The framework of the equilibrium point hypothesis has been used to analyse motor control processes for single-joint movements. Virtual trajectories and joint stiffness were reconstructed for different movement speeds and distances when subjects were instructed either to move "as fast as possible" or to intentionally vary movement speed. These instructions are assumed to be associated with similar or different rates of change of hypothetical central control variables (corresponding to the speed-sensitive and speed-insensitive strategies). The subjects were trained to perform relatively slow, moderately fast and very fast (nominal movement times 800, 400 and 250 ms) single-joint elbow flexion movements against a constant extending torque bias. They were instructed to reproduce the motor command for a series of movements white ignoring possible changes in the external torque which could slowly and unpredictably increase, decrease, or remain constant. The total muscle torque was calculated as a sum of external and inertial components. Fast movements over different distances were made with the speed-insensitive strategy. They were characterized by an increase in joint stiffness near the midpoint of the movements which was relatively independent of movement amplitude. Their virtual trajectories had a non-monotonic N-shape. All three arms of the N-shape scaled with movement amplitude. Movements over one distance at different speeds were made with a speed-sensitive strategy. They demonstrated different patterns of virtual trajectories and joint stiffness that depended on movement speed. The N-shape became less apparent for moderately fast movements and virtually disappeared for the slow movements. Slow movements showed no visible increase in joint stiffness. The joint stiffness patterns have been shown to depend upon central control processes rather than on actual values of joint velocity. We conclude that the two basic central strategies have characteristic patterns by which the equilibrium trajectory and joint stiffness are modulated.
All Science Journal Classification (ASJC) codes