Postural control: learning to balance and responses to mechanical and sensory perturbations

The purpose of the current research was to examine how a novel balance task is learnt by individuals with a mature neurological system, and to investigate the responses of experienced hand balancers to mechanical and sensory perturbations. Balance in each posture was assessed by various techniques, including: traditional measures of centre of pressure, nonlinear time series analysis of centre of pressure, estimates of feedback time delay from cross correlations and delayed regression models, and calculation of small, medium, and large movement corrections. Data from this study suggests that the best balance metric for distinguishing between each of the balance conditions was the traditional balance measure of sway velocity. However, sway velocity cannot provide any further information on the underlying process of balance. Nonlinear measures of balance offer insight into the underlying deterministic processes that control balance, offering measures of system determinism, complexity, and predictability. Assessments of feedback time delay and movement corrections provide both an insight into the control of posture and help distinguish one condition from another. Both feedback time delay and movement corrections and magnitudes may be used simultaneously to delve further into the control of posture. Delayed regression models seem to be an appropriate and useful tool for estimating feedback time delays during balance. Findings support the use of the third term in the adapted regression model as a means of estimating the effect of passive stiffness on feedback time delay. Generally, with increased duration in handstand subjects displayed reduced sway as measured by traditional measures of balance. A more marked change in nonlinear measures of balance can be seen, with quicker reductions in variance for some nonlinear measures of balance than in the traditional measures. It may be that more pronounced changes in nonlinear measures represent changes in the subjects underlying process of postural control, whereas less pronounced changes in traditional measures relate more to their general ability or performance in the balance task.