OBJECTIVE: To evaluate the cognitive effects of topiramate (TPM) and gabapentin (GBP). METHODS: Forty healthy volunteers were randomized to a 12-week course of TPM, GBP, or placebo. Doses were gradually escalated over 10 weeks to a maximum of 400 mg/day of TPM or 3,600 mg/day of GBP or to the highest tolerated dose. Subjects were interviewed and examined biweekly. Cognitive testing was performed prior to initiating the drug and again 12 weeks later, at least 2 weeks after achieving plateau dosing. For each subject and cognitive measure, test-retest Z scores were calculated based on regression equations derived from 73 healthy volunteers. Group comparisons utilized the Wilcoxon test. RESULTS: There were significant TPM vs GBP and TPM vs placebo differences in test-retest Z scores for four of six target cognitive measures (Digit Symbol, Story Recall, Selective Reminding, Controlled Oral Word Association), always indicating worse retest performance for subjects receiving TPM. Overall, 12 of 24 cognitive measures were similarly affected. TPM effects were large, and several target measures averaged >2 SD of negative change. One measure was significantly affected by GBP. CONCLUSIONS: Topiramate (TPM) impaired cognitive test performance, whereas gabapentin had minimal effects. The effects of TPM were of sufficient magnitude potentially to affect daily and occupational function.