Divisively normalized neuronal processing of uncertain visual feedback for visuomotor learning

Abstract

When encountering a visual error during a reaching movement, the motor system improves the motor command for the subsequent trial. The degree of improvement is well-known to be degraded with visual error uncertainty, which is considered evidence that the motor system optimally estimates the error. However, it remains unclear how such statistical computation is accomplished. Here, we propose an alternative motor learning scheme implemented with a divisive normalization (DN) mechanism. This scheme assumes that when an uncertain visual error is provided by multiple cursors, the motor system processes the error conveyed by each cursor and integrates the information using DN. The DN scheme can reproduce the patterns of learning response when 1, 2, or 3 cursors are concurrently displayed and predict how the increase in the visual error uncertainty degrades the learning response. Our proposed scheme provides a novel view of how the motor learning system updates the motor command according to uncertain visual error information.