Self-Supervised Learning in Self-Touch (Follow Up Experiment)

The present study aims to investigate the flexibility of touch and movement as supervisory signals for estimating spatial distance of a self-touch movement, examining whether one signal consistently dominates or whether their supervisory roles are flexible. In previous experiments based on between participants design (unpublished) we investigated how internal models of spatial perception are acquired and updated through self-touch and visual feedback using a robot set up. Manipulating the motor:tactile (M:T) gain in a robotic self-touch setup leads to temporary recalibration between movement and tactile perception, producing an aftereffect in which perception was biased based on the learned movement-touch relationship. Whereas visual feedback reduced the supervisory effect of movement on the judgement of the spatial distance of touch, the reverse did not seem to be the case. That is, visual feedback did not modulate the aftereffects when participants had to judge the length of the movement. Based on this, one could hypothesize that touch and movement’s supervisory roles are flexible, with touch being a stronger supervisory signal than movement. To test this, the current study will use a within-subject design, in which each participant judges spatial extent in both sensations and in two conditions (gain and gain + feedback). This design minimizes the influence of individual differences, allowing for more direct comparisons across conditions and sensations.