ABSTRACT
BACKGROUND AND HYPOTHESIS: Social cognition training (SCT), an intervention for social cognition and social functioning, might be improved by using virtual reality (VR), because VR may offer better opportunities to practice in a potentially more realistic environment. To date, no controlled studies have investigated VR-SCT. This study investigated a VR-SCT, "DiSCoVR". We hypothesized that DiSCoVR would improve social cognition and social functioning. STUDY DESIGN: Participants were randomized to DiSCoVR (nâ =â 41) or VR relaxation ('VRelax', nâ =â 40), an active control condition, and completed 16 twice-weekly sessions. Three assessments (baseline, posttreatment, and 3-month follow-up) were performed by blinded assessors. The primary outcome was social cognition (emotion perception and theory of mind). Secondary outcomes included social functioning (measured with an interview and experience sampling), psychiatric symptoms, information processing, and self-esteem. Data were analyzed using mixed-models regression analysis. Treatment effects were evaluated by the time by condition interaction terms. STUDY RESULTS: No significant time by condition interactions were found for any of the outcome variables, indicating an absence of treatment effects. Between-group effect sizes ranged from negligible to moderate (Cohen's dâ <â |0.53|). Main effects of time were found for several outcomes. CONCLUSIONS: These results suggest that DiSCoVR was not effective, possibly because of inadequate simulation of emotional expressions in VR. This lack of efficacy may indicate that current SCT protocols are relatively unsuitable for improving social functioning. Previous studies showed small to moderate effects on higher order social cognition, but the SCT approach may need critical reevaluation, as it may not sufficiently lead to functional improvement.
