AbstractTo study implicit sensorimotor adaptation, Morehead et al. (2017) used clamped feedback where the feedback cursor follows an invariant trajectory relative to the target, irrespective of reach direction. Despite participants' knowledge of the manipulation, reaches gradually drift in the direction opposite of the clamped feedback, revealing how the sensorimotor system automatically responds to sensory prediction errors (SPE), the mismatch between the expected and observed location of the cursor. An unresolved question is why adaptation eventually levels off despite the persistent presence of a visual SPE. One proposal is that as the arm moves further from the target, a proprioceptive SPE, the mismatch between expected and actual proprioceptive feedback, may counteract the effect of the visual SPE. We tested this idea by introducing proprioceptive noise, hypothesizing that by "weakening" the proprioceptive SPE, the asymptote in response to clamped visual feedback should increase. Participants (n=34) adapted to a 15° clamp while experiencing biceps-triceps vibration during one phase of the experiment and no vibration during a second phase, with the order counterbalanced across participants. Focusing on the first phase, the mean asymptote was larger for participants who experienced vibration (20.1° [15.6 24.5]) compared to those who started with no vibration (15.5° [12.3 18.7]). Surprisingly, there was little change during the second phase when vibration condition was reversed for the two groups (e.g., introduction of vibration did not lead to an increase in asymptote). These preliminary results point to an influence of proprioceptive uncertainty on the upper bound of adaptation (Tsay et al. 2022).
Acknowledgments: Supported by Natural Sciences and Engineering Research Council of Canada