Relaxing in virtual reality: one synthetic agent relaxes all

Virtual reality-based interventions have gained attention as innovative approaches in clinical settings. However, the use of virtual-based relaxation in reducing psychological distress and physiological activation, a common strategy in traditional interventions, is not well documented. This study aims at exploring the role of a non-familiar synthetic agent (SyA) as a resource to promote relaxation in a virtual environment (VE). Sixty-nine healthy participants were randomly assigned to three conditions: relaxing in VE while listening to relaxing instructions delivered by a SyA (n = 23), relaxing in the same VE while listening to the same instructions but aired by a radio set (n = 23; ‘active’ control group) and waiting to the end of the experience without relaxing instructions (n = 23; ‘passive’ control group). The instruction was preceded by an activation task (i.e., a matching game within a limited time). Our hypothesis claims that the presence of a humanoid-like figure that is strange to the participant (SyA) hinders the relaxing process. Data from several self-reports (Presence, Immersion, Cybersickness and emotional response) and from psychophysiology (respiratory rate—RR) revealed that no differences were found between the two groups that listened to the relaxing instructions (SyA and radio). Additionally, a significant decrease in RR recordings was only significant for these two relaxation conditions (SyA and radio), but not for the ‘passive’ control group. Results suggest that the presence of a non-familiar humanoid character was not perceived as a dissonant element in the VE setting and did not negatively influence the relaxation outcome. This study sets the ground for future studies that may provide an insight into the optimal characteristics of a SyA, contributing to the development of accessible and beneficial digital applications to a wide range of individuals in clinical and non-clinical contexts.