Self-efficacy as a predictor of behavior change: Interaction with type of training for pain tolerance

The relationship between self-efficacy expectations and the training effects of skill-based and non-skill-based procedures for pain reduction was investigated. Both pretraining self-efficacy magnitude and strength measures, as well as changes in self-efficacy magnitude and strength, were employed to predict posttraining behavioral tolerance and subjective discomfort to a cold pressor pain stimulus. Eighty university women received brief training in either skill-based (relaxation, attention diversion, and self-instruction skills) or non-skill-based (rationale, exposure, and demands for change) procedures. Results indicated that pretraining magnitude, as well as change in magnitude scores, significantly predicted posttraining tolerance and discomfort for the nonskills group. Changes in self-efficacy magnitude and strength both predicted posttraining tolerance for the skills group. For the discomfort measure, the nonskills procedure produced significantly lower scores than the skills procedure for subjects who demonstrated relatively large increases in self-efficacy magnitude. All predictive relationships in this study were independent of pretraining performance levels. The implications of these data for the selection and treatment of subjects is considered.