In a double-blind fourfold crossover design, 11 subjects were randomly assigned to placebo, 10 mg diazepam, 75 mg amitriptyline, and 75 mg chlorpromazine. During a simple vigilance task, 12 midline and left hemisphere leads were recorded before and 3 h after drug administration. The EEG was quantified by spectrum analysis, the topographic structure displayed by brain mapping techniques, and the results compared with earlier studies which used the same design and drugs. Diazepam showed the expected increase in beta; however, fast beta was increased as much as slow beta. Amitriptyline showed an increase of slow wave power and a reduction of alpha. In contrast to earlier studies, a decrease of fast beta was found. In addition, the spatial pattern of alpha changed from an occipital to a parietal maximum. Chlorpromazine showed an increase in the theta band. In occipital regions, there was a small decrease of fast beta; however, centrally there was an increase of both slow and fast beta. These results were confirmed by a multivariate analysis of variance.