Look at the Audience? A Randomized Controlled Study of Shifting Attention From Self-Focus to Nonsocial vs. Social External Stimuli During Virtual Reality Exposure to Public Speaking in Social Anxiety

Background: Enhanced self-focused attention plays a central role in the maintenance and treatment of Social Anxiety and is targeted in contemporary cognitive behavioral therapy. Actual developments use Virtual Reality (VR) for behavioral training. However, no VR attention training combining exposure to public speaking with shifting attention from self-focus to external focus has been investigated, and no experimental evidence exists on different kinds of external cues as targets of attention. Therefore, we investigated the effects of an attention training during public speaking in VR and examined differential effects of an external focus on nonsocial vs. social stimuli. Methods: In this randomized controlled study, highly socially anxious participants were instructed to focus on either objects or the audience within a virtual speech task. We assessed the pre-post effects on affective reactions, self-perception, and attentional processes during public speaking as well as general Social Anxiety using subjective, physiological, and eye-tracking measures. Repeated-measures analyses of variance (ANOVAs) were calculated to detect changes from pretest to posttest over both groups, and time × group interaction effects. Results: Within the analysis sample (n = 41), anxiety during public speaking and fear of negative evaluation significantly decreased, with no significant differences between groups. No significant time effect, but a significant time × group effect, was found for the looking time proportion on the audience members' heads. Follow-up tests confirmed a significant increase in the social-focus group and a significant decrease in the nonsocial-focus group. For all other variables, except external focus and fear of public speaking, significant improvements were found over both groups. Further significant time x group effects were found for positive affect during public speaking, with a significant increase in the social focus, and no significant change in the nonsocial-focus group. Conclusion: Our findings suggest that attention training to reduce self-focus can be successfully conducted in VR. Both training versions showed positive short-term effects in the highly socially anxious, with particular advantages of an external social focus concerning eye contact to the audience and positive affect. Further research should investigate whether social focus is even more advantageous long term and if reinterpretations of dysfunctional beliefs could be achieved by not avoiding social cues.