Modelling therapy for test anxiety: The role of model affect and consequences

This study assessed the effectiveness of modelling therapy as a treatment for test anxiety and investigated the role of model affect and consequences in determining that effectiveness. Twenty-four test anxious subjects were shown one of four one-hour modelling displays of intelligence testing sessions. The displays represented combinations of calm or anxious model affect and positive or negative consequences. Two self report and two performance measures of test anxiety were analyzed by means of a 2 × 2 vs. control group factorial design. The results indicated an overall significant improvement on modelled intelligence tests for the experimental groups vs. the control group. Differential effects were also found depending upon the characteristics of the modelling display. Viewing anxious models lead to a significant increase in introductory psychology grades and viewing negative consequences lead to a decrease in self-reported test anxiety. The present results question the notion that the most effective modelling scenes are ones displaying calm models and positive consequences (Bandura, 1969). Parallels were drawn between the results of the research and the recent success of implosively-oriented behaviour therapies.