The first study involved 48 simulators, 4 amnesics, and 27 nonsimulating controls. Experimental tasks were an autobiographical interview, a wordstem completion task, and a word association task. All the subjects who participated in the first study participated in the second study along with an additional 53 subjects that included 21 patients with memory disorders resulting from brain damage and an additional 32 nonsimulating controls. Two 20-item lists of concrete nouns were used in this study as the experimental task. The studies were conducted because there is a consensus in the legal community that amnesia is easily faked and practically impossible to disprove and that many who claim to be amnesic are malingering. In the two studies, simulators displayed performance patterns different from those of memory-impaired subjects. The results suggest that laypersons have inaccurate beliefs about the cognitive features of amnesia and do not distinguish among etiologically distinct amnesic disorders. Tasks that exploit laypersons' inaccurate beliefs about amnesia appear promising for the detection of malingering. 6 tables, 5 figures, and 55 references. (Author abstract modified)