The Bayesian Causal Inference of Body Ownership Model: Use in VR and Plausible Parameter Choices

Abstract

Experiencing virtual body ownership is an important component of user experience in virtual reality applications with embodied avatars. A functioning model of body ownership could allow designers of such applications to predict the occurrence of body ownership illusions in users. One attempt at such a model, the Bayesian Causal Inference of Body Ownership (BCIBO) model, explains body ownership as inference about the causes of sensory signals. When sensory signals under consideration (e.g. tactile and visual signals) are attributed to a single object (e.g. a rubber hand), then this object is incorporated into the body. We investigate an unrealistic choice of parameter values in the original specification of the BCIBO model and make some suggestions for improvements.