The effects of perturbation frequency, magnitude, and uncertainty during static and dynamic tracking on the estimated level of muscle co-contraction

Abstract

Two experiments were performed to investigate the predictability of antagonistic muscle forces during static and dynamic tracking tasks with perturbations. In the first experiment, perturbations to the isometric tracking were effected by sudden momentary changes in the gain. In the second experiment, perturbations to the tracking consisted of momentary changes in the resistance of the lever arm used to track the icons.

In the first experiment with isometric tracking the effects of perturbation frequency, perturbation magnitude, tracking gain, tracking phase and the absence or presence of preview information on the level of antagonistic co-contraction about the elbow were investigated. The effects of these variables on the tracking error were also examined. In the second experiment, the effects of perturbation frequency, perturbation magnitude, tracking phase, and the absence or presence of preview information were investigated using a dynamic tracking task. Measures of antagonistic co-contraction and tracking error were again computed and used as dependent measures.

Optimization-based biomechanical models have been used by a number of authors to calculate calculate individual muscle forces. The models require that certain lower bounds of the muscle forces be specified. The optimization criteria typically used result in a solution in which the antagonistic muscle forces are at their lower bounds. This lower bound is typically assumed to zero, but could very well be specified at some other nonnegative value.

Two experiments were designed to test whether the level of antagonistic co-contraction varies in a regular predictable way with the factors under study. The aim of the study was to establish a theoretically-derived and experimentally confirmed basis for selecting lower bounds greater than zero for antagonistic muscles.

The two experiments showed that perturbations were effective in impairing tracking performance and subjects evidenced a limited ability to respond to the perturbations by increasing the antagonistic muscle force so that it would minimize the effects perturbations. The study also examined the effects of movement direction, and level of force required on the level of observed antagonistic co-contraction and confirmed the findings of previous authors.

The results of the study do not support different lower bounds for tasks which vary in perturbation frequency, or preview information. The results do indicate significantly different levels of antagonistic activity with tracking gain and direction of exertion in the isometric task. The results also indicate significantly different levels of antagonistic activity with perturbation magnitude in the dynamic tracking task.