Date of Award:


Document Type:


Degree Name:

Doctor of Philosophy (PhD)



Committee Chair(s)

Michael Bertoch


Michael Bertoch


Edward K. Crossman


Reed S. Morrill


C. Jay Skidmore


E. Wayne Wright


This study evaluated the use of a token, operant conditioning technique as a treatment procedure in a group setting with chronic, hospitalized, psychiatric patients. Fifteen patients were randomly assigned to two experimental groups and to one control group; each group included five patients. The two experimental groups received tokens during phases of contingent reinforcement for "Therapeutic Responses" and under a yoked-control phase of non-contingent token reinforcement; the sequence of these phases was the major difference between the two experimental groups. The control group met under conditions of no tokens, and the frequency of "Therapeutic Responses" was recorded on those subjects.

"Therapeutic Responses" were characterized as open, confrontive, and problem-solving oriented interaction among group members as defined by Quadrant IV of the Hill Interaction Matrix (HIM).

Results clearly supported the hypotheses that therapeutic responses did occur significantly more frequently and for longer durations in both the experimental groups in the following predicted directions: (1) under conditions of contingent token reinforcement as compared to conditions of non-contingent token reinforcement and to baseline and extinction phases; (2) in both of the experimental groups as compared to the control group. Reversal procedures demonstrated the expected extinction effects.

Also, all three groups were compared on follow-up outcome data which included: (1) pre- and post-test ratings by hospital attendants of the group member's behavior changes on the ward as measured by The Behavioral Adjustment Scale; (2) pre- and post-test scores on the HIM-B (an unpublished instrument based on the HIM measuring attitudes toward group interaction). There were no differential effects among the three groups' MACC Behavioral Adjustment Scale and HIM-B posttest scores.

These results were discussed in respect to the implications of applied research in the area of verbal conditioning, and implications for the treatment of chronic psychiatric patients. Recommendations for future research to examine possible generalization effects were offered.



Included in

Psychology Commons