Nelson's (1976) Modified Wisconsin Card Sorting Test (MCST) was evaluated and the utility of the conventional criterion and new threshold scores for identifying patients with frontal lobe disorder assessed. Measures of perseveration and other indices of performance on the MCST were not found to discriminate between patients with frontal and non-frontal lobe lesions, using either the conventional criterion or new threshold scores. In addition there were no significant differences between left- and right-sided lesioned patients. However, the task was found to have high specificity (98.7 per cent) and good sensitivity (45.6 per cent) to lesions irrespective of site indicating that it may be a useful screening test. The present findings are consistent with recent research with the original Wisconsin Test, that suggests that card-sorting performance may not be as differentially sensitive to frontal lobe disorder as has been thought hitherto.
