Physiology-based Virtual Reality Training for Social Skills in Schizophrenia

Social impairments present a major barrier toward good outcome in schizophrenia (SZ). Social skills and social competence protect against stress-related exacerbation of symptoms while supporting interpersonal interactions and social affiliation that help enhance the quality of life. Improving social functions could lead to better outcome and reduce healthcare costs and societal burden. Pharmacotherapy has shown to be inadequate in improving social impairments in schizophrenia, and although existing psychosocial interventions can be moderately effective, the time- and effort-intensive nature (high-burden), low adherence, and weak transfer of skills to real life present significant hurdles. Thus, there is a strong need to develop effective and low-burden social interventions that can be personalized to optimize learning and increase adherence.

Virtual reality (VR) technology offers a very promising alternative to conventional interventions, with several advantages. Controllable complexity of the VR world provides a simplified and low-stress social interaction environment with targeted opportunities to simulate, exercise and reinforce basic elements of social skills and to repeat exposure to naturalistic situations from multiple angles. Simulation of a wide variety of social situations that allows for targeted practice is crucial for learning and transfer of skills. In our proposed VR system, the capacity for personalized social skills training is achieved by automatic and adaptive feedback, based on the affective and attentional states of the participant in real-time, using closed-loop, emotion-sensing VR technology that integrates physiological signals, eye-tracking data and performance data from the participants to rapidly adjust the task parameters. This approach enables realistic, and adaptive interactions in VR games to provide scaffolded social skills exercises in an engaging and enjoyable manner.

The major aim of this proposal is to implement a personalized VR-based social skills training program with high efficacy for individuals with SZ by capitalizing on innovations in adaptive, emotion-sensing technology, designed to increase engagement and learning. This work will be conducted in two phases. The R21 phase will focus on implementing an adaptive social intervention VR game, optimizing the dose and measuring targeted social cognitive outcome. Upon meeting the milestones of the R21 (see 'Approach'), the R33 project will evaluate the feasibility of recruitment, and retention of SZ participants in a pilot randomized controlled trial (RCT) of VR training compared with an active control condition and assess long-term outcome.

The long-term goal of this proposal is to develop and implement a personalized social intervention method for individuals with schizophrenia that is efficacious, low-burden, enjoyable, and with generalizable and enduring benefits. Within the R21 phase, a physiology-based, closed-loop, adaptive VR task will be developed to determine the optimal dose for improving targeted social functions in individuals with SZ. If these goals are met for the R21 phase, a pilot RCT will be conducted in R33 to compare the social, cognitive, neural effects of adaptive social VR versus an active control condition. Furthermore, the physiological and eye tracking data and the outcome data from both R21 and R33 phases will be used to refine and improve this VR protocol and technology. It is hoped that the results of this study will lead to a large scale RCT in the future, and the implementation of a portable, low-cost and widely available intervention that can be personalized for optimal learning. In short, the investigators will implement an efficacious, engaging, low-burden, VR intervention to ameliorate social impairments of SZ and improve functional outcome.