al. (1998) used the GT among a group of 12 patients with schizophrenia and 30 normal controls. The results of this study did not show a statistically significant difference on the performance outcomes of the GT between the groups. Due to the small sample size used by Wilder et al., the relative paucity of literature in this area, and the persistent ambiguity about possible global vs. regional
L.M. Ritter et al. / Schizophrenia Research 68 (2004) 65–73
frontocortical dysfunction in schizophrenia, we sought to replicate Wilder et al.’s study by comparing GT and WCST performance among a larger cohort with schizophrenia or schizoaffective disorder vs. comparison subjects. We included those with either schizophrenia or schizoaffective disorder, depressed subtype, as previous studies have suggested that schizoaffective disorder, depressed subtype, may be more closely related to schizophrenia than to mood disorders (Andreasen et al., 1987). We hypothesized that subjects with schizophrenia would show poorer performance on both the GT and WCST than comparison subjects, and that among those with schizophrenia, poor performance on these tasks would correlate with negative symptomatology.
2. Methods This study was approved by the Internal Review Boards of both the VA Ann Arbor Healthcare System and the University of Michigan. Psychiatric and comparison subjects were recruited from the Outpatient Psychiatry Clinics and General Medicine Clinics, respectively, at the VA Ann Arbor Healthcare System. Psychiatric subjects met the following inclusion criteria: (1) age 18 –65; (2) DSM-IV diagnosis by Structured Clinical Interview for DSM-IV (SCID) (First et al., 1997) of chronic schizophrenia or schizoaffective disorder, depressed subtype; (3) psychiatrically stable (no medication changes or psychiatric inpatient admissions) for at least 3 months prior to study. Comparison subjects were between the ages of 18 – 65 and had no DSM-IV diagnosis (based on MINI-Structured Interview) (Sheehan et al., 1998). Potential psychiatric and comparison subjects were excluded if they had an active substance use disorder diagnosis in the 3 months prior to screening, a history of serious head injury, seizure, or other neurological disorder, or stroke. After providing written informed consent, individuals from both groups underwent an assessment battery consisting of the North American Adult Reading Test (an IQ test of premorbid function) (Blair and Spreen, 1989), computerized versions of the GT and WCST, and standardized symptom ratings in a single 1 –2 h session. Symptom rating scales included the Brief Psychiatric Rating Scale (BPRS) (Overall and Gorham, 1962; Woerner et al., 1988) and
67
the Schedule for the Assessment of Negative Symptoms (SANS) (Andreasen, 1982a,b). All subjects received $15 compensation for their participation. 2.1. Gambling Task Subjects turn over cards from one of four different decks (A, B, C, D; 60 cards per deck) until 100 cards are chosen. Before starting the task, the subject is told that the goal of the task is to maximize the profit on a loan of $2000 in play money received at the start of the task. After each card turn, the subject receives a variable play money ‘‘reward’’. Some cards also carry a monetary ‘‘penalty’’. A fixed schedule determines when penalties will occur and how much money is deducted from the subject’s money total. Decks A and B carry higher rewards, but larger penalties, while decks C and D carry lower rewards, but lower penalties. A greater selection of C and D cards will result in a net gain, while a greater selection of A and B cards will result in a net loss. Subjects are free to switch from deck to deck throughout the task and are not told when the task will end. Individuals with VMPFC damage display difficulty anticipating results, typically make riskier choices (A + B>C + D), choose more disadvantage cards (A, B) over the course of the task, and tend to lose more money compared to control subjects (Bechara et al., 1994, 1998, 2000, 2001). 2.2. Wisconsin Card Sort Test The WCST was administered in a computerized format according to the Heaton (1981) protocol in which subjects sort response cards until they have matched six categories or sorted all 128 cards. Cards are matched on the basis of color, shape, and number, and the rule to which cards are matched changes after 10 consecutive correct card sorts. This administration approach differs somewhat from other studies in this area (Weinberger et al., 1988) in that the test is terminated after the successful completion of six categories rather than continuing until all 128 cards have been sorted. The test was designed to measure mental flexibility through the subject’s ability to shift between sets (Berg, 1948). Results typically reported on the WCST include perseverative errors and categories completed. These measures are felt to most directly reflect DLPFC function (Milner, 1963). In the
68
L.M. Ritter et al. / Schizophrenia Research 68 (2004) 65–73
computerized version employed here, subjects used the keyboard to match response cards to the chosen category. 2.3. Data analysis Data analyses of demographic variables, neuropsychological factors, and symptom ratings were done using the chi-square test (ethnicity, history of substance use disorders), group t-tests (age, IQ, income, education, GT and WCST outcome variables), simple regression (symptom rating correlation with neuropsychological test factors), and two-way analysis of variance (ANOVA; group by deck interaction effects). Post-hoc analyses were done using the Newman – Keuls test for significance.
3. Results Demographic variables are summarized in Table 1. There were no significant differences in age, IQ, education, ethnic composition, and annual income (earned and/or disability) between groups. We compared the rates of the past history of substance use disorders between groups, recognizing that previous significant substance use might have long-lasting deleterious effects on cortical function. The rates of Table 1 Subject demographics
Age (years) Gender Ethnicity/race
IQ Income (earned and disability) Education (years) History of drug/alcohol use disorder Cocaine Alcohol
Psychiatric subjects N = 20
Comparison subjects N = 15
48.5 F 6.0 100% Male 5% African American; 95% Caucasian 104.8 F 10.0 $17,098 F 9582
47.1 F 10.2 100% Male 13% African American; 87% Caucasian 100.2 F 12.3 $25,464 F 18,889
13.8 F 1.70
13.3 F 1.45
5/20 11/20
1/15 3/15*
Demographic comparisons between psychiatric and comparison subjects were nonsignificant except for history of alcohol use disorder. * Chi-square 4.38; df = 1; p < 0.05.
a past history of cocaine abuse/dependence were not significantly different between the groups (Fisher’s Exact p>0.20). Rates of a past history of alcohol dependence, however, were significantly different. Fifty-five percent of the psychiatric subjects had a past history of alcohol abuse or dependence vs. 33% of the comparison group (Fisher’s Exact p < 0.05). Analysis of GT and WCST data including history of alcohol abuse/dependence as a covariate did not alter the results in any way. Hence, only the straight group comparisons are presented. In the group of subjects with psychiatric illness, 15 subjects had a DSM-IV diagnosis of schizophrenia and 5 subjects were diagnosed as schizoaffective. Seventeen were prescribed atypical antipsychotic agents, and the remaining were given typical antipsychotic agents. Lastly, by means of self-report, the average number of years psychiatrically ill at the time of study was 25.8 F 5.1 years. On multiple GT outcome measures, individuals with schizophrenia performed significantly worse than comparison subjects (Table 2). Individuals with schizophrenia chose significantly more disadvantaged than advantaged cards ( F = 4.91; df = 1; p < 0.04) and had a significantly larger differential between disadvantaged and advantaged cards selected ( F = 4.91; df = 1; p < 0.04). While subjects with schizophrenia gained significantly more money ( F = 4.10; df = 1; p = 0.05), they also lost significantly more money ( F = 4.55; df = 1; p < 0.05). A deck by group interaction ( F = 3.68; df = 3; p = 0.01) resulted from individuals with schizophrenia choosing more Deck B cards than comparison individuals choosing Deck A cards ( p < 0.001) and comparison subjects choosing Deck D cards more than subjects with schizophrenia choosing Deck A cards ( < 0.02) (Fig. 1). Examining chronological card choice in blocks of 20 cards, the subjects with schizophrenia began picking more disadvantaged cards, and only slowly improved their selection mix compared to the control group (Fig. 2). A main effect for diagnosis was found ( F = 7.56; df = 1; p < 0.007), but there was not a diagnosis by card block interaction. Significant differences were not found between the two groups on WCST outcome variables (Table 2), although both groups performed poorly. Additionally, performance on the GT was not correlated to WCST performance for either group.
L.M. Ritter et al. / Schizophrenia Research 68 (2004) 65–73
69
Table 2 GT and WCST outcome measures
Gambling task
Wisconsin card sorting test
advantaged minus disadvantaged cards advantaged cards selected disadvantaged cards selected money gained money lost net gain % error % perseverative error % conceptual level categories completed failure to maintain set
Psychiatric subjects N = 20
Comparison subjects N = 15
5.2 F 19.7* 47.4 F 9.84* 52.6 F 9.84* $8388 F 632y $9643 F 1316* $1156 F 783* 35.9% F 20.0 18.8% + 15.0 53.0% F 25.8 3.5 F 2.4 1.8 F 1.8
14.4 F 32.5 57.2 F 16.25 42.8 F 16.25 $7870 F 882 $8390 F 2147 $480 F 1333 36.5% F 19.7 19.7% F 15.9 53.3% F 27.4 3.8 F 2.3 1.4 F 1.4
Wisconsin Card Sorting Test and Gambling Task results for psychiatric and comparison subjects. Significance between group differences is indicated. * p < 0.05. y p = 0.05.
Not surprisingly, subjects with schizophrenia were significantly more symptomatic on the BPRS and SANS standardized rating scales (BPRS: comparison subjects 20.3 F 1.9 vs. schizophrenic subjects 32.6 F 5.1, p < 0.0001; SANS: comparison subjects 4.1 F 3.6 vs. schizophrenic subjects 21.4 F 7.5, p < 0.0001). Total scores on both the BPRS and SANS did not correlate to GT or WCST performance for the schizophrenic group. However, selected BPRS and SANS
subscale scores did correlate to WCST performance for individuals with schizophrenia. The hostility subscale of the BPRS was negatively correlated to percent error (r = 0.489; df = 19; p < 0.03) and positively correlated with percent conceptual level (r = 0.484; df = 19; p < 0.04). None of the other BPRS subscales (i.e. positive, negative, or activation) were correlated to performance on either the WCST or GT outcome variables. The SANS subscales of attention
Fig. 1. Card choices of psychiatric and comparison subjects during the 100-card task.
Fig. 2. Individuals with schizophrenia shifted more slowly than comparison subjects towards choosing more advantaged cards.
70
L.M. Ritter et al. / Schizophrenia Research 68 (2004) 65–73
(r = 0.445; df = 19; p < 0.05) and alogia ( p < 0.01) were positively correlated to the failure to maintain set WCST variable for individuals with schizophrenia.
4. Discussion In this study, we found that based on performance on a measure felt to reflect VMPFC function, subjects with schizophrenia performed poorly compared to a nonpsychiatrically ill comparison group. Gambling Task performance was not related to subjects’ psychiatric symptoms, or performance on the WCST, a measure more reflective of DLPFC function. Interestingly, both subjects with schizophrenia and the comparison group demonstrated impaired performance on the WCST. Further, there was some relationship between negative symptom subscale scores and WCST in the group with schizophrenia. While the differences between groups by card block were not significant, comparison subjects moved quickly to pick more advantageous than disadvantageous cards by the second block of 20 cards (Fig. 2) and maintained that bias throughout the rest of the task. Subjects with schizophrenia began with a greater preference for more disadvantageous cards, and took until block 4 (cards 61– 80) to demonstrate a shift in preference to more advantageous cards. In contrast, among subjects with VMPFC lesions, choice of disadvantageous cards persists throughout the task (Bechara et al., 2001) suggesting that they never adjust their card choice based on feedback from previous choices. In studies examining GT performance in other cohorts with psychiatric diagnoses, pathological gamblers (Cavedini et al., 2002a) and individuals OCD (Cavedini et al., 2002b) appear not to shift to more advantageous cards (like those with VMPFC lesions), while those with antisocial personality (Mazas et al., 2000) or substance use disorders (Bechara et al., 2001), like our schizophrenic subjects, slowly but eventually shift to more advantageous cards. Subjects with VMPFC cortical lesions appear insensitive to future consequences and are guided more by the immediate possibility of reward (Bechara et al., 2000). Our subjects with schizophrenia appear more sensitive to future consequences, but take a longer time to incorporate and act on the information fed
back from each card selection. While processing of gains and losses clearly involves the VMPFC, there is also evidence to suggest that the anterior cingulate cortex is also involved in this processing. Abnormalities in ACC function have been implicated in schizophrenia (Carter et al., 2001; Suhara et al., 2002), and the ACC has been shown to be involved in a similar monetary decision-making task (Gehring and Willoughby, 2002). Hence, it is possible that the impaired GT performance in schizophrenia represents abnormal function in other structures connected to the VMPFC cortex and involved in the complex decision-making process tested by the GT. We did not anticipate that we would be unable to find a difference in WCST performance between our subjects with schizophrenia and a well-matched comparison group. Our lack of a difference between the two groups would appear to be accounted for by the poorer than expected performance in the comparison group compared to controls in other reports (Bechara et al., 2001; Koren et al., 1998). At the same time, the GT performance of our comparison subjects is quite comparable to that of control groups from other studies. Similar to Cavedini et al. (2002a), our comparison subjects selected comparable numbers of advantageous and disadvantageous cards. Additionally, our comparison subjects had a positive advantageous minus disadvantageous differential after the first block of 20 cards and continued to be positive for the remainder of the task, following the same performance pattern of the controls in Bechara et al. (2001) and Cavedini et al. (2002a). Based on our results, the effect size for the GT findings ranged from 0.7 to 0.8. These GT effects sizes are similar to those calculated from data presented in Grant et al. (2000), but are 30 –50% greater than those calculated from data reported by Wilder et al. (1998), who studied a smaller group with schizophrenia (N = 12) compared to a larger control group (N = 30). In contrast, the effect sizes for our Wisconsin Card Sort Findings were much smaller, ranging from less than 0.1 to 0.3, despite variances that were comparable to studies of nonschizophrenic substance abusers (Grant et al., 2000) as well as those with schizophrenia (Saoud et al., 1998). Hence, we are confident that the lack of a difference in the WCST results was not the result of being underpowered.
L.M. Ritter et al. / Schizophrenia Research 68 (2004) 65–73
Although we did not find a difference in WCST performance between our schizophrenic and comparison groups, we did find that positive and negative symptom subscale scores were correlated with WCST measures for individuals with schizophrenia. This finding is in agreement with previous reports that suggest that increased negative symptomatology is associated with poor WCST performance (Basso et al., 1998; Berman et al., 1997). We did not find that symptom subscales were correlated to GT performance for this same group, again suggesting that performance of these two neuropsychological tests may map to different prefrontal cortical areas, and that the negative symptoms of schizophrenia are likely not the result of homogenous impairment of the prefrontal cortex. Several limitations to these data should be considered. Our psychiatric sample was not diagnostically homogenous. Differences in GT performance between those with schizophrenia vs. schizoaffective disorder, depressed subtype, could not be excluded with our sample size, and might have affected our results. Future studies might specifically compare GT performance between these two diagnoses. In addition, schizophrenia and schizoaffective disorder are by no means purely frontal lobe disorders. The broader array of cortical and subcortical dysfunction seen in these conditions likely also impacts on performance on neurocognitive tests like the GT and WCST. Likewise, the GT or WCST are not purely tests of frontal cortical function. Indeed, some evidence suggests that the specificity of the WCST for frontal lobe damage is limited (Anderson et al., 1991). While our data support the notion that regional cortical dysfunction can be discerned in schizophrenia, further investigation is needed. Finally, our use of the keyboard response, computerized version of the WSCT, might have affected our results. One study in normal subjects (Feldstein et al., 1999) indicates that performance on various computer versions of the WCST can differ from that on the standard (manual) version (Feldstein et al., 1999). Indeed, the ‘‘keyboard’’ computerized version, similar to the one we used in our study, was found to result in lower scores when compared to the standard version. Nonetheless, both groups in our study used the same version, and so should have been comparably affected. Moreover, the scores of our comparison group were remarkably close to those of
71
the psychiatric subjects, and lower than control or comparison groups in other studies regardless of version administered. In summary, we conclude that GT performance in schizophrenia is impaired. This may reflect impairment in VMPFC function and/or abnormalities in the cortico-subcortical circuitry putatively involved in this decision-making task. At the same time, GT performance was unrelated to severity of schizophrenic symptomatology, or a measure reflective, in part, of DLPFC function. This suggests that regionally specific deficits in schizophrenia appear more likely than global prefrontal cortical dysfunction. Acknowledgements The authors would like to thank Dr. Antoine Bechara for providing a computerized version of the Gambling Task and helpful consultation. We also thank Ingrid Lund for her help with data acquisition. Earlier versions of this work were presented at the 40th annual meeting of the American College of Neuropsychopharmocology, Kona, HI, December 2001 and the 57th annual meeting of the Society of Biological Psychiatry, Philadelphia, PA, May 2002. References Abbruzzese, M., Bellodi, L., Ferri, S., Scarone, S., 1995. Frontal lobe dysfunction in schizophrenia and obsessive – compulsive disorder: a neuropsychological study. Brain Cogn. 27, 202 – 212. Anderson, S.W., Damasio, H., Jones, R.D., Tranel, D., 1991. Wisconsin Card Sorting Test performance as a measure of frontal lobe damage. J. Clin. Exp. Neuropsychol. 13, 909 – 922. Andreasen, N.C., 1982a. Negative symptoms in schizophrenia: definition and reliability. Arch. Gen. Psychiatry 38, 784 – 788. Andreasen, N.C., 1982b. Negative versus positive schizophrenia: definition and validation. Arch. Gen. Psychiatry 38, 788 – 794. Andreasen, N.C., Rice, J., Endicott, J., Coryell, W., Grove, W.M., Reich, T., 1987. Familial rates of affective disorder. A report from the National Institute of Mental Health Collaborative Study. Arch. Gen. Psychiatry 45, 461 – 469. Bartzokis, G., Lu, P.H., Beckson, M., Rapoport, R., Grant, S., Wiseman, E.J., London, E.D., 2000. Abstinence from cocaine reduces high-risk gambling responses on a Gambling Task. Neuropsychopharmacology 22, 102 – 103. Basso, M.R., Nasrallah, H.A., Olson, S.C., Bornstein, R.A., 1998. Neuropsychological correlates of negative, disorganized and psychotic symptoms in schizophrenia. Schizophr. Res. 31, 99 – 111.
72
L.M. Ritter et al. / Schizophrenia Research 68 (2004) 65–73
Bechara, A., Damasio, A.R., Damasio, H., Anderson, S.W., 1994. Insensitivity to future consequences following damage to human prefrontal cortex. Cognition 50, 7 – 15. Bechara, A., Damasio, H., Tranel, D., Anderson, S.W., 1998. Dissociation of working memory from decision-making within the human prefrontal cortex. J. Neurosci. 18, 428 – 437. Bechara, A., Tranel, D., Damasio, H., 2000. Characterization of the decision-making deficit of patients with ventromedial prefrontal cortex lesions. Brain 123, 2189 – 2202. Bechara, A., Dolan, S., Denburg, N., Hindes, A., Anderson, S.W., Nathan, P.E., 2001. Decision-making deficits, linked to a dysfunctional ventromedial prefrontal cortex, revealed in alcohol and stimulant abusers. Neuropsychologia 39, 376 – 389. Berg, E.A., 1948. A simple objective technique for measuring flexibility in thinking. J. Gen. Psych. 19, 15 – 22. Berman, K.F., Zec, R.F., Weinberger, D.R., 1986. Physiological dysfunction of dorsolateral prefrontal cortex in schizophrenia: II. Role of neuroleptic treatment, attention, and mental effort. Arch. Gen. Psychiatry 43, 126 – 135. Berman, K.F., Torrey, E.F., Daniel, D.G., Weinberger, D.R., 1992. Regional cortical blood flow in monozygotic twins discordant and concordant for schizophrenia. Arch. Gen. Psychiatry 49, 927 – 934. Berman, K.F., Ostrem, J.L., Randolph, C., Gold, J., Goldberg, T.E., Coppola, R., Carson, R.E., Herscovitch, P., Weinberger, D.R., 1995. Physiological activation of cortical networks during performance of the Wisconsin Card Sorting Test: a positron emission tomography study. Neuropsychologia 33 (8), 1027 – 1046. Berman, I., Viegner, B., Mason, A., Allan, E., Pappas, D., Green, A.I., 1997. Differential relationships between positive and negative symptoms and neuropsychological deficits in schizophrenia. Schizophr. Res. 25, 1 – 10. Blair, J.R., Spreen, O., 1989. Predicting premorbid IQ: a revision of the National Adult Reading Test. Clin. Neuropsychol. 3, 129 – 136. Blanchard, J.J., Neale, J.M., 1994. The neuropsychological signature of schizophrenia: generalized or differential deficit? Am. J. Psychiatry 151, 40 – 48. Braff, D.L., Heaton, R., Kuck, J., Cullum, M., Moranville, J., Grant, I., Zisook, S., 1991. The generalized pattern of neuropsychological deficits in outpatients with chronic schizophrenia with heterogeneous Wisconsin Card Sorting Test results. Arch. Gen. Psychiatry 48, 891 – 898. Carter, C.S., MacDonald, A.W., Ross, L.L., Stenger, V.A., 2001. Anterior cingulate cortex activity and impaired self-monitoring of performance in patients with schizophrenia: an event-related fMRI study. Am. J. Psychiatry 158 (9), 1423 – 1428. Cavedini, P., Giovanna, R., Keller, R., D’Annucci, A., Bellodi, L., 2002a. Frontal lobe dysfunction in pathological gambling patients. Biol. Psychiatry 51, 334 – 341. Cavedini, P., Riboldi, G., D’Annucci, A., Belotti, P., Cisima, M., Bellodi, L., 2002b. Decision-making heterogeneity in obsessive – compulsive disorder: ventromedial prefrontal cortex function predicts different treatment outcomes. Neuropsychologia 40 (2), 205 – 211. Chelune, G.J., Heaton, R.K., Leham, R.A.W., Robinson, A., 1979. Level versus pattern of neuropsychological performance among
schizophrenic and diffusely brain-damaged patients. J. Consult. Clin. Psychol. 47, 155 – 163. Feldstein, S.N., Keller, F.R., Portman, R.E., Durham, R.L., Klebe, K.J., Hasker, P.D., 1999. A comparison of computerized and standard versions of the Wisconsin Card Sorting Test. Clin. Neuropsychol. 13, 303 – 313. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 1997. Structured Clinical Interview for DSM-IV Axis I DisordersClinician Version (SCID-CV). American Psychiatric Press, Washington, DC. Gehring, W.J., Willoughby, A.R., 2002. The medial frontal cortex and the rapid processing of monetary gains and losses. Science 295, 2279 – 2282. Gold, J.M., Harvey, P.D., 1993. Cognitive deficits in schizophrenia. Psychiatr. Clin. North Am. 16, 295 – 312. Goldberg, T.E., Weinberger, D.R., Berman, K.F., Pliskin, N.H., Podd, M.H., 1987. Further evidence for dementia of the prefrontal type in schizophrenia? 1987. A controlled study of teaching Wisconsin Card Sorting Test. Arch. Gen. Psychiatry 44, 1008 – 1014. Goldberg, T.E., Kelsoe, J.R., Weinberger, D.R., Pliskin, N.H., Kirwin, P.D., Berman, K.F., 1988. Performance of schizophrenic patients on putative neuropsychological tests of frontal lobe function. Int. J. Neurosci. 42, 51 – 58. Goldberg, T.E., Ragland, J.D., Torrey, E.F., Gold, J.M., Bigelow, L.B., Weinberger, D.R., 1990. Neuropsychological assessment of monozygotic twins discordant for schizophrenia. Arch. Gen. Psychiatry 47, 1066 – 1072. Goldberg, T.E., Torrey, E.F., Gold, J.M., Ragland, J.D., Bigelow, L.B., Weinberger, D.R., 1993. Learning and memory in monozygotic twins discordant for schizophrenia. Psychol. Med. 23, 71 – 85. Grant, S., Contoreggi, C., London, E.D., 2000. Drug abusers show impaired performance in a laboratory test of decision making. Neuropsychologia 38, 1180 – 1187. Heaton, R.K., 1981. Wisconsin Card Sorting Test Manual. PAR, Odessa, FL. Koren, D., Harrision, W.S., Kremen, W.S., Goldstein, J.M., Faraone, S.V., Seidman, L.J., Lyons, M.J., Caplan, B., Tsuang, M.T., 1998. Factor structure of the Wisconsin Card Sorting Test: dimensions of deficit in schizophrenia. Neuropsychology 12, 289 – 302. Marenco, S., Coppola, R., Daniel, D.G., Zigun, J.R., Weinberger, D.R., 1993. Regional cerebral blood flow during the Wisconsin Card Sorting Test in normal subjects as studied by xenon-133 dynamic SPECT: comparison of absolute values, percent distribution values, and covariance analysis. Psychiatry Res. 50, 177 – 192. Matarazzo, J.D., 1972. Wechsler’s Measurement and Appraisal of Adult Intelligence, 5th ed. Williams & Wilkins, Baltimore. Mazas, C., Finn, P.R., Steinmetz, J.E., 2000. Decision-making biases, antisocial personality, and early-onset alcoholism. Alcohol., Clin. Exp. Res. 24, 1036 – 1040. Milner, B., 1963. Effects of different brain lesions on card sorting. Arch. Neurol. 9, 100 – 110. Overall, J.E., Gorham, D.R., 1962. The Brief Psychiatric Rating Scale. Psychol. Rep. 10, 799 – 812.
L.M. Ritter et al. / Schizophrenia Research 68 (2004) 65–73 Russell, E.W., Neuringer, C., Goldstein, G., 1970. Assessment of Brain Damage: A Neuropsychological Key Approach. Wiley, New York. Saoud, M., Hueber, T., Mandran, H., Dalery, J., d’Amato, T., 1998. Olfactory identification deficiency and WCST performance in men with schizophrenia. Psychiatry Res. 81, 251 – 257. Sheehan, D.V., Lecrubier, Y., Sheehan, K.H., Amorin, P., Janavs, J., Weiller, E., Hergueta, T., Baker, R., Dunbar, G.C., 1998. The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic interview for DSM-IV and ICD-10. J. Clin. Psychiatry 59, 22 – 33. Suhara, T., Okubo, Y., Yasuno, F., Sudo, Y., Inoue, M., Ichimiya, T., Nakashima, Y., Nakayama, K., Tanada, S., Suzuki, K., Halldin, C., Farde, L., 2002. Decreased dopamine D2 receptor binding in the anterior cingulate cortex in schizophrenia. Arch. Gen. Psychiatry 59, 25 – 30. Tollefson, G.D., 1996. Cognitive function in schizophrenic patients. J. Clin. Psychiatry 57 (Suppl. 11), 31 – 39.
73
Weinberger, D.R., Berman, K.F., Zec, R.F., 1986. Physiological dysfunction of dorsolateral prefrontal cortex in schizophrenia: I. Regional cerebral blood flow (rCBF) evidence. Arch. Gen. Psychiatry 43, 114 – 125. Weinberger, D.R., Berman, K.F., Illowsky, B.P., 1988. Physiological dysfunction of dorsolateral prefrontal cortex in schizophrenia: III. A new cohort and evidence for a monoaminergic mechanism. Arch. Gen. Psychiatry 45, 609 – 615. Wilder, K.E., Weinberger, D.R., Goldberg, T.E., 1998. Operant conditioning and orbitofrontal cortex in schizophrenic patients: unexpected evidence for intact functioning. Schizophr. Res. 30, 169 – 174. Woerner, M.G., Mannuzza, S., Kane, J.M., 1988. Anchoring the BPRS: an aid to improved reliability. Psychopharmacol. Bull. 24, 112 – 117.
Neurocognitive measures of prefrontal cortical dysfunction in schizophrenia Laura M. Ritter a,b, James H. Meador-Woodruff b,c, Gregory W. Dalack a,b,* a
Psychiatry Service, VA Ann Arbor Health Care System, Ann Arbor, MI 48105, USA b Department of Psychiatry, University of Michigan, Ann Arbor, MI 48109, USA c Mental Health Research Institute, University of Michigan, Ann Arbor, MI 48109, USA Received 4 September 2002; received in revised form 18 February 2003; accepted 26 February 2003
Abstract Frontal lobe dysfunction in individuals with schizophrenia has frequently been detected using both neuroimaging and neuropsychological testing. Results from previous studies vary in the findings of regional specificity vs. generalized frontal cortical dysfunction. We sought to examine potential regional differences in frontal cortical functioning among patients with schizophrenia vs. a comparison group using two different neurocognitive tasks: the Gambling Task (GT) and the Wisconsin Card Sorting Test (WCST). In general, the GT is thought to reflect function of the ventromedial prefrontal cortex (VMPFC), while the WCST reflects function of the dorsolateral prefrontal cortex (DLPFC). Twenty individuals with schizophrenia or schizoaffective disorder and 15 nonpsychiatrically ill comparison subjects underwent an assessment battery consisting of the GT, WCST, and positive and negative symptom ratings. Patients with schizophrenia performed worse on the GT with respect to total monetary gain ( p = 0.05) and total monetary loss ( p < 0.05). They also preferred disadvantaged vs. advantaged cards ( p < 0.04). Surprisingly, WCST performance was poor in both groups and was not significantly different between groups. These findings are at some variance with those in the previously reported literature, but nonetheless support the idea that prefrontal cortical areas mediating different cognitive tasks may be distinguished by specific neurocognitive assessments. D 2003 Elsevier B.V. All rights reserved. Keywords: Gambling Task; Ventromedial prefrontal cortex; Wisconsin Card Sorting Test; Dorsolateral prefrontal cortex; Schizophrenia
1. Introduction Abnormal cortical function, particularly in the area of the frontal lobes, has long been recognized to be one important component of a broad spectrum of impairments seen in schizophrenia. Indeed, memory, * Corresponding author. Department of Psychiatry, Ann Arbor VA Medical Center, 2215 Fuller Road, 116A, Ann Arbor, MI 48105, USA. Tel.: +1-734-761-7926; fax: +1-734-769-7410. E-mail address: [email protected] (G.W. Dalack). 0920-9964/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0920-9964(03)00086-0
attention, and executive function are attributes of cognition that have been reported to be affected by frontal lobe dysfunction (Tollefson, 1996). These cognitive abnormalities have been assessed by neurocognitive tasks, with varying specificity for frontal cortex areas. Neuropsychological tests such as Wisconsin Card Sorting Test (WCST, Heaton, 1981), Trail Making B (Russell et al., 1970), and Wechsler Adult Intelligence Scale-Revised (WAIS-R) Digit Symbol subtest (Matarazzo, 1972) have been used to identify impairments in set shifting, judgment, attention, con-
66
L.M. Ritter et al. / Schizophrenia Research 68 (2004) 65–73
centration, mental sequencing, planning and anticipation, and concept formation among individuals with schizophrenia (Gold and Harvey, 1993; Goldberg et al., 1988, 1990, 1993). Of the different neurocognitive tasks used in identifying cortical dysfunction in schizophrenia, the WCST has received the most attention. This task is considered a somewhat more specific measure of dorsolateral prefrontal cortex (DLPFC) function. The WCST requires subjects to sort a deck of cards on the basis of a series of unknown categories. To determine the category by which to sort the cards, the individual must use feedback that accompanies each sort (i.e. ‘‘correct’’ or ‘‘wrong’’). As the test progresses and the subject sorts cards correctly, the card-sorting category changes. Executive functions are employed in determining the correct sorting category and adapting when the category changes. Performance on the WCST has been correlated to brain activity in the DLPFC region in neuroimaging studies in both comparison and schizophrenic subjects, and is typically impaired in individuals with schizophrenia (Berman et al., 1986, 1995; Marenco et al., 1993; Weinberger et al., 1986). For example, while performing the WCST, the DLPFC of comparison subjects is physiologically activated above baseline, but it is not activated to the same extent in subjects with schizophrenia (Berman et al., 1992; Weinberger et al., 1986, 1988). Further, individuals with schizophrenia show deficits in performance outcomes of the WCST (i.e. fewer categories completed, more total errors, and more perseverative errors) (Basso et al., 1998; Berman et al., 1997; Koren et al., 1998), and poor performance is typically correlated with negative symptom ratings (Basso et al., 1998; Berman et al., 1997). While some reports suggest that prefrontal cortical dysfunction in schizophrenia is regionally specific (Goldberg et al., 1987; Wilder et al., 1998), other data support generalized bilateral dysfunction (Blanchard and Neale, 1994; Braff et al., 1991; Chelune et al., 1979; Goldberg et al., 1990). The use of tasks with putative regional specificity, such as the Gambling Task (GT) (Bechara et al., 1994), that have been applied to other psychiatric populations (Abbruzzese et al., 1995; Cavedini et al., 2002b) could be useful in addressing the question of regionally specific deficits of the prefrontal cortical functioning in schizophrenia.
The GT was developed as an assessment tool of decision-making process of patients suffering from ventromedial prefrontal cortex (VMPFC) lesions. These patients appear to have intact intellectual function, but demonstrate impairments in personal and social decision making where integration of negative consequences and learning from previous actions are required. The GT is an attempt to mimic real-life decision making in which the patient is presented with options, receives feedback from decisions made, and has the opportunity to incorporate that information into subsequent decisions. Numerous studies have demonstrated the specificity of this task for bilateral VMPFC dysfunction in patients with lesions in this region from stroke or tumor resection (Bechara et al., 1994, 1998, 2000). The GT has been applied to some psychiatric populations to assess VMPFC function. These studies have included subjects with obsessive – compulsive disorder (OCD) (Cavedini et al., 2002b), antisocial personality disorder (Mazas et al., 2000), substance use disorders (alcohol, stimulants, and opiates) (Bartzokis et al., 2000; Bechara et al., 2001; Grant et al., 2000), and pathological gambling (Cavedini et al., 2002a). The studies focusing on decision making in individuals with substance use disorders suggest that active substance use was associated with VMPFC dysfunction. Individuals with antisocial personality disorder also demonstrated impaired performance on the GT. In the OCD and pathological gambling groups, patients with active symptoms (untreated or not responsive to treatment) demonstrated impairment on the task, indicative again of VMPFC impairment. Interestingly, OCD patients who responded to treatment performed like normal controls, suggesting that, for some diagnoses, abnormal GT performance may be a state-dependent phenomenon. Assessment of VMPFC dysfunction in schizophrenia has been limited. To our knowledge, only one study has previously studied the GT in individuals with this disorder. Wilder et al. (1998) used the GT among a group of 12 patients with schizophrenia and 30 normal controls. The results of this study did not show a statistically significant difference on the performance outcomes of the GT between the groups. Due to the small sample size used by Wilder et al., the relative paucity of literature in this area, and the persistent ambiguity about possible global vs. regional
L.M. Ritter et al. / Schizophrenia Research 68 (2004) 65–73
frontocortical dysfunction in schizophrenia, we sought to replicate Wilder et al.’s study by comparing GT and WCST performance among a larger cohort with schizophrenia or schizoaffective disorder vs. comparison subjects. We included those with either schizophrenia or schizoaffective disorder, depressed subtype, as previous studies have suggested that schizoaffective disorder, depressed subtype, may be more closely related to schizophrenia than to mood disorders (Andreasen et al., 1987). We hypothesized that subjects with schizophrenia would show poorer performance on both the GT and WCST than comparison subjects, and that among those with schizophrenia, poor performance on these tasks would correlate with negative symptomatology.
2. Methods This study was approved by the Internal Review Boards of both the VA Ann Arbor Healthcare System and the University of Michigan. Psychiatric and comparison subjects were recruited from the Outpatient Psychiatry Clinics and General Medicine Clinics, respectively, at the VA Ann Arbor Healthcare System. Psychiatric subjects met the following inclusion criteria: (1) age 18 –65; (2) DSM-IV diagnosis by Structured Clinical Interview for DSM-IV (SCID) (First et al., 1997) of chronic schizophrenia or schizoaffective disorder, depressed subtype; (3) psychiatrically stable (no medication changes or psychiatric inpatient admissions) for at least 3 months prior to study. Comparison subjects were between the ages of 18 – 65 and had no DSM-IV diagnosis (based on MINI-Structured Interview) (Sheehan et al., 1998). Potential psychiatric and comparison subjects were excluded if they had an active substance use disorder diagnosis in the 3 months prior to screening, a history of serious head injury, seizure, or other neurological disorder, or stroke. After providing written informed consent, individuals from both groups underwent an assessment battery consisting of the North American Adult Reading Test (an IQ test of premorbid function) (Blair and Spreen, 1989), computerized versions of the GT and WCST, and standardized symptom ratings in a single 1 –2 h session. Symptom rating scales included the Brief Psychiatric Rating Scale (BPRS) (Overall and Gorham, 1962; Woerner et al., 1988) and
67
the Schedule for the Assessment of Negative Symptoms (SANS) (Andreasen, 1982a,b). All subjects received $15 compensation for their participation. 2.1. Gambling Task Subjects turn over cards from one of four different decks (A, B, C, D; 60 cards per deck) until 100 cards are chosen. Before starting the task, the subject is told that the goal of the task is to maximize the profit on a loan of $2000 in play money received at the start of the task. After each card turn, the subject receives a variable play money ‘‘reward’’. Some cards also carry a monetary ‘‘penalty’’. A fixed schedule determines when penalties will occur and how much money is deducted from the subject’s money total. Decks A and B carry higher rewards, but larger penalties, while decks C and D carry lower rewards, but lower penalties. A greater selection of C and D cards will result in a net gain, while a greater selection of A and B cards will result in a net loss. Subjects are free to switch from deck to deck throughout the task and are not told when the task will end. Individuals with VMPFC damage display difficulty anticipating results, typically make riskier choices (A + B>C + D), choose more disadvantage cards (A, B) over the course of the task, and tend to lose more money compared to control subjects (Bechara et al., 1994, 1998, 2000, 2001). 2.2. Wisconsin Card Sort Test The WCST was administered in a computerized format according to the Heaton (1981) protocol in which subjects sort response cards until they have matched six categories or sorted all 128 cards. Cards are matched on the basis of color, shape, and number, and the rule to which cards are matched changes after 10 consecutive correct card sorts. This administration approach differs somewhat from other studies in this area (Weinberger et al., 1988) in that the test is terminated after the successful completion of six categories rather than continuing until all 128 cards have been sorted. The test was designed to measure mental flexibility through the subject’s ability to shift between sets (Berg, 1948). Results typically reported on the WCST include perseverative errors and categories completed. These measures are felt to most directly reflect DLPFC function (Milner, 1963). In the
68
L.M. Ritter et al. / Schizophrenia Research 68 (2004) 65–73
computerized version employed here, subjects used the keyboard to match response cards to the chosen category. 2.3. Data analysis Data analyses of demographic variables, neuropsychological factors, and symptom ratings were done using the chi-square test (ethnicity, history of substance use disorders), group t-tests (age, IQ, income, education, GT and WCST outcome variables), simple regression (symptom rating correlation with neuropsychological test factors), and two-way analysis of variance (ANOVA; group by deck interaction effects). Post-hoc analyses were done using the Newman – Keuls test for significance.
3. Results Demographic variables are summarized in Table 1. There were no significant differences in age, IQ, education, ethnic composition, and annual income (earned and/or disability) between groups. We compared the rates of the past history of substance use disorders between groups, recognizing that previous significant substance use might have long-lasting deleterious effects on cortical function. The rates of Table 1 Subject demographics
Age (years) Gender Ethnicity/race
IQ Income (earned and disability) Education (years) History of drug/alcohol use disorder Cocaine Alcohol
Psychiatric subjects N = 20
Comparison subjects N = 15
48.5 F 6.0 100% Male 5% African American; 95% Caucasian 104.8 F 10.0 $17,098 F 9582
47.1 F 10.2 100% Male 13% African American; 87% Caucasian 100.2 F 12.3 $25,464 F 18,889
13.8 F 1.70
13.3 F 1.45
5/20 11/20
1/15 3/15*
Demographic comparisons between psychiatric and comparison subjects were nonsignificant except for history of alcohol use disorder. * Chi-square 4.38; df = 1; p < 0.05.
a past history of cocaine abuse/dependence were not significantly different between the groups (Fisher’s Exact p>0.20). Rates of a past history of alcohol dependence, however, were significantly different. Fifty-five percent of the psychiatric subjects had a past history of alcohol abuse or dependence vs. 33% of the comparison group (Fisher’s Exact p < 0.05). Analysis of GT and WCST data including history of alcohol abuse/dependence as a covariate did not alter the results in any way. Hence, only the straight group comparisons are presented. In the group of subjects with psychiatric illness, 15 subjects had a DSM-IV diagnosis of schizophrenia and 5 subjects were diagnosed as schizoaffective. Seventeen were prescribed atypical antipsychotic agents, and the remaining were given typical antipsychotic agents. Lastly, by means of self-report, the average number of years psychiatrically ill at the time of study was 25.8 F 5.1 years. On multiple GT outcome measures, individuals with schizophrenia performed significantly worse than comparison subjects (Table 2). Individuals with schizophrenia chose significantly more disadvantaged than advantaged cards ( F = 4.91; df = 1; p < 0.04) and had a significantly larger differential between disadvantaged and advantaged cards selected ( F = 4.91; df = 1; p < 0.04). While subjects with schizophrenia gained significantly more money ( F = 4.10; df = 1; p = 0.05), they also lost significantly more money ( F = 4.55; df = 1; p < 0.05). A deck by group interaction ( F = 3.68; df = 3; p = 0.01) resulted from individuals with schizophrenia choosing more Deck B cards than comparison individuals choosing Deck A cards ( p < 0.001) and comparison subjects choosing Deck D cards more than subjects with schizophrenia choosing Deck A cards ( < 0.02) (Fig. 1). Examining chronological card choice in blocks of 20 cards, the subjects with schizophrenia began picking more disadvantaged cards, and only slowly improved their selection mix compared to the control group (Fig. 2). A main effect for diagnosis was found ( F = 7.56; df = 1; p < 0.007), but there was not a diagnosis by card block interaction. Significant differences were not found between the two groups on WCST outcome variables (Table 2), although both groups performed poorly. Additionally, performance on the GT was not correlated to WCST performance for either group.
L.M. Ritter et al. / Schizophrenia Research 68 (2004) 65–73
69
Table 2 GT and WCST outcome measures
Gambling task
Wisconsin card sorting test
advantaged minus disadvantaged cards advantaged cards selected disadvantaged cards selected money gained money lost net gain % error % perseverative error % conceptual level categories completed failure to maintain set
Psychiatric subjects N = 20
Comparison subjects N = 15
5.2 F 19.7* 47.4 F 9.84* 52.6 F 9.84* $8388 F 632y $9643 F 1316* $1156 F 783* 35.9% F 20.0 18.8% + 15.0 53.0% F 25.8 3.5 F 2.4 1.8 F 1.8
14.4 F 32.5 57.2 F 16.25 42.8 F 16.25 $7870 F 882 $8390 F 2147 $480 F 1333 36.5% F 19.7 19.7% F 15.9 53.3% F 27.4 3.8 F 2.3 1.4 F 1.4
Wisconsin Card Sorting Test and Gambling Task results for psychiatric and comparison subjects. Significance between group differences is indicated. * p < 0.05. y p = 0.05.
Not surprisingly, subjects with schizophrenia were significantly more symptomatic on the BPRS and SANS standardized rating scales (BPRS: comparison subjects 20.3 F 1.9 vs. schizophrenic subjects 32.6 F 5.1, p < 0.0001; SANS: comparison subjects 4.1 F 3.6 vs. schizophrenic subjects 21.4 F 7.5, p < 0.0001). Total scores on both the BPRS and SANS did not correlate to GT or WCST performance for the schizophrenic group. However, selected BPRS and SANS
subscale scores did correlate to WCST performance for individuals with schizophrenia. The hostility subscale of the BPRS was negatively correlated to percent error (r = 0.489; df = 19; p < 0.03) and positively correlated with percent conceptual level (r = 0.484; df = 19; p < 0.04). None of the other BPRS subscales (i.e. positive, negative, or activation) were correlated to performance on either the WCST or GT outcome variables. The SANS subscales of attention
Fig. 1. Card choices of psychiatric and comparison subjects during the 100-card task.
Fig. 2. Individuals with schizophrenia shifted more slowly than comparison subjects towards choosing more advantaged cards.
70
L.M. Ritter et al. / Schizophrenia Research 68 (2004) 65–73
(r = 0.445; df = 19; p < 0.05) and alogia ( p < 0.01) were positively correlated to the failure to maintain set WCST variable for individuals with schizophrenia.
4. Discussion In this study, we found that based on performance on a measure felt to reflect VMPFC function, subjects with schizophrenia performed poorly compared to a nonpsychiatrically ill comparison group. Gambling Task performance was not related to subjects’ psychiatric symptoms, or performance on the WCST, a measure more reflective of DLPFC function. Interestingly, both subjects with schizophrenia and the comparison group demonstrated impaired performance on the WCST. Further, there was some relationship between negative symptom subscale scores and WCST in the group with schizophrenia. While the differences between groups by card block were not significant, comparison subjects moved quickly to pick more advantageous than disadvantageous cards by the second block of 20 cards (Fig. 2) and maintained that bias throughout the rest of the task. Subjects with schizophrenia began with a greater preference for more disadvantageous cards, and took until block 4 (cards 61– 80) to demonstrate a shift in preference to more advantageous cards. In contrast, among subjects with VMPFC lesions, choice of disadvantageous cards persists throughout the task (Bechara et al., 2001) suggesting that they never adjust their card choice based on feedback from previous choices. In studies examining GT performance in other cohorts with psychiatric diagnoses, pathological gamblers (Cavedini et al., 2002a) and individuals OCD (Cavedini et al., 2002b) appear not to shift to more advantageous cards (like those with VMPFC lesions), while those with antisocial personality (Mazas et al., 2000) or substance use disorders (Bechara et al., 2001), like our schizophrenic subjects, slowly but eventually shift to more advantageous cards. Subjects with VMPFC cortical lesions appear insensitive to future consequences and are guided more by the immediate possibility of reward (Bechara et al., 2000). Our subjects with schizophrenia appear more sensitive to future consequences, but take a longer time to incorporate and act on the information fed
back from each card selection. While processing of gains and losses clearly involves the VMPFC, there is also evidence to suggest that the anterior cingulate cortex is also involved in this processing. Abnormalities in ACC function have been implicated in schizophrenia (Carter et al., 2001; Suhara et al., 2002), and the ACC has been shown to be involved in a similar monetary decision-making task (Gehring and Willoughby, 2002). Hence, it is possible that the impaired GT performance in schizophrenia represents abnormal function in other structures connected to the VMPFC cortex and involved in the complex decision-making process tested by the GT. We did not anticipate that we would be unable to find a difference in WCST performance between our subjects with schizophrenia and a well-matched comparison group. Our lack of a difference between the two groups would appear to be accounted for by the poorer than expected performance in the comparison group compared to controls in other reports (Bechara et al., 2001; Koren et al., 1998). At the same time, the GT performance of our comparison subjects is quite comparable to that of control groups from other studies. Similar to Cavedini et al. (2002a), our comparison subjects selected comparable numbers of advantageous and disadvantageous cards. Additionally, our comparison subjects had a positive advantageous minus disadvantageous differential after the first block of 20 cards and continued to be positive for the remainder of the task, following the same performance pattern of the controls in Bechara et al. (2001) and Cavedini et al. (2002a). Based on our results, the effect size for the GT findings ranged from 0.7 to 0.8. These GT effects sizes are similar to those calculated from data presented in Grant et al. (2000), but are 30 –50% greater than those calculated from data reported by Wilder et al. (1998), who studied a smaller group with schizophrenia (N = 12) compared to a larger control group (N = 30). In contrast, the effect sizes for our Wisconsin Card Sort Findings were much smaller, ranging from less than 0.1 to 0.3, despite variances that were comparable to studies of nonschizophrenic substance abusers (Grant et al., 2000) as well as those with schizophrenia (Saoud et al., 1998). Hence, we are confident that the lack of a difference in the WCST results was not the result of being underpowered.
L.M. Ritter et al. / Schizophrenia Research 68 (2004) 65–73
Although we did not find a difference in WCST performance between our schizophrenic and comparison groups, we did find that positive and negative symptom subscale scores were correlated with WCST measures for individuals with schizophrenia. This finding is in agreement with previous reports that suggest that increased negative symptomatology is associated with poor WCST performance (Basso et al., 1998; Berman et al., 1997). We did not find that symptom subscales were correlated to GT performance for this same group, again suggesting that performance of these two neuropsychological tests may map to different prefrontal cortical areas, and that the negative symptoms of schizophrenia are likely not the result of homogenous impairment of the prefrontal cortex. Several limitations to these data should be considered. Our psychiatric sample was not diagnostically homogenous. Differences in GT performance between those with schizophrenia vs. schizoaffective disorder, depressed subtype, could not be excluded with our sample size, and might have affected our results. Future studies might specifically compare GT performance between these two diagnoses. In addition, schizophrenia and schizoaffective disorder are by no means purely frontal lobe disorders. The broader array of cortical and subcortical dysfunction seen in these conditions likely also impacts on performance on neurocognitive tests like the GT and WCST. Likewise, the GT or WCST are not purely tests of frontal cortical function. Indeed, some evidence suggests that the specificity of the WCST for frontal lobe damage is limited (Anderson et al., 1991). While our data support the notion that regional cortical dysfunction can be discerned in schizophrenia, further investigation is needed. Finally, our use of the keyboard response, computerized version of the WSCT, might have affected our results. One study in normal subjects (Feldstein et al., 1999) indicates that performance on various computer versions of the WCST can differ from that on the standard (manual) version (Feldstein et al., 1999). Indeed, the ‘‘keyboard’’ computerized version, similar to the one we used in our study, was found to result in lower scores when compared to the standard version. Nonetheless, both groups in our study used the same version, and so should have been comparably affected. Moreover, the scores of our comparison group were remarkably close to those of
71
the psychiatric subjects, and lower than control or comparison groups in other studies regardless of version administered. In summary, we conclude that GT performance in schizophrenia is impaired. This may reflect impairment in VMPFC function and/or abnormalities in the cortico-subcortical circuitry putatively involved in this decision-making task. At the same time, GT performance was unrelated to severity of schizophrenic symptomatology, or a measure reflective, in part, of DLPFC function. This suggests that regionally specific deficits in schizophrenia appear more likely than global prefrontal cortical dysfunction. Acknowledgements The authors would like to thank Dr. Antoine Bechara for providing a computerized version of the Gambling Task and helpful consultation. We also thank Ingrid Lund for her help with data acquisition. Earlier versions of this work were presented at the 40th annual meeting of the American College of Neuropsychopharmocology, Kona, HI, December 2001 and the 57th annual meeting of the Society of Biological Psychiatry, Philadelphia, PA, May 2002. References Abbruzzese, M., Bellodi, L., Ferri, S., Scarone, S., 1995. Frontal lobe dysfunction in schizophrenia and obsessive – compulsive disorder: a neuropsychological study. Brain Cogn. 27, 202 – 212. Anderson, S.W., Damasio, H., Jones, R.D., Tranel, D., 1991. Wisconsin Card Sorting Test performance as a measure of frontal lobe damage. J. Clin. Exp. Neuropsychol. 13, 909 – 922. Andreasen, N.C., 1982a. Negative symptoms in schizophrenia: definition and reliability. Arch. Gen. Psychiatry 38, 784 – 788. Andreasen, N.C., 1982b. Negative versus positive schizophrenia: definition and validation. Arch. Gen. Psychiatry 38, 788 – 794. Andreasen, N.C., Rice, J., Endicott, J., Coryell, W., Grove, W.M., Reich, T., 1987. Familial rates of affective disorder. A report from the National Institute of Mental Health Collaborative Study. Arch. Gen. Psychiatry 45, 461 – 469. Bartzokis, G., Lu, P.H., Beckson, M., Rapoport, R., Grant, S., Wiseman, E.J., London, E.D., 2000. Abstinence from cocaine reduces high-risk gambling responses on a Gambling Task. Neuropsychopharmacology 22, 102 – 103. Basso, M.R., Nasrallah, H.A., Olson, S.C., Bornstein, R.A., 1998. Neuropsychological correlates of negative, disorganized and psychotic symptoms in schizophrenia. Schizophr. Res. 31, 99 – 111.
72
L.M. Ritter et al. / Schizophrenia Research 68 (2004) 65–73
Bechara, A., Damasio, A.R., Damasio, H., Anderson, S.W., 1994. Insensitivity to future consequences following damage to human prefrontal cortex. Cognition 50, 7 – 15. Bechara, A., Damasio, H., Tranel, D., Anderson, S.W., 1998. Dissociation of working memory from decision-making within the human prefrontal cortex. J. Neurosci. 18, 428 – 437. Bechara, A., Tranel, D., Damasio, H., 2000. Characterization of the decision-making deficit of patients with ventromedial prefrontal cortex lesions. Brain 123, 2189 – 2202. Bechara, A., Dolan, S., Denburg, N., Hindes, A., Anderson, S.W., Nathan, P.E., 2001. Decision-making deficits, linked to a dysfunctional ventromedial prefrontal cortex, revealed in alcohol and stimulant abusers. Neuropsychologia 39, 376 – 389. Berg, E.A., 1948. A simple objective technique for measuring flexibility in thinking. J. Gen. Psych. 19, 15 – 22. Berman, K.F., Zec, R.F., Weinberger, D.R., 1986. Physiological dysfunction of dorsolateral prefrontal cortex in schizophrenia: II. Role of neuroleptic treatment, attention, and mental effort. Arch. Gen. Psychiatry 43, 126 – 135. Berman, K.F., Torrey, E.F., Daniel, D.G., Weinberger, D.R., 1992. Regional cortical blood flow in monozygotic twins discordant and concordant for schizophrenia. Arch. Gen. Psychiatry 49, 927 – 934. Berman, K.F., Ostrem, J.L., Randolph, C., Gold, J., Goldberg, T.E., Coppola, R., Carson, R.E., Herscovitch, P., Weinberger, D.R., 1995. Physiological activation of cortical networks during performance of the Wisconsin Card Sorting Test: a positron emission tomography study. Neuropsychologia 33 (8), 1027 – 1046. Berman, I., Viegner, B., Mason, A., Allan, E., Pappas, D., Green, A.I., 1997. Differential relationships between positive and negative symptoms and neuropsychological deficits in schizophrenia. Schizophr. Res. 25, 1 – 10. Blair, J.R., Spreen, O., 1989. Predicting premorbid IQ: a revision of the National Adult Reading Test. Clin. Neuropsychol. 3, 129 – 136. Blanchard, J.J., Neale, J.M., 1994. The neuropsychological signature of schizophrenia: generalized or differential deficit? Am. J. Psychiatry 151, 40 – 48. Braff, D.L., Heaton, R., Kuck, J., Cullum, M., Moranville, J., Grant, I., Zisook, S., 1991. The generalized pattern of neuropsychological deficits in outpatients with chronic schizophrenia with heterogeneous Wisconsin Card Sorting Test results. Arch. Gen. Psychiatry 48, 891 – 898. Carter, C.S., MacDonald, A.W., Ross, L.L., Stenger, V.A., 2001. Anterior cingulate cortex activity and impaired self-monitoring of performance in patients with schizophrenia: an event-related fMRI study. Am. J. Psychiatry 158 (9), 1423 – 1428. Cavedini, P., Giovanna, R., Keller, R., D’Annucci, A., Bellodi, L., 2002a. Frontal lobe dysfunction in pathological gambling patients. Biol. Psychiatry 51, 334 – 341. Cavedini, P., Riboldi, G., D’Annucci, A., Belotti, P., Cisima, M., Bellodi, L., 2002b. Decision-making heterogeneity in obsessive – compulsive disorder: ventromedial prefrontal cortex function predicts different treatment outcomes. Neuropsychologia 40 (2), 205 – 211. Chelune, G.J., Heaton, R.K., Leham, R.A.W., Robinson, A., 1979. Level versus pattern of neuropsychological performance among
schizophrenic and diffusely brain-damaged patients. J. Consult. Clin. Psychol. 47, 155 – 163. Feldstein, S.N., Keller, F.R., Portman, R.E., Durham, R.L., Klebe, K.J., Hasker, P.D., 1999. A comparison of computerized and standard versions of the Wisconsin Card Sorting Test. Clin. Neuropsychol. 13, 303 – 313. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 1997. Structured Clinical Interview for DSM-IV Axis I DisordersClinician Version (SCID-CV). American Psychiatric Press, Washington, DC. Gehring, W.J., Willoughby, A.R., 2002. The medial frontal cortex and the rapid processing of monetary gains and losses. Science 295, 2279 – 2282. Gold, J.M., Harvey, P.D., 1993. Cognitive deficits in schizophrenia. Psychiatr. Clin. North Am. 16, 295 – 312. Goldberg, T.E., Weinberger, D.R., Berman, K.F., Pliskin, N.H., Podd, M.H., 1987. Further evidence for dementia of the prefrontal type in schizophrenia? 1987. A controlled study of teaching Wisconsin Card Sorting Test. Arch. Gen. Psychiatry 44, 1008 – 1014. Goldberg, T.E., Kelsoe, J.R., Weinberger, D.R., Pliskin, N.H., Kirwin, P.D., Berman, K.F., 1988. Performance of schizophrenic patients on putative neuropsychological tests of frontal lobe function. Int. J. Neurosci. 42, 51 – 58. Goldberg, T.E., Ragland, J.D., Torrey, E.F., Gold, J.M., Bigelow, L.B., Weinberger, D.R., 1990. Neuropsychological assessment of monozygotic twins discordant for schizophrenia. Arch. Gen. Psychiatry 47, 1066 – 1072. Goldberg, T.E., Torrey, E.F., Gold, J.M., Ragland, J.D., Bigelow, L.B., Weinberger, D.R., 1993. Learning and memory in monozygotic twins discordant for schizophrenia. Psychol. Med. 23, 71 – 85. Grant, S., Contoreggi, C., London, E.D., 2000. Drug abusers show impaired performance in a laboratory test of decision making. Neuropsychologia 38, 1180 – 1187. Heaton, R.K., 1981. Wisconsin Card Sorting Test Manual. PAR, Odessa, FL. Koren, D., Harrision, W.S., Kremen, W.S., Goldstein, J.M., Faraone, S.V., Seidman, L.J., Lyons, M.J., Caplan, B., Tsuang, M.T., 1998. Factor structure of the Wisconsin Card Sorting Test: dimensions of deficit in schizophrenia. Neuropsychology 12, 289 – 302. Marenco, S., Coppola, R., Daniel, D.G., Zigun, J.R., Weinberger, D.R., 1993. Regional cerebral blood flow during the Wisconsin Card Sorting Test in normal subjects as studied by xenon-133 dynamic SPECT: comparison of absolute values, percent distribution values, and covariance analysis. Psychiatry Res. 50, 177 – 192. Matarazzo, J.D., 1972. Wechsler’s Measurement and Appraisal of Adult Intelligence, 5th ed. Williams & Wilkins, Baltimore. Mazas, C., Finn, P.R., Steinmetz, J.E., 2000. Decision-making biases, antisocial personality, and early-onset alcoholism. Alcohol., Clin. Exp. Res. 24, 1036 – 1040. Milner, B., 1963. Effects of different brain lesions on card sorting. Arch. Neurol. 9, 100 – 110. Overall, J.E., Gorham, D.R., 1962. The Brief Psychiatric Rating Scale. Psychol. Rep. 10, 799 – 812.
L.M. Ritter et al. / Schizophrenia Research 68 (2004) 65–73 Russell, E.W., Neuringer, C., Goldstein, G., 1970. Assessment of Brain Damage: A Neuropsychological Key Approach. Wiley, New York. Saoud, M., Hueber, T., Mandran, H., Dalery, J., d’Amato, T., 1998. Olfactory identification deficiency and WCST performance in men with schizophrenia. Psychiatry Res. 81, 251 – 257. Sheehan, D.V., Lecrubier, Y., Sheehan, K.H., Amorin, P., Janavs, J., Weiller, E., Hergueta, T., Baker, R., Dunbar, G.C., 1998. The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic interview for DSM-IV and ICD-10. J. Clin. Psychiatry 59, 22 – 33. Suhara, T., Okubo, Y., Yasuno, F., Sudo, Y., Inoue, M., Ichimiya, T., Nakashima, Y., Nakayama, K., Tanada, S., Suzuki, K., Halldin, C., Farde, L., 2002. Decreased dopamine D2 receptor binding in the anterior cingulate cortex in schizophrenia. Arch. Gen. Psychiatry 59, 25 – 30. Tollefson, G.D., 1996. Cognitive function in schizophrenic patients. J. Clin. Psychiatry 57 (Suppl. 11), 31 – 39.
73
Weinberger, D.R., Berman, K.F., Zec, R.F., 1986. Physiological dysfunction of dorsolateral prefrontal cortex in schizophrenia: I. Regional cerebral blood flow (rCBF) evidence. Arch. Gen. Psychiatry 43, 114 – 125. Weinberger, D.R., Berman, K.F., Illowsky, B.P., 1988. Physiological dysfunction of dorsolateral prefrontal cortex in schizophrenia: III. A new cohort and evidence for a monoaminergic mechanism. Arch. Gen. Psychiatry 45, 609 – 615. Wilder, K.E., Weinberger, D.R., Goldberg, T.E., 1998. Operant conditioning and orbitofrontal cortex in schizophrenic patients: unexpected evidence for intact functioning. Schizophr. Res. 30, 169 – 174. Woerner, M.G., Mannuzza, S., Kane, J.M., 1988. Anchoring the BPRS: an aid to improved reliability. Psychopharmacol. Bull. 24, 112 – 117.