Date of Award


Degree Name

Doctor of Philosophy



First Advisor

Dr. Dale M. Brethower

Second Advisor

Dr. William K. Redmon

Third Advisor

Dr. Lisa Baker


This study compared multiple sequential performances on a computerized version of the Wisconsin Card Sorting Test (Grant & Berg, 1948; Harris, 1986) by adult men receiving institutional inpatient treatment for chronic schizophrenia and related disorders. Participants were exposed to differential test conditions of minimal post-session non-contingent token reinforcement (baseline phases) and enhanced intra-session performance-contingent token reinforcement (experimental phase). Each major participant resided at the same continuing treatment unit at Kalamazoo Regional Psychiatric Hospital receiving similar social therapies but idiosyncratic psychotropic medication regimens, changes in which were tracked in relation to test performances. A simple single-subject reversal design was employed with each of the 6 fully participating subjects who were engaged in at least 5 sessions under each phase o f their study. Each schizophrenic participant was rated by the same interdisciplinary treatment team using the Scale for the Assessment of Positive Symptoms and the Scale for the Assessment of Negative Symptoms (Andreasen, 1982; Andreasen & Olsen, 1982; Andreasen, 1984a, 1984b).

The results show that multiple consecutive administrations can be given to impaired individuals without problematic interference from learning effects. Further, schizophrenic participants produce test performance measures that are like those of participants with residual cognitive impairments from exogenous frontal lobe brain damage. Contingent reinforcement failed to improve test performances and, in some cases, apparently worsened perseveration. However, all participants displayed overt signs of increased interest, attentional efforts and reward concern as well as elated affect and animation during this phase. While later test measures produced different performance profiles than initial measures, substantial cognitive impairment was still reflected for all but one participant. Except for that same participant overall negative symptom scale item percentages closely predicted later perseverative error test performance percentages.

Serendipitous results support common conclusions found in the literature in these areas. Greater variability across performances was generally seen in schizophrenic participants. In one organic study modelling produced remarkable performance improvements when practice, contingent reinforcement and instruction had not. In both schizophrenia-related studies involving the atypical antipsychotic medication clozapine, dramatic test performance improvements and lower positive and negative symptom item ratings resulted. The implications of these findings are discussed.

Access Setting

Dissertation-Open Access