Virtual Reality Social Cognition Training Not Effective in Schizophrenia

A virtual reality (VR)-based social cognition training (SCT) intervention was not found to be effective at improving social cognition and functioning among patients with schizophrenia. These findings were published in the Schizophrenia Bulletin.

This single-blind study was conducted at 5 mental health care centers in the Netherlands between 2018 and 2020. Patients (N=81) with a diagnosed psychotic disorder with deficits in social cognition were randomly assigned to receive Dynamic Interactive Social Cognition Training in Virtual Reality (DiSCoVR; n=41) or VR relaxation therapy (VRelax; n=40).

Both interventions comprised 16 biweekly 45 to 60-minute sessions delivered on site. The DiSCoVR program included 3 modules covering facial affect recognition, social perception theory of mind, and applications of social cognition in social interactions. The outcomes of interest were changes to the Awareness of Social Inference Task (TASIT) and Personal and Social Performance (PSP) scores.

VRelax sought to relax participants, help them cope, and reduce stress. Before the VR sessions participants were taught about stress, coping, rumination, and stress reduction. Half of each session focused on VR relaxation with participants looking at relaxing VR videos of scenes from nature. In addition, VR relaxation exercises with audio guidance were shown while looking at relaxing nature scenes.

The DiSCoVR and VRelax cohorts included patients who had a mean age of 35.9 (SD, 10.4) and 39.7 (SD, 12.4) years, 73.2% and 65.0% were men, 61.0% and 59.0% were diagnosed with schizophrenia, 53.7% and 65.0% lived independently, and 58.5% and 60.0% had a family history of psychosis, respectively.

No time-by-condition outcomes were significant, indicating no treatment effects. Similarly, nearly all between-group effect sizers were small. However, the effect sizes for Perceived Stress Scale (PSS; d, 0.59), Positive and Negative Syndrome Scale (PANSS)-Negative (d, 0.41), and Beck Anxiety Inventory (BAI; d, 0.33) scores tended to favor DiSCoVR whereas the experience sampling method (ESM) positive affect favored VRelax (d, -0.38).

Significant time effects were observed for the Self Esteem Rating Scale (SERS; P =.002), ESM (P <.001), PSP (P =.027), and the Ekman 60 Faces test (P =.019) scores.

An analysis of blinding found that the blinding was unsuccessful (P =.033).

A total of 2 patients in the DiSCoVR and 1 patient in the VRelax group had a serious adverse event of psychiatric hospitalization. However, all were deemed unrelated with the study. The DiSCoVR recipients reported more nausea than the VRelax recipients (P =.036).

The limitations of this study included interruption due to the COVID-19 causing the planned sample size to not be reached and a higher dropout rate (28%) than anticipated (~13%).

Study authors concluded, “In this randomized trial, we compared a VR-SCT (DiSCoVR) to an active control VR relaxation condition (VRelax), to investigate its effects on social cognition, social functioning, and other clinical outcomes. No significant treatment effects were found. An analysis of completers showed no relevant differences from the intention-to-treat analysis.”

Disclosure: An author declared affiliations with industry. Please refer to the original article for a full list of disclosures.