Trier Social Stress Test in vivo and in virtual reality: Dissociation of response domains

- The Trier Social Stress Test (TSST) was successfully implemented in virtual reality (VR).
- There was a dissociation in the reactions in salivary cortisol and the physiological and psychological stress reactions.
- Adding competitor to the avatar in the VR-TSST did not elevate the stress effect.

The Trier Social Stress Test (TSST) is considered a reliable paradigm for inducing psychosocial stress. Virtual reality (VR) has successfully been applied to ensure a greater degree of efficiency and standardization in the TSST. Studies using the TSST in VR (VR-TSST) have reported significant stress reactions, with subjective and peripheral physiological reactions comparable to those in response to the in vivo TSST and with lower cortisol reactions. The current study examined whether an additional virtual competitive factor triggers larger stress responses than a standard VR-TSST. Forty-five male participants were randomly assigned to either in vivo TSST, VR-TSST (VR) or VR-TSST with a virtual competitor (VR +). A significant increase of self-reported stress, electrodermal activity, and heart rate indicated a pronounced stress reaction with no differences between groups. For salivary cortisol, however, responder rates differed significantly between groups, with in vivo participants showing overall higher response rates (86%) than participants of both VR groups (VR: 33%, VR +: 47%). In contrast, participants of both VR groups judged the task significantly more challenging than did in vivo TSST participants. In sum, our results indicate successful stress induction in all experimental conditions, and a marked dissociation of salivary cortisol levels on the one hand, and the physiological and psychological stress reactions on the other hand. The competitive scenario did not significantly enhance stress reactions. VR technology may serve as a standardized tool for inducing social stress in experimental settings, but further research is needed to clarify why the stress reaction as assessed by cortisol differs from peripheral and subjective stress reactions in VR.