A novel virtual reality fear conditioning paradigm to investigate the influence of expectancy violation on fear extinction

Exposure therapy is an efficient treatment for pathological anxiety, yet its underlying mechanisms are not fully understood. Prediction error models suggest that optimizing the violation of threat-related expectancies improves treatment outcomes, however, causal evidence is still sparse. The aim of the current study was therefore to provide causal evidence for the influence of the extent of expectancy violations on extinction retention using a novel virtual reality fear conditioning paradigm. In total, 100 participants completed a two-day fear conditioning paradigm in which the approach behavior of an animated stimulus was differentially reinforced with an electrical stimulus (i.e., closer distances were associated with a higher probability for receiving the aversive stimulation). To experimentally manipulate the extent of expectancy violations during fear extinction, participants were presented only with distances to the conditioned stimulus that either weakly (i.e., far distances) or strongly predicted the aversive outcome (i.e., close distances), resulting in low vs. high expectancy violations. We found successful fear acquisition and extinction, as well as an influence of the extent of expectancy violations on US-expectancy and threat ratings after extinction on day 1. On the second day, at a spontaneous recovery and reinstatement test, however, there was only weak evidence for improved extinction retention in the high expectancy violation condition. Optimizing expectancy violations might be a necessary but not sufficient condition for improved extinction retention. Future research needs to address the conditions under which expectancy violations lead to robust expectancy adjustments and how these conditions can be met in exposure therapy for anxiety disorders.