Differential responses on cognitive function of young (18-45 years, n = 17) and elderly (60-75 years, n = 9) subjects were compared before and after a challenge with lorazepam (2 mg oral) or placebo. Tests of memory and frontal lobe function were administered to determine the specificity of the amnesic deficit produced and the interaction of drug and the aging brain. Anterograde amnesia, as assessed by recall of a word list, was noted in both groups and was greater in the elderly. Whereas lorazepam produced only impaired recall in the young group, the elderly group manifested a further range of cognitive impairments, including decreased verbal fluency on the controlled oral word association test and reduced performance on the Wisconsin card-sorting test. These impairments in the elderly group could not be attributed to performance deficits pre-drug. Short-term memory, as evidenced by digit span and the copy trial of the Rey figure, was not impaired by lorazepam. It is concluded that whilst lorazepam produces a relatively `pure' amnesia in young to middle aged adults, in the elderly there is an admixture of deficits in some frontal lobe functions. This is presumed to reflect age-related changes in the brain and may be analogous to the spectrum of results noted in Korsakoff's psychosis as opposed to 'pure' diencephalic amnesia.