There is increasing evidence that the P 100 peak of the pattern-reversal visual-evoked potential (VEP-PR) is delayed by drug-induced dopamine antagonism and in Parkinson's disease. Recent studies have reported that components of the flash-VEP (VEP-F) are delayed by an anticholinergic which does not affect the VEP-PR. The present study found that a single dose of chlorpromazine increased the latencies of the VEP-PR and of the VEP-F and increased the VEP-F P2 amplitude. Trihexyphenidyl increased the VEP-PR amplitude but had only minor effects on the VEP-F. There was a tendency for imipramine to increase VEP-F latencies, especially the N3 peak, but had no effect on the VEP-PR. Both VEPs were unaffected by diazepam. These VEP findings add further support to the role of dopamine in the human visual system. Possible reasons are advanced for the failure of trihexyphenidyl to cause previously reported VEP changes associated with hyoscine hydrobromide. Several important issues need to be addressed by future research.