When subjects are presented with distorted visual feedback of their hand during a goal-directed movement (i.e. subjects view a cursor representing their hand that is rotated from their hand’s actual position while reaching in a virtual reality environment), they typically adapt their movements so that the cursor is brought to the target, thus reducing reaching errors. In addition to motor adaptation, it has recently been shown that reaching with distorted visual feedback of the hand results in sensory changes, such that proprioceptive estimates of hand position are shifted in the direction of the visual feedback (Cressman and Henriques 2009). The current study looked to establish how quickly these sensory changes arise while training to reach with distorted visual feedback of the hand. Additionally, by comparing sensory to motor changes across time, we looked to determine the relationship between their underlying processes. Subjects trained to reach to a single visual target while seeing a cursor that was aligned with their actual hand position (50 trials: aligned reach training), or rotated 30° clockwise (CW) relative to their actual hand position (150 trials: rotated reach training). Reach errors and proprioceptive estimates of felt hand position were assessed following the aligned reach training trials and at 7 different times during the rotated reach training trials by having subjects reach to the target without visual feedback, and provide estimates of the position of their hand relative to a visual reference marker respectively. Results revealed a slow change in proprioceptive estimates over the course of reach training with the rotated cursor relative to estimates after the aligned reach training, and in fact, significant sensory changes were not observed until after 70 trials. In contrast, reach adaptation showed a much steeper increase and significant adaptation after a limited number of reach training trials with a rotated cursor. These different time courses suggest that proprioceptive recalibration and reach adaptation arise due to separate neural processes.

Proprioception is recalibrated following reaches with misaligned visual feedback of the hand, such that one’s sense of felt hand position is shifted in the direction of the visual feedback provided (Cressman & Henriques 2009). In the current experiment, we examined the ability of proprioceptive recalibration to be retained over an extended period of time (i.e. 4 days), and the benefits of additional training on retention in the form of recall and savings (i.e. faster re-learning on subsequent testing days). Twenty-four participants trained to reach to a target while seeing a cursor that was rotated 30° clockwise relative to their hand on an initial day of testing. Half of the participants then completed additional reach training trials on 4 subsequent testing days (Training group), whereas the second half of participants did not complete additional training (Non-Training group). Participants provided estimates of their felt hand position on all 5 testing days to establish retention of proprioceptive recalibration. Results revealed that proprioceptive recalibration was recalled 24 hours after initial training and that there was no benefit of additional training. Retention in the form of savings was observed on all days for the Training group and on Day 5 in the Non-Training group. These results reveal that proprioceptive recalibration does not benefit from additional training but is retained in the form of recall and savings. Taken together, results from the two groups of participants showed that the sensory system’s ability to change over time appeared to saturate early on, within two days of training. Moreover, the different time scales (i.e. 1 day for recall versus 4 days for savings), suggested that distinct processes may underlie recall and savings of proprioceptive recalibration.