Evoking stress reactivity in virtual reality: A systematic review and meta-analysis

Virtual reality (VR) research probes stress environments that are infeasible to create in the real world. However, because research simulations are applied to narrow populations, it remains unclear if VR simulations can stimulate a broadly applicable stress-response. This systematic review and meta-analysis was conducted on studies using VR stress tasks and biomarkers. Included papers (N = 52) measured cortisol, heart rate (HR), galvanic skin response (GSR), systolic blood pressure (SBP), diastolic blood pressure (DBP), respiratory sinus arrhythmia (RSA), parasympathetic activity (RMSSD), sympathovagal balance (LF/HF), and/or salivary alpha-amylase (sAA). Effect sizes (ES) and confidence intervals (CI) were calculated based on standardized mean change of baseline-to-peak biomarker levels. From baseline-to-peak (ES, CI), analyses showed a statistically significant change in cortisol (0.56, 0.28–0.83), HR (0.68, 0.53–0.82), GSR (0.59, 0.36–0.82), SBP (.55, 0.19–0.90), DBP (.64, 0.23–1.05), RSA (−0.59, −0.88 to −0.30), and sAA (0.27, 0.092–0.45). There was no effect for RMSSD and LF/HF. VR stress tasks elicited a varied magnitude of physiological stress reactivity. VR may be an effective tool in stress research. •Across a range of biomarkers, VR worked at inducing a stress response for the majority of biomarkers.•Sympathetic biomarkers had especially large effect sizes, contrary to expectations.•Cortisol showed consistent effect sizes, but not as large as sympathetic biomarkers.•Even relatively inexpensive immersive VR system were efficacious.