Reduced feedback barely slows down proprioceptive recalibration

Abstract

Introducing altered visual feedback of the hand results in quick adaptation of reaching movements. And while this may be partly due to explicit strategies, our lab has shown that implicit changes like reach aftereffects and shift in estimates of the unseen hand, can also emerge and even saturate within a few training trials. The goal of the current study is to determine whether these rapid changes in unseen hand position that occur during classical visuomotor adaptation are diminished or slowed when feedback during training is reduced. We reduced feedback by either providing visual feedback only at the end of the reach (terminal feedback) or constraining hand movements to reduce efferent contribution (exposure). We measured changes as participants completed reaches with a 30{degrees} rotation, a -30{degrees} rotation and clamped visual feedback, with these two "impoverished" training conditions, along with classical visuomotor adaptation training, while continuously estimating their felt hand position. Classic continuous-cursor training produced exemplary learning curves and rapid and robust shifts in felt hand position. Training with terminal feedback slightly reduced the initial rate of change in overall adaptation and but not the magnitude of shifts in felt hand position. Finally using a robot to constrain and deviate hand movement direction, called exposure training, only delayed saturation of proprioceptive changes by a single trial and these changes were slightly smaller than those during classical training. Taken together, adaptation and shifts in felt hand position are a rapid and robust responses to sensory mismatches and are only slightly modulated when feedback is reduced. This means that, given a visuo-proprioceptive mismatch, the resulting shift in sense of limb position can contribute to movements from the start of adaptation.