Card sorting test performance : the role of visual working memory and the effect of visual feedback

Poor performance on the Wisconsin Card Sorting Test (WCST: Heaton, Chelune, Talley, Kay, & Curtis, 1993), a test of executive function, has been attributed to deficits in information processing, cognitive flexibility, inhibitory control, and working memory. Previous WCST studies, involving student participants, have indicated a primary role for the phonological loop and/or central executive components of working memory, whereas studies involving individuals with schizophrenia or schizotypic traits have indicated spatial working memory involvement. So far, no study has investigated the role of the visual and spatial subcomponents of the visuospatial sketchpad or has included all components of working memory within the one study. The visual aspects of the WCST have also been overlooked in many studies, although Cinan and Tanor's (2002) findings suggested that the presence or absence of visual feedback from previous response cards might differentially affect performance. Therefore, the aim of this thesis was to clarify the role of working memory in performance of the WCST, using Baddeley and Hitch's (1974) working memory model as a conceptual framework. The series of experiments examined the contribution to WCST performance by the phonological loop, the central executive, and the visual and spatial subcomponents of the visuospatial sketchpad. It also investigated the effect of visual feedback on the individual working memory components. Participants included university undergraduates psychometrically identified as having schizotypal personality traits (schizotypy) and mild to moderate head-injured patients. Experiments IA and 1B explored the relationship between schizotypy, working memory, and card sorting test performance using computerised versions of the WCST and the Madrid Card Sorting Test (MCST: BarcelO & Knight, 2002). No significant differences were found between high and low schizotypic scorers on any dependent measure. However, low scorers on working memory tasks performed significantly worse than high scorers on all card sorting test variables. Additionally, performance was differentially affected by the presence (WCST) or absence (MCST) of visual feedback. After piloting the appropriate secondary tasks in Experiment 2, a dual-task paradigm was used with a nonclinical population in Experiment 3. Results indicated significant involvement of the visual subcomponent when visual feedback is absent and greater involvement of the phonological loop when visual feedback is present. Experiment 4 further explored this dissociation in a group of head-injured patients with differential impairments in executive functioning, auditory working memory, and visual working memory. The results showed card sorting test performance was significantly worse for patients with visual working memory impairment than for any other group. Overall, the results from this thesis suggest that the visual subcomponent of the visuospatial sketchpad, which has previously been overlooked, contributes significantly to card sorting test performance; and that the degree of contribution of the individual components maybe differentially affected by the presence or absence of visual feedback. These findings have implications for both the methodology used in future experimental studies and for the interpretation of WCST performance in clinical practice. A redesign of the WCST procedure is proposed to enable the clinician to differentiate between poor performance due to perseveration and that due to distractibility.