An experimental analysis of generalization in assertive training

Unassertive psychiatric patients matched on age, diagnosis, and self-reported assertiveness were assigned to one of five conditions, with 10 subjects in each group: Test-Retest, Practice-Control, Practice-Control with Generalization Instructions, Modeling and Instructions, Modeling and Instructions with Generalization Instructions. All subjects were videotaped (Pre- and Post-Test) while responding to ten interpersonal situations requiring assertive responses. Following the Pre-test, subjects in the experimental groups received training on five of the ten situations (training scenes). On the remaining five situations subjects received no training (generalization scenes). Results indicated that the Modeling and Instructions groups effected the greatest changes on seven of the eight components for the training scenes but on only five of the eight components for the generalization scenes. No differences were found between either of the two Practice-Control groups and the the Test-Retest group. Although transfer of training effects were obtained on the generalization scenes, these effects were minimal on an independent in vivo test of generalization. Moreover, an instructional set designed to facilitate generalization to different situations was only partially effective.