A relapse in clinical anxiety following exposure therapy for social anxiety disorder is prevalent and causes serious problems. According to the fear conditioning theory of social anxiety disorder, a part of this relapse can be caused by the renewal effect. This study aimed to investigate whether three renewal effects occur in a fear conditioning procedure that uses social stimuli as both unconditioned and conditioned stimuli, which is an analog preparation of acquisition of social anxiety and reduction by exposure therapy. Sixty-four participants were randomly allocated to four groups (AAA, ABA, ABC, and AAB). They received 9 pairings with a natural face and a negative comment during the acquisition phase and then received 18 pairings with the same face and a neutral comment from the person in the extinction phase. Following extinction, the testing phase was conducted. Context, defined as background colors, used in each phase was different between groups. We conducted two analyses, the ANOVA and Bayesian modeling, to investigate whether three types of renewal effects occur, whether the individual data can be described by an associative model and whether individual differences in learning are related to social anxiety. The ANOVA showed the occurrence of three renewals in the procedure, although the size of their effects was the same. The Bayesian modeling indicated that individual data were generally consistent with the model, and there were some relationships among the estimated parameters and between their parameters and social anxiety. These findings suggest that the relapse following exposure therapy is related to renewal effects, the effect of each exposure session can be represented by mathematical associative models, and some features of learning in the procedure are related to other features and social anxiety.
Elevated Fear Conditioning to Socially Relevant Unconditioned Stimuli in Social Anxiety Disorder