Forty mentally retarded adolescents were divided into two divisions on the basis of prestudy assessments that identified subjects as thioridazine hydrochloride responders or nonresponders. Within each division, the participants were assigned randomly to one of four groups. Members of each group received three administrations of the Leiter international performance scale. Four conditions were examined: testing under standard conditions and testing under reinforcement conditions while receiving and not receiving thioridazine. Condition sequences were random in all four groups. Assessments were double-blind and placebo controlled. The results showed that while both drug responders and nonresponders were receiving thioridazine there were no major differences in test scores obtained under standard and reinforcement conditions. While subjects were not receiving thioridazine there were considerable increases in scores obtained under the reinforcement conditions.