Pupillometric Analysis of Cognitive Load in VR Environments: The Impact of Visual and Auditory Fidelity

This study investigates the effects of auditory and visual quality in virtual reality (VR) environments on cognitive load and engagement through pupillometric measurement. It provides a comprehensive consideration of their influence on physiological processes collectively. The experiment controlled both auditory (low-fidelity muffled, normal dynamic, and high-volume monotonous) and visual (less detailed, normal, and richly detailed) stimuli systematically in three VR museum environments. Pupil diameter measurements were employed as primary indicators of cognitive processing, with rigorous preprocessing to ensure data integrity. For visual conditions, low-quality settings with less detail consistently generated maximum stress responses, while normal-quality visuals were determined to optimize the balance of engagement and comfort. High-quality visuals in dim settings had slightly more engagement but had to be calibrated precisely to avoid overstimulation. The audio results indicated that normal-quality dynamic sound produced the most stable pupil responses across all visual contexts and served as a cognitive anchor. Poor-quality audio consistently degraded experience quality, supporting the fact that visual richness cannot compensate for poor auditory design. Notably, combining normal-quality audio with normal-quality visuals provided the optimal physiological profile. These findings guide theoretical advancement in multisensory integration in VR and offer design recommendations of utility: (1) prefer balanced modulation of the senses, (2) avoid visual extremes of simplicity with high luminance, and (3) seek harmonious mixtures of normal-fidelity sensory and visual content. Future research needs to explore individual differences and long-term adaptation in order to further refine these guidelines for a variety of educational, training, and therapy uses in VR.