Whether or not subjects can simulate mental retardation, a consideration that has implications in criminal cases, was studied using 21 adult Caucasian males between 20 and 30 years of age, largely comprised of students and staff employees of the University of New Mexico. Subjects were asked to give genuine and simulated responses to two major test batteries of intelligence, the Stanford Binet Fourth Edition and the Wechsler Adult Intelligence Scale Revised. Individual subjects did not appear able to simulate mental retardation consistently on both subtests and achieve statistically similar results. Latency times for simulating appeared to be significantly longer than latency times for genuine responding. Findings further suggested that genuine responses will yield similar scores on both subtests. Qualitative data suggest that it is more difficult to simulate mental retardation than to give genuine answers. Implications for participants in the criminal justice system may be significant, particularly for defendants accused of falsifying a test. Fourteen tables present study data. (Contains 145 references.) (SLD)