In the present study, a sensorimotor "patterning" program used with 66 institutionalized, mentally retarded children and adolescents was evaluated. The subjects were randomly assigned to one of three groups: (a) Experimental 1 group, which received a program of mobility exercises including patterning, creeping, and crawling; visual-motor training; and sensory stimulation exercises; (b) Experimental 2 group, which received a program of physical activity, personal attention, and the same sensory stimulation program given to the first group; or (c) Passive Control group, which provided baseline measures but which received no additional programming as part of the study. Experimental 1 group subjects improved more than subjects in the other groups in visual perception, program-related measures of mobility, and language ability. Intellectual functioning did not appear to be enhanced by the procedures, at least during the active phase of the project. The results were discussed with reference to other researchers who have failed to support the patterning approach, and some reasons were suggested for the differences between the present and past investigations.