On sensorimotor function and the relationship between proprioception and motor learning

Research continues to explore the mechanisms that mediate successful motor control. Behaviourally-relevant modulation of muscle commands is dependent on sensory signals. Proprioception -- the sense of body position -- is one signal likely to be crucial for motor learning. The present thesis explores the relationship between human proprioception and motor learning. First we investigated changes to sensory function during the adaptation of arm movements to novel forces. Subjects adapted movements in the presence of directional loads over the course of learning. Psychophysical estimates of perceived hand position showed that motor learning resulted in sensed hand position becoming \emph{biased} in the direction of the experienced load. This biasing of perception occurred for four different perturbation directions and remained even after washout movements. Therefore, motor learning can result in systematic changes to proprioceptive function. In a second experiment we investigated proprioceptive changes after subjects learned highly accurate movements to targets. Subjects demonstrated improved acuity of the hand\u27s position following this type of motor learning. Interestingly, improved acuity did not generalize to the entire workspace but was instead restricted to local positions within the region of the workspace where motor learning occurred. These results provide evidence that altered sensory function from motor learning may also include sensory acuity improvements. Subsequently the duration of acuity improvements was assessed. Improved acuity of hand position was observed immediately after motor learning and 24h later, but was not reliably different from baseline at 1h or 4h. Persistent sensory change may thus be similar to retention of motor learning and may involve a sleep-dependent component. In the fourth study we investigated the ability of proprioceptive training to improve motor learning. Subjects had to match the position and speed of desired trajectories. At regular intervals during motor motor learning, subjects were presented with the desired trajectory either only visually, or with both vision and and passive proprioceptive movement through the desired trajectory using a robot. Subjects who received proprioceptive guidance indeed performed better in matching both velocity and position of desired movements, suggesting a role for passive proprioceptive training in improving motor learning