We compared in a placebo controlled, double blind, cross-over within-subject design, the amplitude and area integral of contingent negative variation (CNV) using a 2 second interstimulus interval in migraine patients between attack and interval before and after treatment. The study was conducted on 14 female subjects suffering from migraine without aura. The measurements were performed in a balanced sequence at four different times on each patient, twice during the migraine interval and once in each of two migraine attacks. The CNV in the patients was measured first (baseline), then medication was administered on a double-blind basis with an auto-injector, using either 6 mg sumatriptan or a placebo solution. Thirty minutes after administration the CNV parameters were measured again and the changes between pre-and post-treatment were taken as dependent variables. CNV amplitude baseline readings did not differ significantly between the four conditions, Neither administration of placebo nor sumatriptan led to a significant change in CNV parameters independent of whether significant clinical improvement of migraine headache occurred or not. According to our findings CNV-mechanisms between attack and interval are not subject to short-term changes, even though a small, not significant tendency towards · decrease in CNV amplitude during migraine attacks appears to exist. Therefore, it can be assumed that changes in the systems which are depicted by CNV readings are not involved in initiating and terminating acute migraine attacks.