Movement variability response to change in the rate of hopping

Abstract

The purpose of this study was to compare two methods of measuring movement variability that provide information at the task level, and their capacities to illuminate the neuromotor control system's response to change in hopping rate. The typical task-level movement variability measure of the standard deviation of vertical limb length was compared to uncontrolled manifold analysis. We examined the relationship between change scores in deviation from spring-mass model-type behavior and these two variability measures for the shift from typical (2.3 Hz) to slow (1.7 Hz) hopping. The change scores for deviation from spring-mass model-type behavior and vertical limb length standard deviation demonstrated no correlation ( p = 0.784, R = 0.051). In contrast, the change scores for deviation from spring-mass model-type behavior and the uncontrolled manifold analysis measure demonstrated a moderate correlation ( p = 0.004, R = 0.502). Uncontrolled manifold analysis considers not just variability in the sense of error, but illustrates how the neuromotor control system distributes movement variability into performance-irrelevant and performance-destabilizing subspaces. As such, this type of analysis may be more effective at illuminating global control aspects of movement variability than the typical variability measure of limb length standard deviation.