BACKGROUND: Epilepsy and antiepileptic drugs may be associated with cognitive dysfunction. P300 event-related potentials have been used to assess cognitive function. In this study we investigated the effect of antiepileptic drugs on P300 event-related potentials in epileptic patients. METHODS: P300 latency in event-related potentials was recorded during an auditory oddball paradigm in 29 patients, 10 with generalized epilepsy and 19 with partial epilepsy. The relationships between antiepileptic drugs and P300 latency prolongation were analyzed. RESULTS: P300 latency was prolonged in 31% (9/29) patients with epilepsy. There were no significant relationships between etiology of epilepsy, type of seizure, or type of antiepileptic drugs and P300 latency prolongation. The number or serum level of antiepileptic drugs and treatment duration were related to latency prolongation, but these relations were not significant. CONCLUSION: The prolongation of P300 latency in event-related potentials is not significantly related to antiepileptic drug. This finding suggests that P300 may not be a sensitive additional procedure to assess the cognitive status in epileptic patients.