The aims of this study were to: (a) examine the consistency of the published Wisconsin Card Sorting Test (WCST) factor structures; (b) determine the factor structure of the WCST in a large, heterogenous sample; and (c) compare the WCST factor analytically with other neuropsychological procedures. Two WCST factors (concept formation/perseveration and Failure-to-Maintain-Set [FMS]) were consistently reported in the literature. Our analysis of data from 473 clinical cases replicated the two factors previously reported and revealed a third on which nonperseverative errors (NPE) was the sole salient variable. This pattern was maintained in three of four diagnostically distinct subgroups. These factors are potentially clinically meaningful, with each seeming to reflect one of three qualitatively different performance styles. In the construct validation factor analysis, WCST scores loaded independently of other neuropsychological variables, indicating that the WCST contributes uniquely to neuropsychological evaluation. Nevertheless, despite the rational interpretation of the factors, the cognitive processes underlying WCST performance remain poorly understood. Future directions for the application of these factor analytic findings are discussed.