Cybersickness, sensory feedback, and embodiment in XR–BMI systems

Abstract Embodiment is a core construct in immersive human–computer interaction, yet it is often assumed that increasing multisensory feedback linearly enhances embodied experience. Using an extended reality brain–machine interface (XR–BMI) task, we investigated whether sensory load modulates embodiment and whether affective, experiential, and dispositional factors provide a more informative psychological account of individual differences. Participants performed a BMI task under visual, audiovisual, and audiovisual-tactile conditions, while embodiment subcomponents (ownership, agency, self-location), cybersickness, anxiety, personality traits, and task performance were assessed. Contrary to prevailing assumptions, no statistically significant differences in global embodiment or its subcomponents were observed across sensory conditions. Complementary Bayesian analyses provided moderate evidence favoring models that excluded an effect of sensory condition on embodiment. Instead, cybersickness showed the most consistent negative association with reduced embodiment across conditions. This association remained significant in combined sensitivity analyses, although its independent contribution in adjusted models was attenuated. Mediation analyses identified an indirect statistical pathway consistent with the proposed model, whereby anxiety was associated with lower embodiment via increased cybersickness. Stable personality traits showed limited explanatory value beyond cybersickness. Global embodiment was not associated with BMI performance, although exploratory analyses suggested that spatial self-location may influence the relationship between cybersickness and task performance. Overall, these findings suggest that, under the present experimental conditions, cybersickness is a more consistent correlate of embodiment than increasing sensory load, although smaller effects of sensory enrichment cannot be ruled out. These findings highlight the importance of minimizing cybersickness and promoting sensorimotor coherence when designing immersive XR–BMI systems.