Initial arm muscle activation in a planar ballistic arm movement with varying external force directions: Aa simulation study

It has been shown in previous research that the initial phase of EMG for a punching movement remained almost unchanged regardless of whether an external force was applied to the arm. The purpose of the present study was to explain this finding with the help of simulations. A two-dimensional model of the arm actuated by 6 Hill-type muscles was used to simulate a punching movement in the horizontal plane from a prescribed starting position with 90degrees elbow flexion. Input to the model was the stimulation of the muscles, and output were, among others, muscle forces and segmental accelerations. A genetic algorithm was used to determine the muscle onset times that minimized movement duration and targeting error. In a subsequent forward simulation, the optimized muscle onset times for an unloaded punching movement were superimposed on the isometric stimulation necessary to hold the arm in the starting position while an external force was applied to the arm. The resulting movement was only slightly different from the unloaded movement. It appeared that because of the low level of isometric muscle force prior to the movement, and the high level of stimulation during the movement, muscle force was increased at a rate that was almost independent of the prior force level. These results confirmed the suggestion that the initial phase of EMG in ballistic movements is more related to the rate of change of force than to the absolute force level. It is hypothesized that this may simplify the task of the nervous system in the choice of initial muscle activity in ballistic arm movements because no adjustments to varying external forces are required.