The recovery cycle of the P1 component of the auditory evoked potential (50-70 ms latency) has been reported as abnormal in both unmedicated and medicated patients with schizophrenia when a paired stimuli protocol is used to examine recovery. However, findings have been mixed when a stimulus train protocol is used. This study examined the effects of medication history on P1 abnormalities in schizophrenia assessed by a stimulus train protocol. Auditory evoked potentials were recorded from 14 normal controls, 14 neuroleptic naive patients with schizophrenia and 14 previously medicated patients in response to binaural clicks presented at three stimulus rates: 1/s, 5/s and 10/s. Neuroleptic naive patients showed a smaller P1 at the baseline rate (1/s) and greater recovery (less amplitude suppression) at faster rates than did normal controls or previously medicated patients. Additional analyses suggested that this latter effect was not due to smaller baseline P1 amplitudes. Greater recovery in neuroleptic naive patients was not associated with clinical symptomatology contrary to previous findings of the authors for a mixed sample of neuroleptic naive and previously medicated patients. Medication status appears to account for some of the variability in P1 abnormalities in schizophrenia although identification of the underlying mechanism responsible requires further study.