- Email: [email protected]
Exploring logical reasoning abilities in schizophrenia patients
Schizophrenia Research 127 (2011) 178–180
Contents lists available at ScienceDirect
Schizophrenia Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c h r e s
Exploring logical reasoning abilities in schizophrenia patients Dario Mirian a,⁎, R. Walter Heinrichs a, Stephanie McDermid Vaz b a b
Department of Psychology, York University, 4700 Keele Street, Toronto, Ontario Canada M3J 1P3 Cleghorn Early Intervention in Psychosis Program, St Joseph's Healthcare Hamilton, 25 Charlton Avenue East, Suite 703, Hamilton, Ontario, Canada L8N 1Y2
a r t i c l e
i n f o
Article history: Received 29 October 2010 Received in revised form 11 January 2011 Accepted 13 January 2011 Keywords: Deductive logic Cognition Symptoms Schizophrenia Performance
a b s t r a c t Objective: To assess deductive reasoning in schizophrenia patients with special reference to whether accuracy varies across type of stimulus problem. Previous research suggests that patients, unlike healthy controls, are insensitive to emotionally provocative (salient) problem content. Method: A syllogistic reasoning task consisting of five argument types varying in salience, congruence with commonly held beliefs and meaningfulness was administered along with standard intellectual and symptom measures to 25 schizophrenia patients and 26 healthy control participants. Results: Patients performed below control participants in all reasoning task conditions, but group differences were non significant after controlling for IQ. There were no significant interactions between group and argument type in terms of reasoning accuracy and both patients and controls performed better when reasoning with belief congruent material. In addition, no relation between deduction and paranoid symptoms was found. Conclusion: Formal deductive reasoning abnormalities in schizophrenia are a reflection of the broadly based cognitive impairment documented in the illness. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Although schizophrenia involves severe disturbances in thinking, the formal study of reasoning capacities in patients has not been pursued comprehensively. Research has primarily focused on reasoning biases in relation to symptoms such as delusions (Corcoran, 2003; John and Dodgson, 1994). Patients with delusions show abnormalities and biases in probabilistic reasoning and “jump to conclusions” by making decisions with minimal information (Garety and Freeman, 1999; Garety et al., 2005). In addition, their attributional reasoning about negative events shows a bias towards assigning causes to external factors (Jolley et al., 2006). In contrast, deductive reasoning, which involves evaluating an argument drawn from a given premise, has received relatively little attention (Speechley et al., 2009). The validity of an argument and its conclusion is contingent on the validity of the premises preceding it (Evans and St, 2005). Moreover, validity judgments about an argument are influenced by pre-existing beliefs regarding its conclusion (Goel and Dolan, 2003). These properties of deduction imply potentially fruitful avenues of investigation in relation to the psychopathology of schizophrenia.
⁎ Corresponding author. c/o Dr. R. W. Heinrichs York University, Department of Psychology, 4700 Keele Street, Toronto, ON, Canada, M3J 1P3. Tel.: + 1 647 298 8222. E-mail addresses: [email protected], [email protected] (D. Mirian), [email protected] (R.W. Heinrichs), [email protected] (S.M. Vaz). 0920-9964/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2011.01.007
Previous research suggests that schizophrenia patients may have difficulty with deductive reasoning in general and, in contrast with healthy controls, show equivalent performance in emotionally provocative or salient and non-salient problem variations (Goel et al., 2004; Ho, 1974; Mujica-Parodi et al., 2000; Speechley et al., 2009). Goel et al. (2004), administered syllogisms to schizophrenia patients and controls and found that patients performed equally poorly across all problem types, whereas healthy controls were most proficient in solving emotionally neutral or “non-salient” problems. However, a critical limitation of this study was the presence of a floor effect in the patient group whereby performance approximated chance levels. Another drawback was their study did not assess incongruent and congruent problem types. This distinction reflects reasoning with material that is consistent with prior knowledge and beliefs as contrasted with material consistent with such beliefs. Speechley et al. (2009), showed that schizophrenia patients with delusions were impaired in the incongruent trials. However, the sample sizes were small and it is unclear whether symptom profiles and intensity figure in this impairment. In light of these considerations, we studied deductive reasoning in schizophrenia patients and healthy participants with problem types varying in salience, congruence and meaningfulness. We were interested specifically in the question of whether deductive reasoning is deficient across problem types or occurs more selectively. In addition, we examined possible relations between symptom severity and reasoning performance.
D. Mirian et al. / Schizophrenia Research 127 (2011) 178–180
2. Method
Table 2 Adjusted mean accuracy scores and standard deviations for belief condition.
2.1. Participants The sample comprised 26 healthy control participants and 25 patients meeting DSM-IV-R (American Psychiatric Association Task Force on DSM-IV, 2000) criteria for schizophrenia (80%) or schizoaffective disorder (20%) based on the Structured Clinical Interview for DSM-IV (SCID; First et al., 1996). Table 1 presents the demographic characteristics of the sample. Control participants were recruited from the greater Toronto area and patients were recruited from outpatient settings that included the Hamilton Program for Schizophrenia, the Community Schizophrenia Service and the Cleghorn Program (St. Joseph's Healthcare Hamilton). All patients were receiving antipsychotic medication during the data collection period. The study received approval by the institutional review board at York University and St. Joseph's Healthcare affiliated research sites. 2.2. Clinical and cognitive measures Patients' symptoms were measured with the Positive, Negative and General Psychopathology subscales of the Positive and Negative Syndrome Scale (PANSS; Kay et al., 1987; Opler et al., 1999). General cognitive abilities of both patients and controls were indexed with the Vocabulary and Matrix Reasoning subtests of the Wechsler Abbreviated Scale of Intelligence (WASI; Psychological Corporation, 1999). Syllo-
Table 1 Demographic characteristics of sample. n
%
Gender Male Female
18 33
35.3 64.7
Marital status Single Married Divorced Other
39 7 4 1
76.5 13.7 7.8 2.0
Employment status Full-time Part-time Unemployed Volunteer Student
4 11 22 1 13
7.8 21.6 43.1 2.0 25.5
Hospitalization in past 2 years None 1–2 times 3+2
17 2 3.9
33.3 3.9
Criminal record Yes No
6 15
28.6 71.4
Smoking Yes No
14 37
27.5 72.5
Alcohol use Yes No
25 26
49.0 51.0
Drug use Yes No
8 43
15.7 84.3
M
Patients Controls
M SD M SD
Non-belief accuracy (%)
Belief salient accuracy (%)
Belief-non salient accuracy (%)
54.67 15.61 71.15 19.18
62.67 18.18 78.85 17.36
65.33 23.03 73.72 19.39
gisms, which include two premises and a conclusion, were presented to assess deductive reasoning abilities and beliefs. Participants were initially given verbal instructions and practice in deductive problemsolving with neutral stimuli. The experimental task included 18 syllogisms: 6 belief-salient (e.g., No drunks are saints; some Irishmen are saints; therefore some Irishmen are not drunks), 6 belief-non-salient (e.g., All crunchy tubers are vegetables; some carrots are crunchy tubers; therefore some carrots are not vegetables) and 6 non-belief (e.g., Some ploms are dihedric; no moduleses are ploms; therefore some moduleses are dihedric). Within the 12 belief-based syllogisms, 6 problems were belief-congruent (e.g., All flying birds have feathers; no people have feathers; therefore some people are flying birds) and 6 were beliefincongruent (e.g., All handicapped are capable; all amputees are handicapped; therefore some amputees are not capable). In each case participants were required to indicate whether the conclusion was correct or incorrect. The syllogisms were selected from an item bank incorporating a range of difficulty (Dickstein, 1978). In order to avoid floor effects, syllogisms of relatively low difficulty, correctly answered by 90% of healthy controls in the original item bank, were chosen. Participants were allowed a maximum of 60 s to solve each item.
SD
3. Results Pearson correlations were calculated to index relationships between percentage correct in the reasoning tasks and five subscales from the PANSS (thought disturbance, activation, anergia, paranoid belligerence, and depression). One statistically significant correlation was found: correct responses on non-belief items were negatively related to depression scores, r(23)= −.43, pb.05. Two analyses of covariance (ANCOVA) with repeated measures were carried out on reasoning test accuracy data. The first ANCOVA was done to test for main effects of group (schizophrenia vs. control) and belief problem type (belief-salient, belief-non-salient, non-belief) and for group by belief interaction, while controlling for IQ. Table 2 displays the covariate-adjusted descriptive statistics for each group under each belief condition. The assumptions of ANCOVA (i.e., homogeneity-of-regression (slopes), sphericity, and homogeneity of variance) were evaluated and indicated no violations. After adjusting for IQ, there were no significant group differences across belief problem types F(1, 47) = .21, p = .65. There was also no main effect of belief problem type on accuracy scores F(2, 94) = 1.12, p =.33, and the interaction of group and belief problem type was nonsignificant, F(2, 94) = .66, p = 0.52. The second ANCOVA was carried out to test for the main effects of group and congruence problem type (i.e., congruent and incongruent). Table 3 displays the adjusted descriptive statistics for each group under each trail condition. Again, all statistical assumptions were satisfied. After controlling for IQ, there
Table 3 Adjusted mean accuracy scores and standard deviations for trial conditions.
Patients Age Education Age of psychiatric diagnosis
179
37.47 14.33 22.05
13.60 2.89 4.90
Controls
M SD M SD
Incongruent accuracy (%)
Congruent accuracy (%)
44.00 27.17 74.36 22.72
82.00 15.15 78.21 18.11
180
D. Mirian et al. / Schizophrenia Research 127 (2011) 178–180
Table 4 Unadjusted mean accuracy scores and standard deviations for belief condition.
Patients Controls
M SD M SD
Non-belief accuracy (%)
Belief salient accuracy (%)
Non-salient belief accuracy (%)
54.67 15.61 71.15 19.18
62.67 18.18 78.85 17.36
65.33 23.03 73.72 19.39
Table 5 Unadjusted mean accuracy scores and standard deviations for trial conditions.
Patients Controls
M SD M SD
Incongruent accuracy (%)
Congruent accuracy (%)
44.00 27.17 74.36 22.72
82.00 15.15 78.21 18.11
was no significant difference between the two groups across problem type F(1, 47) = .06, p = .81. However, there was a main effect of congruence problem type F(1, 47) = 10.17, p = .003. Specifically, both patients and controls were more accurate in solving congruent (M = 80.07, SD = 16.67) than incongruent problems (M = 59.48, SD = 29.11). For comparative purposes, Tables 4 and 5 present reasoning performance scores unadjusted for IQ. 4. Discussion The present study examined deductive logic in patients with schizophrenia and in healthy comparison subjects by varying the content of syllogisms, while statistically controlling for general intellectual ability. Results indicate that schizophrenia patients are indistinguishable from healthy controls in terms of deductive problem solving once the contribution of general ability to reasoning accuracy is removed. We found no support for the conjecture that the illness involves a selective insensitivity to emotionally provocative problem content. Moreover, belief congruence, or reasoning with material consistent with general knowledge about the world, facilitates problem solving for both patients and controls. There is no evidence that schizophrenia involves insensitivity to this kind of general knowledge and belief system or any preferential processing of idiosyncratic and norm-divergent problem content. Accordingly, it seems likely that deductive reasoning impairments in the illness reflect the broadlybased cognitive dysfunction described in the literature over the past two decades (e.g. Heinrichs and Zakzanis, 1998) and not a specific weakness in deduction. Furthermore, we found no evidence that defective deductive reasoning with any type of problem content associates with the kind of implausible beliefs expressed in persecutory delusions, although some aspects of problem content may vary with depression. There are reports suggesting that depression leads to decreased cognitive ability especially a lack of motivation in the schizophrenia population (Shah et al., 1999). Overall, therefore, the present study does not support findings by Goel et al. (2004) and Speechley et al. (2009) suggesting a specific weakness in deduction in schizophrenia and an abnormal sensitivity to problem content. Poor deduction skills probably reflect the general weakness in cognitive performance observed in the illness rather than any special difficulty in evaluating the logic and validity of arguments. Future studies with larger sample sizes and
matched case/control designs are warranted. In addition, prospective studies should include a larger number of syllogisms to assess broader deductive reasoning abilities. Role of funding source Funding for this study was provided by the Ontario Mental Health Foundation (OMHF) and by the Community Schizophrenia Vocational Rehabilitation Foundation (CSVR). The OMHF and CSVR Foundation had no role in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication. Contributors W. Heinrichs designed the study and wrote the protocol. S. McDermid Vaz collected the data. W. Heinrichs and D. Mirian conducted the analyses. All authors contributed to and approved the final manuscript. Conflicts of interest There are no conflicts of interest with respect to this manuscript. Acknowledgements We express our appreciation to Ashley Oman, B.A., and the staff of the Community Schizophrenia Service, the Cleghorn Program, St. Joseph's Healthcare Hamilton, the Hamilton Program for Schizophrenia, and the Canadian Mental Health Association Toronto Branch.
References American Psychiatric Association Task Force on DSM-IV-R, 2000. Diagnostic and Statistical Manual of Mental Disorders (4th ed revised). American Psychiatric Association, Washington. Corcoran, R., 2003. Inductive reasoning and the understanding of intention in schizophrenia. Cognitive Neuropsychiatry 8, 223–235. Dickstein, L.S., 1978. The effect of figure on syllogistic reasoning. Memory & Cognition 6, 76–83. Evans, J.St.B.T., 2005. Deductive Reasoning. In: Holyoak, K.J., Morrison, R.G. (Eds.), The Cambridge Handbook of Thinking and Reasoning. Cambridge University Press, New York, pp. 169–184. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 1996. Structured Clinical Interview for DSM-IV Axis I Disorders. Non-Patient Edition (SCID-I/NP). New York: New York State Psychiatric Institute, Biometrics Research. Garety, P.A., Freeman, D., 1999. Cognitive approaches to delusions: a critical review of theories and evidence. The British Journal of Clinical Psychology 38, 113–154. Garety, P.A., Freeman, D., Jolley, S., Bebbington, P.E., Kuipers, E., Dunn, G., Fowler, D.G., Dudley, R., 2005. Reasoning, emotions, and delusional conviction in psychosis. Journal of Abnormal Psychology 114, 373–384. Goel, V., Dolan, R.J., 2003. Reciprocal neural response within lateral and ventral medial prefrontal cortex during hot and cold reasoning. Neuroimage 20, 2314–2321. Goel, V., Bartolo, A., St. Clair, D., Venneri, A., 2004. Logical reasoning deficits in schizophrenia. Schizophrenia Research 66, 87–88. Heinrichs, R.W., Zakzanis, K.K., 1998. Neurocognitive deficit in schizophrenia: a quantitative review of the evidence. Neuropsychology 12, 426–445. Ho, D.F., 1974. Modern logic and schizophrenia thinking. Genetic Psychology Monographs 89, 145–165. John, C., Dodgson, G., 1994. Inductive reasoning in delusional thought. Journal of Mental Health 3, 31–49. Jolley, S., Garety, P., Bebbington, P., Graham, D., Freeman, D., Kuipers, E., Fowler, D., Hemsley, D., 2006. Attributional style in psychosis — the role of affect and belief type. Behaviour Research and Therapy 44, 1597–1607. Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The Positive and Negative Syndrome Scale (PANSS) for Schizophrenia. Schizophrenia Bulletin 12, 261–276. Mujica-Parodi, L.R., Malaspina, D., Sackeim, H.A., 2000. Logical processing, affect and delusional thought in schizophrenia. Harvard Review of Psychiatry 8, 73–83. Opler, L.A., Kay, S.R., Lindenmayer, J.P., Fiszbeln, A., 1999. Structured Clinical Interview for the Positive and Negative Syndrome Scale (SCI-PANSS). Multi-Health Systems, North Tonawanda: New York. Psychological Corporation, 1999. Wechsler Abbreviated Scale of Intelligence (WASI) Manual. San Antonio, TX: Author. Shah, P.J., O'Carroll, R.R., Rogers, A., Moffoot, A.P.R., Ebmeier, K.P., 1999. Abnormal response to negative feedback in depression. Psychological Medicine 29, 63–72. Speechley, W.J., Murray, C.B., McKay, R.M., Munz, M.T., Ngan, E.C., 2009. A failure of conflict to modulate dual-stream processing may underlie the formation and maintenance of delusions. European Psychiatry 25, 80–86.
Contents lists available at ScienceDirect
Schizophrenia Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c h r e s
Exploring logical reasoning abilities in schizophrenia patients Dario Mirian a,⁎, R. Walter Heinrichs a, Stephanie McDermid Vaz b a b
Department of Psychology, York University, 4700 Keele Street, Toronto, Ontario Canada M3J 1P3 Cleghorn Early Intervention in Psychosis Program, St Joseph's Healthcare Hamilton, 25 Charlton Avenue East, Suite 703, Hamilton, Ontario, Canada L8N 1Y2
a r t i c l e
i n f o
Article history: Received 29 October 2010 Received in revised form 11 January 2011 Accepted 13 January 2011 Keywords: Deductive logic Cognition Symptoms Schizophrenia Performance
a b s t r a c t Objective: To assess deductive reasoning in schizophrenia patients with special reference to whether accuracy varies across type of stimulus problem. Previous research suggests that patients, unlike healthy controls, are insensitive to emotionally provocative (salient) problem content. Method: A syllogistic reasoning task consisting of five argument types varying in salience, congruence with commonly held beliefs and meaningfulness was administered along with standard intellectual and symptom measures to 25 schizophrenia patients and 26 healthy control participants. Results: Patients performed below control participants in all reasoning task conditions, but group differences were non significant after controlling for IQ. There were no significant interactions between group and argument type in terms of reasoning accuracy and both patients and controls performed better when reasoning with belief congruent material. In addition, no relation between deduction and paranoid symptoms was found. Conclusion: Formal deductive reasoning abnormalities in schizophrenia are a reflection of the broadly based cognitive impairment documented in the illness. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Although schizophrenia involves severe disturbances in thinking, the formal study of reasoning capacities in patients has not been pursued comprehensively. Research has primarily focused on reasoning biases in relation to symptoms such as delusions (Corcoran, 2003; John and Dodgson, 1994). Patients with delusions show abnormalities and biases in probabilistic reasoning and “jump to conclusions” by making decisions with minimal information (Garety and Freeman, 1999; Garety et al., 2005). In addition, their attributional reasoning about negative events shows a bias towards assigning causes to external factors (Jolley et al., 2006). In contrast, deductive reasoning, which involves evaluating an argument drawn from a given premise, has received relatively little attention (Speechley et al., 2009). The validity of an argument and its conclusion is contingent on the validity of the premises preceding it (Evans and St, 2005). Moreover, validity judgments about an argument are influenced by pre-existing beliefs regarding its conclusion (Goel and Dolan, 2003). These properties of deduction imply potentially fruitful avenues of investigation in relation to the psychopathology of schizophrenia.
⁎ Corresponding author. c/o Dr. R. W. Heinrichs York University, Department of Psychology, 4700 Keele Street, Toronto, ON, Canada, M3J 1P3. Tel.: + 1 647 298 8222. E-mail addresses: [email protected], [email protected] (D. Mirian), [email protected] (R.W. Heinrichs), [email protected] (S.M. Vaz). 0920-9964/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2011.01.007
Previous research suggests that schizophrenia patients may have difficulty with deductive reasoning in general and, in contrast with healthy controls, show equivalent performance in emotionally provocative or salient and non-salient problem variations (Goel et al., 2004; Ho, 1974; Mujica-Parodi et al., 2000; Speechley et al., 2009). Goel et al. (2004), administered syllogisms to schizophrenia patients and controls and found that patients performed equally poorly across all problem types, whereas healthy controls were most proficient in solving emotionally neutral or “non-salient” problems. However, a critical limitation of this study was the presence of a floor effect in the patient group whereby performance approximated chance levels. Another drawback was their study did not assess incongruent and congruent problem types. This distinction reflects reasoning with material that is consistent with prior knowledge and beliefs as contrasted with material consistent with such beliefs. Speechley et al. (2009), showed that schizophrenia patients with delusions were impaired in the incongruent trials. However, the sample sizes were small and it is unclear whether symptom profiles and intensity figure in this impairment. In light of these considerations, we studied deductive reasoning in schizophrenia patients and healthy participants with problem types varying in salience, congruence and meaningfulness. We were interested specifically in the question of whether deductive reasoning is deficient across problem types or occurs more selectively. In addition, we examined possible relations between symptom severity and reasoning performance.
D. Mirian et al. / Schizophrenia Research 127 (2011) 178–180
2. Method
Table 2 Adjusted mean accuracy scores and standard deviations for belief condition.
2.1. Participants The sample comprised 26 healthy control participants and 25 patients meeting DSM-IV-R (American Psychiatric Association Task Force on DSM-IV, 2000) criteria for schizophrenia (80%) or schizoaffective disorder (20%) based on the Structured Clinical Interview for DSM-IV (SCID; First et al., 1996). Table 1 presents the demographic characteristics of the sample. Control participants were recruited from the greater Toronto area and patients were recruited from outpatient settings that included the Hamilton Program for Schizophrenia, the Community Schizophrenia Service and the Cleghorn Program (St. Joseph's Healthcare Hamilton). All patients were receiving antipsychotic medication during the data collection period. The study received approval by the institutional review board at York University and St. Joseph's Healthcare affiliated research sites. 2.2. Clinical and cognitive measures Patients' symptoms were measured with the Positive, Negative and General Psychopathology subscales of the Positive and Negative Syndrome Scale (PANSS; Kay et al., 1987; Opler et al., 1999). General cognitive abilities of both patients and controls were indexed with the Vocabulary and Matrix Reasoning subtests of the Wechsler Abbreviated Scale of Intelligence (WASI; Psychological Corporation, 1999). Syllo-
Table 1 Demographic characteristics of sample. n
%
Gender Male Female
18 33
35.3 64.7
Marital status Single Married Divorced Other
39 7 4 1
76.5 13.7 7.8 2.0
Employment status Full-time Part-time Unemployed Volunteer Student
4 11 22 1 13
7.8 21.6 43.1 2.0 25.5
Hospitalization in past 2 years None 1–2 times 3+2
17 2 3.9
33.3 3.9
Criminal record Yes No
6 15
28.6 71.4
Smoking Yes No
14 37
27.5 72.5
Alcohol use Yes No
25 26
49.0 51.0
Drug use Yes No
8 43
15.7 84.3
M
Patients Controls
M SD M SD
Non-belief accuracy (%)
Belief salient accuracy (%)
Belief-non salient accuracy (%)
54.67 15.61 71.15 19.18
62.67 18.18 78.85 17.36
65.33 23.03 73.72 19.39
gisms, which include two premises and a conclusion, were presented to assess deductive reasoning abilities and beliefs. Participants were initially given verbal instructions and practice in deductive problemsolving with neutral stimuli. The experimental task included 18 syllogisms: 6 belief-salient (e.g., No drunks are saints; some Irishmen are saints; therefore some Irishmen are not drunks), 6 belief-non-salient (e.g., All crunchy tubers are vegetables; some carrots are crunchy tubers; therefore some carrots are not vegetables) and 6 non-belief (e.g., Some ploms are dihedric; no moduleses are ploms; therefore some moduleses are dihedric). Within the 12 belief-based syllogisms, 6 problems were belief-congruent (e.g., All flying birds have feathers; no people have feathers; therefore some people are flying birds) and 6 were beliefincongruent (e.g., All handicapped are capable; all amputees are handicapped; therefore some amputees are not capable). In each case participants were required to indicate whether the conclusion was correct or incorrect. The syllogisms were selected from an item bank incorporating a range of difficulty (Dickstein, 1978). In order to avoid floor effects, syllogisms of relatively low difficulty, correctly answered by 90% of healthy controls in the original item bank, were chosen. Participants were allowed a maximum of 60 s to solve each item.
SD
3. Results Pearson correlations were calculated to index relationships between percentage correct in the reasoning tasks and five subscales from the PANSS (thought disturbance, activation, anergia, paranoid belligerence, and depression). One statistically significant correlation was found: correct responses on non-belief items were negatively related to depression scores, r(23)= −.43, pb.05. Two analyses of covariance (ANCOVA) with repeated measures were carried out on reasoning test accuracy data. The first ANCOVA was done to test for main effects of group (schizophrenia vs. control) and belief problem type (belief-salient, belief-non-salient, non-belief) and for group by belief interaction, while controlling for IQ. Table 2 displays the covariate-adjusted descriptive statistics for each group under each belief condition. The assumptions of ANCOVA (i.e., homogeneity-of-regression (slopes), sphericity, and homogeneity of variance) were evaluated and indicated no violations. After adjusting for IQ, there were no significant group differences across belief problem types F(1, 47) = .21, p = .65. There was also no main effect of belief problem type on accuracy scores F(2, 94) = 1.12, p =.33, and the interaction of group and belief problem type was nonsignificant, F(2, 94) = .66, p = 0.52. The second ANCOVA was carried out to test for the main effects of group and congruence problem type (i.e., congruent and incongruent). Table 3 displays the adjusted descriptive statistics for each group under each trail condition. Again, all statistical assumptions were satisfied. After controlling for IQ, there
Table 3 Adjusted mean accuracy scores and standard deviations for trial conditions.
Patients Age Education Age of psychiatric diagnosis
179
37.47 14.33 22.05
13.60 2.89 4.90
Controls
M SD M SD
Incongruent accuracy (%)
Congruent accuracy (%)
44.00 27.17 74.36 22.72
82.00 15.15 78.21 18.11
180
D. Mirian et al. / Schizophrenia Research 127 (2011) 178–180
Table 4 Unadjusted mean accuracy scores and standard deviations for belief condition.
Patients Controls
M SD M SD
Non-belief accuracy (%)
Belief salient accuracy (%)
Non-salient belief accuracy (%)
54.67 15.61 71.15 19.18
62.67 18.18 78.85 17.36
65.33 23.03 73.72 19.39
Table 5 Unadjusted mean accuracy scores and standard deviations for trial conditions.
Patients Controls
M SD M SD
Incongruent accuracy (%)
Congruent accuracy (%)
44.00 27.17 74.36 22.72
82.00 15.15 78.21 18.11
was no significant difference between the two groups across problem type F(1, 47) = .06, p = .81. However, there was a main effect of congruence problem type F(1, 47) = 10.17, p = .003. Specifically, both patients and controls were more accurate in solving congruent (M = 80.07, SD = 16.67) than incongruent problems (M = 59.48, SD = 29.11). For comparative purposes, Tables 4 and 5 present reasoning performance scores unadjusted for IQ. 4. Discussion The present study examined deductive logic in patients with schizophrenia and in healthy comparison subjects by varying the content of syllogisms, while statistically controlling for general intellectual ability. Results indicate that schizophrenia patients are indistinguishable from healthy controls in terms of deductive problem solving once the contribution of general ability to reasoning accuracy is removed. We found no support for the conjecture that the illness involves a selective insensitivity to emotionally provocative problem content. Moreover, belief congruence, or reasoning with material consistent with general knowledge about the world, facilitates problem solving for both patients and controls. There is no evidence that schizophrenia involves insensitivity to this kind of general knowledge and belief system or any preferential processing of idiosyncratic and norm-divergent problem content. Accordingly, it seems likely that deductive reasoning impairments in the illness reflect the broadlybased cognitive dysfunction described in the literature over the past two decades (e.g. Heinrichs and Zakzanis, 1998) and not a specific weakness in deduction. Furthermore, we found no evidence that defective deductive reasoning with any type of problem content associates with the kind of implausible beliefs expressed in persecutory delusions, although some aspects of problem content may vary with depression. There are reports suggesting that depression leads to decreased cognitive ability especially a lack of motivation in the schizophrenia population (Shah et al., 1999). Overall, therefore, the present study does not support findings by Goel et al. (2004) and Speechley et al. (2009) suggesting a specific weakness in deduction in schizophrenia and an abnormal sensitivity to problem content. Poor deduction skills probably reflect the general weakness in cognitive performance observed in the illness rather than any special difficulty in evaluating the logic and validity of arguments. Future studies with larger sample sizes and
matched case/control designs are warranted. In addition, prospective studies should include a larger number of syllogisms to assess broader deductive reasoning abilities. Role of funding source Funding for this study was provided by the Ontario Mental Health Foundation (OMHF) and by the Community Schizophrenia Vocational Rehabilitation Foundation (CSVR). The OMHF and CSVR Foundation had no role in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication. Contributors W. Heinrichs designed the study and wrote the protocol. S. McDermid Vaz collected the data. W. Heinrichs and D. Mirian conducted the analyses. All authors contributed to and approved the final manuscript. Conflicts of interest There are no conflicts of interest with respect to this manuscript. Acknowledgements We express our appreciation to Ashley Oman, B.A., and the staff of the Community Schizophrenia Service, the Cleghorn Program, St. Joseph's Healthcare Hamilton, the Hamilton Program for Schizophrenia, and the Canadian Mental Health Association Toronto Branch.
References American Psychiatric Association Task Force on DSM-IV-R, 2000. Diagnostic and Statistical Manual of Mental Disorders (4th ed revised). American Psychiatric Association, Washington. Corcoran, R., 2003. Inductive reasoning and the understanding of intention in schizophrenia. Cognitive Neuropsychiatry 8, 223–235. Dickstein, L.S., 1978. The effect of figure on syllogistic reasoning. Memory & Cognition 6, 76–83. Evans, J.St.B.T., 2005. Deductive Reasoning. In: Holyoak, K.J., Morrison, R.G. (Eds.), The Cambridge Handbook of Thinking and Reasoning. Cambridge University Press, New York, pp. 169–184. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 1996. Structured Clinical Interview for DSM-IV Axis I Disorders. Non-Patient Edition (SCID-I/NP). New York: New York State Psychiatric Institute, Biometrics Research. Garety, P.A., Freeman, D., 1999. Cognitive approaches to delusions: a critical review of theories and evidence. The British Journal of Clinical Psychology 38, 113–154. Garety, P.A., Freeman, D., Jolley, S., Bebbington, P.E., Kuipers, E., Dunn, G., Fowler, D.G., Dudley, R., 2005. Reasoning, emotions, and delusional conviction in psychosis. Journal of Abnormal Psychology 114, 373–384. Goel, V., Dolan, R.J., 2003. Reciprocal neural response within lateral and ventral medial prefrontal cortex during hot and cold reasoning. Neuroimage 20, 2314–2321. Goel, V., Bartolo, A., St. Clair, D., Venneri, A., 2004. Logical reasoning deficits in schizophrenia. Schizophrenia Research 66, 87–88. Heinrichs, R.W., Zakzanis, K.K., 1998. Neurocognitive deficit in schizophrenia: a quantitative review of the evidence. Neuropsychology 12, 426–445. Ho, D.F., 1974. Modern logic and schizophrenia thinking. Genetic Psychology Monographs 89, 145–165. John, C., Dodgson, G., 1994. Inductive reasoning in delusional thought. Journal of Mental Health 3, 31–49. Jolley, S., Garety, P., Bebbington, P., Graham, D., Freeman, D., Kuipers, E., Fowler, D., Hemsley, D., 2006. Attributional style in psychosis — the role of affect and belief type. Behaviour Research and Therapy 44, 1597–1607. Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The Positive and Negative Syndrome Scale (PANSS) for Schizophrenia. Schizophrenia Bulletin 12, 261–276. Mujica-Parodi, L.R., Malaspina, D., Sackeim, H.A., 2000. Logical processing, affect and delusional thought in schizophrenia. Harvard Review of Psychiatry 8, 73–83. Opler, L.A., Kay, S.R., Lindenmayer, J.P., Fiszbeln, A., 1999. Structured Clinical Interview for the Positive and Negative Syndrome Scale (SCI-PANSS). Multi-Health Systems, North Tonawanda: New York. Psychological Corporation, 1999. Wechsler Abbreviated Scale of Intelligence (WASI) Manual. San Antonio, TX: Author. Shah, P.J., O'Carroll, R.R., Rogers, A., Moffoot, A.P.R., Ebmeier, K.P., 1999. Abnormal response to negative feedback in depression. Psychological Medicine 29, 63–72. Speechley, W.J., Murray, C.B., McKay, R.M., Munz, M.T., Ngan, E.C., 2009. A failure of conflict to modulate dual-stream processing may underlie the formation and maintenance of delusions. European Psychiatry 25, 80–86.