A Study on Baseline Estimation Methods for Emotion Evaluation Using Physiological Indices During VR Experiences

Physiological indices such as electroencephalography (EEG) and heart rate variability (HRV) have attracted increasing attention as objective measures for evaluating emotional responses during experiences in virtual reality (VR) environments. However, to ensure the reliability of such evaluations, it is essential to appropriately determine the baseline that serves as the reference for emotional assessment. This study proposes two baseline estimation methods—the Pre-Stimulus Averaging Method and the Resting-State Prediction Method—which account for the effects of habituation commonly observed during VR viewing. An analysis using physiological data from 41 participants demonstrated that, in particular, the Resting-State Prediction Method resulted in the VR environments being evaluated as eliciting higher arousal compared to real environments.