| An examination of the quantitative and qualitative features of the Wisconsin Card Sorting Test, and a comparison of it with the Halstead Category Test, was conducted with a heterogeneous sample of brain damaged and neurologically normal subjects (N = 100). Conventional ways to score the test were analyzed, as well as original concept shifting scores. Both types of scores were effective in differentiating the brain dysfunction group from the normal group. Consistent patterns of shifting and absence of shifting were displayed in relation to criterion group, and level of test performance. For example, the brain damaged group demonstrated a clear tendency to stay with a category, especially following an error. The normal group displayed a distinctly opposite pattern. The relationships between the test and age, education, and intelligence were explored in the total sample, and in each group separately. Differential patterns of association were manifest in the groups. The diagnostic accuracy of scores, singularly and in combination, was investigated. The findings would support the Wisconsin Card Sorting Test as a potentially useful measure for detecting cerebral dysfunction generally, and in relation to laterality. The construct validity correlation between the Wisconsin Card Sorting Test and the Halstead Category Test was moderate, indicating approximately 40% shared variance. Partial correlation was used to attempt to clarify certain aspects of the relationship between these tests. The possible content and process variables which might account for the shared variance and the absence of it were explored. Suggestive classificatory problem solving strategies were discussed. |
