The aim of this study was to assess cognitive effects of anticonvulsants in a way that would yield results that are most directly applicable to epilepsy populations. This was done with a placebo-controlled, prospective, randomized, double-blind, parallel group study of anticonvulsant withdrawal in a population of subjects taking a single anticonvulsant with completely controlled seizures. Outcomes of this study on cognitive measures from the California Computerized Assessment Package have recently been reported. To aid comparison with results of prior studies, we report outcomes here on several more standard measures of neuropsychological function. The major findings were that, in subjects with therapeutic drug levels at baseline, drug withdrawal was associated with significant improvement in performance on the Controlled Oral Word Association Test and the Stroop Color-Word Interference Test. Comparable results were achieved in the subgroup taking carbamazepine.