Application of contingent physical restraint to suppress stereotyped body rocking of profoundly mentally retarded persons.

The effectiveness of contingent restraint in reducing stereotyped body rocking of profoundly retarded subjects was investigated to determine (a) whether restraint reduced stereotyped body rocking and (b) the degree to which the effectiveness of restraint depended upon where in the behavioral sequence it was administered. Results showed that restraint, in comparison to no restraint, reduced the amount of body rocking. Also, restraint was judged to be generally more effective if it were applied when the subject was bent over in the middle of the response sequence rather than administered while the subject was sitting upright at the end of the response sequence. The results were interpreted in terms of response chaining and postural feedback. Implications for the utilization of physical restraint as a behavioral intervention were discussed.