Arguably, one of the core structural features of the experience of intentionally-directed bodily movement is the presentation of the agent as the “perspectival source” of the motion experienced (Horgan et al. 2003; Marcel 2006). However, a strikingly robust experimental finding is that individuals will correct for a deviation introduced into a movement they perform via a bias in visual input, thereby ensuring the action they intend achieves its goal, whilst nevertheless not reporting such corrections in their movement (Fourneret & Jeannerod 1998; Knoblich & Kircher 2004; Slachevsky et al. 2001). Recent developments of this paradigm have adapted it to test explicit egomotion perspectival experience in walking movements, by using a motion-tracked avatar, observed from the rear. Kannape and colleagues found that by introducing a slight bias into the subject’s visual experience of the trajectory of the avatar, they could induce subjects to perform appropriate corrective movements in walking to a target, whilst not noting the discrepancy between their actual movements and the avatar (Kannape et al. 2010). Again, the corrections went largely unnoticed within a certain range of angular deviation between observed and actual movements.[9] Thus, a natural explanation of the pattern of data is that the mechanisms enabling the experience of agency present bodily movements in a manner that is far more coarse-grained than the level of detail required to make corrective changes in movement trajectory. In short, egomotion perspectives structure experiences of intentionally-directed bodily movement. They do so by specifying what we might call coarse-grained phenomenal grooves, within which a movement must unfold if it is to seem like the movement that the subject intended or is trying to perform.
Strangely, as yet the potential contributions of the vestibular system to the structuring of agentive experience by egomotion perspective have not been manipulated. Moreover, as noted, the work that has been done in this area has been restricted to explicit egomotion perspectival phenomena. A natural further step would be to investigate the nature of vestibular processing in implicit egomotion perspective, by controlling a participant’s optic flow in a manner corresponding to the control of the avatar’s motion in Kannape and colleagues’ original study.
Experiment 4: Study 1: Participants view a textured environment via HMD in which optical flow fields are regulated by their motion-tracked movements. Study 2: Participants control a motion-tracked, real-time avatar seen from behind. In both studies, participants are tasked with walking directly towards a virtual target. All the while, they either receive GVS or sham stimulation and visual feedback (optic flow or avatar position) that is either faithful to motion-tracking or systematically deviated left/right of the participant’s mid-line, as a function of distance from a point of displacement onset.
Participant trajectory could thus be compared to the dynamics of the flow field or avatar trajectory and participants could be asked to rate the degree to which their movements in the virtual environment or the movements of the avatar corresponded to their actual movements, as respective measures of implicit and explicit egomotion perspectival experience. The question would be whether, in trials in which GVS is applied, the range of angular deviation in which participants would judge that movements in the virtual environment correspond to their own would be equal to or larger than trials in which participants receive only biased visual feedback. If the latter occurs, then in the evocative terms used above, it would suggest that vestibular processes are one of the determinants of the coarseness of the phenomenal groove specified by an egomotion perspective.