Generalization patterns for sensory and reach adaptation following exposure to visual-proprioceptive discrepancies

Abstract

The CNS evolved sensory and reach adaptation as types of plasticity to deal with body-growth changes and variability in the surrounding world. Reach adaptation generalize to untrained contexts and transfer between limbs. We examined the extent by which proprioceptive recalibration generalize to the untrained hand, across novel locations in the workspace. In experiment 1, subjects trained to reach with an aligned and translated cursor, we assessed the resulting changes in hand movements (without cursor) and felt hand position for both trained and untrained hand. Reach adaptation transferred between hands, proprioceptive recalibration did not transfer. In experiment 2, we measured reach adaptation and proprioceptive recalibration at novel locations following training with a rotated cursor. Reach and sensory adaptation generalized to novel locations at different distances, however, sensory changes generalized with smaller extent at far-locations. In experiment 3, we removed the motor component during training so that subjects exposed to a proprioceptive-visual discrepancy in which they see the cursor heading to the training target while the robot gradually rotates their unseen hand-path. Subjects reached to one target from a starting-position(S1) then we measured reach and sensory changes at novel locations from S1 and from a novel starting-position (S2). We found proprioceptive recalibration at the trained and novel locations from S1 and S2. Additionally, we found reach adaptation at the same locations but with smaller extent. Our findings suggest that reach and sensory adaptation may be independent, mere exposure to proprioceptive-visual discrepancy results in proprioceptive recalibration which drive partial reach adaptation that follow similar generalization pattern.