3 - (SYM 97) Positive Affect Moderates Expectancy Violation During Exposure Therapy for Social Anxiety

There is some evidence to indicate that comorbidity with depression undermines the efficacy of exposure therapy for anxiety disorders, although evidence exists to the contrary as well. Anhedonia is a transdiagnostic symptom dimension that has historically been considered in relation to depression symptoms, but is also highly related to anxiety. Of note, anhedonia is associated with diminished prediction error signaling, a central mechanism of exposure therapy. Low positive affect, a core feature of anhedonia, is further associated with aberrant fear extinction learning and retrieval and may reduce prediction updating following the non-occurrence of an expected negative outcome. The goal of the present study was to examine if low positive affect is associated with learning mechanisms of exposure therapy among individuals with social anxiety disorder.

The present study was conducted as part of a randomized-controlled trial comparing within-session fear reduction and inhibitory retrieval learning models of exposure therapy. 64 socially anxious individuals completed a nine-session exposure therapy protocol. Of this group, 32 participants underwent treatment focused on inhibitory retrieval learning, during which they rated expectancy of a negative outcome before and after each exposure. 32 participants underwent treatment focused on within-session fear reduction, during which they rated subjective experience of fear before and after each exposure.

In the Inhibitory Retrieval Learning group, positive affect moderated the relationship between pre- and post-expectancy (p=.049), such that individuals with low positive affect updated expectancies less when pre-expectancy was highest (e.g., 80-100%). Negative affect also moderated the relationship between pre- and post-expectancy (p=.033). Specificity analyses indicated that, over and above negative affect, within-participants variability in positive affect moderated the relationship between pre- and post-expectancy (p=.011). No effects of negative affect were significant when controlling for positive affect. In the fear reduction group, neither positive affect nor negative affect moderated the relationship between pre- and post-fear.

Within-person levels of positive affect are associated with updating of expectancy following the non-occurrence of highly anticipated negative outcomes. Exposure therapy may incorporate positive affect interventions to maximize inhibitory retrieval learning.