Modeling the human elbow joint dynamics: estimation of joint stiffness with different loads and movement velocities

Knowledge about joint dynamics during the performance of human joint movements has become very crucial if one is to understand the mode of joint control, how the brain generates the necessary control signals and how movement performance is maintained under varying loading conditions. Experiments were performed on 3 subjects using computer controlled manipulation. The subjects performed large elbow angular cyclical movements with different loads. During the movements random mechanical perturbations were applied to the manipulation. After phase alignment of all the movements to the peak velocity of each cycle, the dynamic parameters of the joint were estimated at each time point from the ensemble of all the movements. 500 movement cycles were used for each of the 2 velocities and 3 loading conditions. Different parameter estimation algorithms were used to calculate the stiffness, viscosity and inertia of the elbow joint. Here, the author demonstrates that despite large variances in the values of the estimated parameters, it is possible to show that joint stiffness varies during a cyclic movement in correlation with movement acceleration. This stiffness is lower during movement then during steady state. The relation of these findings to human movement control theories are discussed.