Abstract
We analyzed the ability of the Wisconsin Card Sorting Test to distinguish between 41 malingering and 31 normal undergraduates, and 17 probable malingering and 16 brain injured patients. A logistic regression consisting of number of categories (CAT) and failure to maintain set (FMS) distinguished malingering and normal undergraduates with 70.7% sensitivity, 87.1% specificity, and 77.8% overall classification, and distinguished patients suspected of malingering from brain injured controls with 82.4% sensitivity, 93.3% specificity, and 87.5% overall classification.