The design, gamification, and physiological evaluation of a VR social interaction simulation

Abstract

Communication skills are essential characteristics required to thrive in a fast-paced and ever-growing society that we currently live in. The ability to socialize in both informal and formal settings is important for the all-around development of individuals. The use of VR simulations for training has provided a novel avenue for individuals to learn such soft skills. In this work, we present a VR-based interaction system designed to simulate four structured social scenarios. The primary objective of this study is to evaluate whether the proposed simulation can elicit measurable physiological responses in users during controlled social interaction tasks. We analyze the participants’ autonomic responses using heart rate variability (HRV) as an index of physiological arousal associated with stress-related responses. Initially, we discuss the various anxiety disorders and expand on social anxiety disorder (SAD). SAD is used here as a theoretical motivation for designing socially evaluative VR environments rather than as a clinical target of evaluation. Thereafter, we study the connection between social skills training (SST) and SAD, as well as the possibility of imparting training as exposure therapy. It is key to note that the present work does not directly evaluate training outcomes or behavioral skill acquisition. Instead, we focus on the role of VR in supporting controlled exposure to social situations rather than demonstrating improvements in confidence or social skills. In addition, we begin by creating a scene through four key scenarios by designing their avatars and adding pre-scripted conversations. Finally, we evaluated the simulation on two key aspects, namely, affective fidelity and construct validity. We use graphical representations to infer that the simulation elicits differentiated patterns of HRV across users and scenarios, and this varies appropriately from one user to another. We therefore infer that the simulation demonstrates its ability to induce and measure stress in structured social interaction contexts, rather than constituting or validating a social skills training intervention.