The regional cerebral effects of an anxiolytic (clorazepate) in 20 patients with generalized anxiety disorder were assessed using 16-channel electroencephalogram (EEG) power spectral estimate maps of the left hemisphere. Patients were studied with double-blind random assignment to placebo or drug and were assessed at baseline, day 7, and day 14 with EEG and Hamilton Anxiety Ratings. Ten age- and sex-matched normal controls were also tested. Hamilton Anxiety Ratings were decreased significantly more in the drug group than in the placebo group. Topographic maps of EEG activity revealed decreases in occipital alpha and parietal delta, together with increases in posterior frontal (central EEG leads) and parietal beta. Decreases in delta are consistent with a lack of sedation; the reciprocal beta increases in parietal cortex are similarly consistent. This pattern of regional EEG changes in the direction of alert attentiveness, together with individual differences, observed in frontal/parietal and occipital alpha activity, suggests the importance of at least these two cortical regions for anxiolytic action. Differences between patients with generalized anxiety disorder and normals were restricted to the occipital and temporal regions. These results suggest the importance of multilead recording in assessing EEG correlates of drug action.