The Effects of Suggestibility and Expectancy on Systematic Desensitization

Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)




The present study investigated the effects of subject suggestibility and instruction induced expectancies in systematic desensitization. It was hypothesized that subject suggestibility, instruction induced expectancies, and the interaction of suggestibility and expectancies would affect the outcome of systematic desensitization. Students in introductory psychology classes were asked to fill out the Suinn Test Anxiety Behavior Scale (STABS). The Barber Suggestibility Scale (BSS) was later administered to 105 high STABS scorers. High and low BSS scorers were then selected for further participation in the study. High and low suggestibility subjects were randomly assigned to each of the three treatment groups: the high expectancy desensitization group, the low expectancy desensitization group, and the no-treatment control group. Ten subjects were assigned to each of the six cells. Subjects in all of the desensitization groups received identical treatment except that the high and low expectancy subjects received different expectancy instructions. Those in the high expectx ancy groups were told that they were receiving an effective treatment for test anxiety. No reference was made to treatment or to test anxiety in the instructions given the low expectancy groups. They were simply told that they were in a study of physiological reactions. All of the desensitization subjects received four tape recorded relaxation training sessions followed by the number of desensitization sessions necessary to complete a 17-item test anxiety hierarchy. Pre- and postmeasures were obtained for three outcome measures: the STABS, a heart rate measure, and scores on exams in their introductory psychology course. The heart rate measure included a one minute resting sample and a one minute sample while imagining a scene designed to arouse test anxiety. The premeasure for the course exam variable included scores on two exams which all subjects had completed prior to the treatment phase of the experiment. The postmeasure for the course exam variable was the final exam in the introductory psychology course. A postexperiment interview was used to ascertain the efficacy of the expectancy instructions.

The data was analyzed by 3 X 2 analyses of variance with covariant adjustments made for pretest scores. For the heart rate measure, successive covariant adjustments were made with the resting prescores covaried out of the anxiety scene prescores and the resting postscores covaried out of the anxiety scene postscores and the pretreatment residuals then covaried out of the posttreatment residuals. Analysis of the STABS scores showed a significant (p<.01) treatment effect but neither the suggestibility effect nor the interaction between suggestibility and treatment reached statistical significance. Internal comparisons showed that the STABS scores of the high expectancy group had been reduced significantly (p<.01) more than those of low expectancy desensitization group and significantly (p<.Q01) more than those of the no-treatment control group but that the STABS scores of the low expectancy desensitization group did not differ from those of the no-treatment control group. The original analysis of the heart rate scores showed no significant differences. A separate analysis of just the four systematic desensitization groups showed a significant (p<.05) expectancy effect. Analysis of the course exam change scores showed no significant differences between groups. These results were discussed in the context of prior theory and research and some implications for clinical practice were suggested.

This document is currently not available here.