Separating Predicted and Perceived Sensory Consequences of Motor Learning

Abstract

During motor adaptation the discrepancy between predicted and actually perceived sensory feedback is thought to be minimized, but it can be difficult to measure predictions of the sensory consequences of actions. Studies attempting to do so have found that self-directed, unseen hand position is mislocalized in the direction of altered visual feedback. However, our lab has shown that motor adaptation also leads to changes in perceptual estimates of hand position, even when the target hand is passively displaced. We attribute these changes to a recalibration of hand proprioception, since in the absence of a volitional movement, efferent or predictive signals are likely not involved. The goal here is to quantify the extent to which changes in hand localization reflect a change in the predicted sensory (visual) consequences or a change in the perceived (proprioceptive) consequences. We did this by comparing changes in localization produced when the hand movement was self-generated ('active localization') versus robot-generated ('passive localization') to the same locations following visuomotor adaptation to a rotated cursor. In this passive version, there should be no predicted consequences of these robot-generated hand movements. We found that although changes in localization were somewhat larger in active localization, the passive localization task also elicited substantial changes. Our results suggest that the change in hand localization following visuomotor adaptation may not be based entirely on updating predicted sensory consequences, but may largely reflect changes in our proprioceptive state estimate.

Fig 1. Setup and experimental design.

A: Participants moved their unseen right hand with visual feedback on hand position provided through a mirror (middle surface) half-way between their hand and the monitor (top surface). A touchscreen located just above the hand was used to collect responses for the localization tasks and calibration (bottom surface). B: Training task. The target, shown as a yellow disc, is located 10 cm away from the home position at 45°. In the rotated training tasks, the cursor (shown here in as a green circle) represents the hand position rotated 30° relative to the home position. C: Display of the no-cursor reach task. Targets are located 10 cm away from the home position at 15°, 25°, 35°, 45°, 55°, 65°, and 75°, shown by the yellow circles here (only one was shown on each trial). While reaching to one of these targets, no visual feedback on hand position is provided. D: Localization task. The participants’ unseen, right hands moved out, and subsequently participants indicated the direction of the hand movement by pointing with their visible left hand at a location on a an arc on a touch screen.

Fig 2. Time course, and reaching results, of the experiment.

A, top: task/block order. Each session started with a longer training block. The four localization tasks were performed after a block of training and before a no-cursor reach block. The order was similar for the aligned and rotated sessions, but the rotated sessions had an extra no-cursor reach block and training block after the initial training block. A, bottom: Hand movement direction while reaching during the rotated session. The average angle at the point of peak velocity over trials (purple area in the training blocks denotes the average across participants ± SEM) shows that participants adapted their reaching movements to rotated visual feedback in step with the gradually introduced rotation of visual feedback. Reach aftereffects are shown in the no-cursor reach blocks. The curves go from the target at 15° (at the left side of the curve) to the target at 75° in order, and the circles denote the aftereffects at the 45° target, the only target used during training. B: Generalization of reach aftereffects. The difference between the average reach endpoint angles in each of the no-cursor blocks done after rotated training, corrected for the average reach endpoint angles across all the no-cursor blocks done after aligned training. Gray areas represent the standard error of the mean. Note that the effect is strongest at the trained target, and decreased for targets further away. Also, the reach aftereffects measured immediately after training are slightly larger than those in the other blocks. C: Time course of reach aftereffects. Left: On the very first trial following training, reach aftereffects are about twice as large as at the end of that block. Right: Except for some initial errors, the reach aftereffects measured on the central three targets seems stable in the no-cursor reach blocks following localization. Gray areas indicate standard error of the mean, dashed lines indicate the 95% confidence interval for responses in the blocks after localization.

Fig 3. Localization results.

The change in the touchscreen responses in the four variations of the localization task, using 10° bins centered on the reach targets (circles) in the no-cursor reach block. Active localization is shown in purple, passive localization in gray. The eye-icons illustrate the direction of the target during training, and the hand icons illustrate the direction of movements required to hit the target with 30° rotated visual feedback. A,C: Delayed localization, B,D: Online localization. A,B: The beginning and end of the arrows show the average deviation from the true reach angle in that bin before and after visuomotor adaptation, respectively. The open arrow illustrates the visuomotor rotation. E: The average change in the direction of hand localization across bins and participants.