Abstract
The Wisconsin Card Sorting Test (WCST) is widely used to assess executive function, specifically mental flexibility. Item Response Theory (IRT) has several advantages over classical test theory, including the invariance of the measure. This study aimed to apply IRT to study the psychometric properties of WCST in control subjects and patients with idiopathic Parkinson's disease (PD) and Alzheimer's disease (AD). The sample consisted of 86 controls, 77 Parkinson's disease subjects, and 155 AD subjects. The Rasch model was applied to binary data from the conversion to percentiles adjusted for age and schooling. The R program was used to calibrate the difficulty parameter. The characteristic curves of the items and the estimation of the difficulty parameter for each diagnostic group were obtained. WCST makes it possible to separate between different skill levels among the three diagnostic entities and its application may be useful for other neuropsychological tests. In conclusion, WCST has good clinical sensitivity and excellent discriminant validity in the groups under study, making it possible to separate them between different levels of ability or latent trait. Its application may be useful for other neuropsychological tests.