Proprioceptive deafferentation slows down the processing of visual hand feedback

Abstract

During visually guided movements both vision and proprioception inform the brain about the position of the hand, so interaction between these two modalities is presumed. Current theories suggest that this interaction occurs by sensory information from both sources being fused into a more reliable, multimodal, percept of hand location. In the literature on perception, however, there is evidence that different sensory modalities interact in the allocation of attention, so that a stimulus in one modality facilitates the processing of a stimulus in a different modality. We investigated whether proprioception facilitates the processing of visual information during motor control. Subjects used a computer mouse to move a cursor to a screen target. In 28% of the trials, pseudorandomly, the cursor was rotated or the target jumped. Reaction time for the trajectory correction in response to this perturbation was compared under conditions with normal and reduced proprioception after 1-Hz rTMS over the hand-contralateral somatosensory cortex. Proprioceptive deafferentation slowed down the reaction time for initiating a motor correction in response to a visual perturbation in hand position, but not to a target jump. Correlation analyses suggested that reaction time was influenced by the size of the visual error rather than the visuo-proprioceptive conflict or the variance in cursor position. We suggest that during movements intact proprioception is necessary for the rapid processing of visual feedback.

Figure 1. Schematic representation of the task

a. unperturbed trial b. trial with a 20 degrees clockwise rotation of cursor trajectory relative to the trajectory of the mouse c. trial with a counterclockwise target jump of 20 degrees along an arc centered on the start position. ! - start position for screen cursor and mouse; ∋ - screen target. Target position and cursor/mouse trajectory without the perturbation are shown with dashed lines and after the perturbation with solid line. The arrows indicate the direction of the perturbation.

Figure 2. Sample finger trajectories

Trajectories from all trials with no perturbation (top row), cursor rotation (middle row) and target jump (bottom row) recorded before (A) and after rTMS (B) in one subject. Blue dots - mouse position. Red circles -time point for maximum acceleration in lateral direction after peak velocity, where a correction was assumed to be initiated. Valid trajectories - blue. Invalid trajectories - green.

Figure 3. Increase in reaction time for cursor rotation after rTMS-induced proprioceptive deafferentation

(mean ± standard error); ! - real rTMS, ∋ - sham rTMS.