Emotion Recognition in Virtual Reality: Investigating the Effect of Gameplay Variations on Affective Responses

Abstract

Virtual reality (VR) technology offers a unique platform for affective computing research, which aims to create intelligent systems capable of recognizing, interpreting, and responding to human emotions in immersive and interactive virtual environments. This study examined how different variations of VR games can elicit different emotional responses in players. Thirty participants played three VR Pong games, and their emotional responses were measured using the Self-Assessment Manikin and physiological measures. Results showed that different gameplay variations evoked varying emotional responses in players, and two features, average heart rate and the non-specific skin conductance responses (in peaks per min), were found to differ significantly between the games. The study suggests that VR gameplay can provide an engaging and interactive platform to study human emotions and behavior, with implications for designing and developing affective computing systems in fields such as healthcare, marketing, and humancomputer interaction.