The validity of physique-based inverted pendulum models of postural sway behaviour

Abstract

On the grounds that the inverted pendulum model of sway behavior necessarily implies that sway behavior will be a linear function of physique, an attempt was made to predict sway behavior from appropriate individual and composite physique variables. Antero-posterior (Fy) and lateral (Fx) horizontal force components of sway behavior were recorded for 58 young adults, 29 male and 29 female, using a multicomponent biomechanical measuring platform. From these records, digitized using an Hewlett Packard Fourier Analyzer, a rectified sway impulse was calculated. Height, weight, lean body mass, ponderal index and body surface area were also determined for each subject. Correlations between sway magnitude and these physique variables were calculated and appropriate multiple regression equations computed. All correlations were low, with those for the men being consistently lower than those for the women, and all regression equations were of little predictive value. Sway behavior, as estimated by the rectified minute impulse, is essentially independent of physique. Inverted pendulum models of postural sway are unacceptable. The body may behave as a dynamic counterpoise system in which continuous compensatory movements occur simultaneously and internally between adjacent body segments in such a way that ground reaction forces are not predictably invoked. Being thus balanced, the absolute magnitudes of the segmental masses and the positions of these masses above their supporting base are not important.