[A new system for the quantitative evaluation of motor commands for neurorehabilitation].

In this study, we introduce a new method to analyze the functional significance of motor commands based on the causal relationship between muscle activities and movement kinematics. We asked 4 healthy subjects to perform 2 movement tasks involving movements of the wrist joint with a manipulandum: (1) a step-tracking movement, and (2) a smooth pursuit movement to follow a smoothly moving target with a cursor. We recorded the movements of the wrist joint and the electromyography (EMG) signals from 4 prime movers of the wrist. We then identified the causal relationship between the muscle activities and the movement kinematics in terms of the torque at the wrist joint torque. Correlation coefficients (R) between the muscle activities and the movement kinematics were surprisingly high for both the step-tracking movement (0.88-0.96) and the smooth pursuit movement (0.89-0.98). Nevertheless, the ratio of the viscosity coefficient B and the elastic coefficient K for the of the causal relationship was systematically different for the 2 movement tasks. For the step-tracking movement, the elastic coefficient K was much higher than the viscosity coefficient B (K>>B, B/K ratio = 0-0.25), suggesting that the muscle activities were correlated mainly with position of the wrist joint. In other words, the central nervous system (CNS) mainly controls the target position during the step-tracking movement. In contrast, for the smooth pursuit movement, the elastic coefficient K and the viscosity coefficient B were similar (B approximately equal to K, B/K ratio = 1.0-2.1), suggesting that the muscle activities were correlated both with the position and velocity of the wrist joint. In other words, the CNS controls the velocity as well as position of the target during the smooth pursuit movement. Overall, the controller for the wrist movement can switch between the position-control mode and the position/velocity-control mode depending on the requirement of the task. Our new method is a unique noninvasive tool for the bedside evaluation of the condition of motor controllers in the CNS. Thus, it is useful for finding evidence of neurorehabilitation.