Date of Award:


Document Type:


Degree Name:

Doctor of Philosophy (PhD)




Marvin F. Daley


In Experiment I, contingency management was employed with five non-institutionalized trainable retardates (mean MA=4.l, CA=9.9, and language age=2.6) in a classroom situation. Empirically determined high probability behaviors were displayed as colored cartoon figures in a reinforcement menu. Tasks from the Peabody Language Development Kit defined the behavior strengthened during 15 one-hour sessions. A quantity of low probability task behavior had to occur in order for subjects to emit 4 minutes of high probability reinforcing activity. Through contingency management, amount of task per reinforcement was shaped from a low ratio to a high ratio. Significant gains in language age (p

In Experiment II, the five children of Experiment I and another trainable retardate served as subjects. Dependent upon the behavioral change desired of the subject in the classroom situation, each subject had to meet the criterion of increasing or decreasing the specified behavior to receive the reinforcement menu and engage in 4 minutes of high probability reinforcing activity. Contingency management over 15 one-hour sessions was successful in reducing and possibly extinguishing the rate of aggressive acts, delays in starting work, shouting, leaving the desk during lessons, and refusing to obey teacher instructions. Frequency of task relevant vocalizations was also increased by this technique. During an unsystematic observation of the subjects 6 months after termination of the experiment, the undesirable behaviors occurred once an hour on the average, a rate far below the baseline frequency. Task relevant vocalizations which had been shaped to a high frequency during the experiment were emitted at a rate higher than the rate obtained during the baseline period.