Since Lindsley and Skinner reported for the first time a method for the experimental analysis of the behavior of psychotic patients, many investigators have successfully utilized operant conditioning procedures to manipulate the behavior of adult psychotics. The behavior of both normal and retarded children has also been shown to be readily controllable through such conditioning procedures. The assumption was made only that the child's behavior had probably been shaped as theorized by Ferster and, at any rate, could be brought under significant control by a social learning therapy. The crux of the therapy was to utilize play situations for the differential reinforcement and extinction of various behaviors. "Therapist-control" was defined as the percentage of commands obeyed by the child with each therapist without the use of candy-reinforcements. Throughout the treatment program, it became increasingly apparent to both the author and the therapists that most valuable data were being collected in the reports which each therapist wrote up following a therapy-session.