Investigating the role of sensorimotor spatial dependencies in shaping conscious access to virtual 3D objects

According to sensorimotor accounts of perceptual experience, the subjective veridicality of an object (the sense of its ‘presence’) builds up gradually as one learns how changes in sensory inputs depend on bodily movements. To investigate how sensorimotor interactions shape visual experience, we designed a virtual-reality-based study that allowed us to manipulate the complexity of spatial dependencies governing interactions with unfamiliar 3D objects. Participants had to learn to manually control fully visible objects that could move in congruent, opposite, novel (orthogonal), or random directions in response to their movements. The sensorimotor control tasks occurred alternately with a continuous flash suppression (CFS) task evaluating the access of stationary objects to visual awareness, operationalised as the time taken for a 3D object to break the interocular suppression. We hypothesised that objects whose motion was experienced as depending on actions in a lawful, and thus encodable, manner (i.e., according to a congruent, opposite, or novel – but not random – dependency) would overcome suppression faster than objects moving randomly in response to actions (for which there is no world-related statistical structure to learn). However, while performance in the sensorimotor tasks consistently decreased along with the difficulty of the conditions (i.e., congruent > opposite > novel > random), the pre-registered analysis yielded no significant differences in breakthrough times of objects manipulated under different coupling rules. An exploratory analysis assessing whether the acquisition of ‘sensorimotor mastery’ was associated with reduced breakthrough times also revealed no significant effects. Thus, our results suggest that one’s knowledge of how an object responds to action does not play a salient role in determining conscious access to visual stimuli. This extends previous evidence for a general ineffectiveness of sensorimotor spatial manipulations in interocular suppression paradigms. Notably, in all conditions, object movement remained tightly coupled (i.e., contingent) to the participant’s actions – and given such stimuli have already been shown to break suppression faster than uncoupled/pre-recorded visual inputs – it is possible that sensorimotor contingency was a sufficiently salient factor to override any influences related to how identifiable specific spatial dependencies were.