Making 'sense' of agency: the bodily self has a time and place

How does the brain distinguish between the signals it produces and the sensations it registers from the environment? To shed light on this, it was investigated if the human mind could be capable of perceiving an avatar’s body in a third person game as one’s own. To create precise, high-quality motion simulations, we uniquely combined a Virtual Reality-setup (Valve Index) with real-time motion capture (Vicon). In doing so, we systematically explored if (predictions of) self-actions allow the brain to infer self-location and to distinguish the body from the environment including other agents. A full-body illusion paradigm (FBI) was developed in VR with three movement conditions: (A) a standard, passive FBI in which people had no motion control; (B) an active FBI in which they made simple voluntary movements; and (C) an immersive game in which they real-time controlled a full-sized human avatar in third person (i.e., the first third person VR-game). Systematic comparisons between measures (implicit, explicit, exit-interview, and temporal binding) revealed a causal relationship between (i) sense of agency, (ii) self-other identification, and (iii) the ability to locate oneself. A loss in sense of agency was reported when movement was restricted, and a shift in self-location and self-identification towards the virtual body was experienced; which did not happen when healthy volunteers were (to some extent) able to voluntarily move. It is confirmed that motor predictions are salient cues for the brain that not only provide a sense of control in self-actions, but also recognition of the self in time and place. People can recognise their movements in a third-person avatar and psychologically align with it (action observation); but do not seem to lose a sense of place (self-location), time (temporal binding), nor who they are (self vs. other), because voluntary action codes the bodily self to a physical location in space. These results provide further evidence for our hypothesis (de Boer et al., 2020) and may shed light on how bodily self-consciousness is constructed. In the future, immersive game simulations could target and strengthen the brain’s control networks in psychosis, neurodegeneration (e.g., dementia, movement disorders) and old age. In addition, real-time motion simulations could help advance future rehabilitation techniques (e.g., to treat nervous system injury) by fine-tuning and personalising the therapeutic setting on demand.