The aim of the present study was to investigate the temporal stability of the Wisconsin Card Sorting Test (WCST) in a clinical population with documented brain pathology, stable cognitive deficits, and for whom repeated testing is common: chronic severe traumatic brain injury (TBI). Participants were 34 patients at least 1 year post severe TBI living at a large residential rehabilitation facility. The WCST was administered in standard fashion with both the standard and 64-card versions scored. All derived scores with norms were examined. Results indicated acceptable temporal stability of most scores for both the standard and short WCST, although the stability of the WCST-64 was poorer than for the standard WCST. Three sets of significant change indices are provided for clinical use.