Task Effects on Three-Dimensional Dynamic Postures during Seated Reaching Movements: An Analysis Method and Illustration

Abstract

This paper presents a new method to empirically investigate the effects of task factors on three-dimensional (3D) dynamic postures during seated reaching movements. The method relies on a statistical model in which the effects of hand location and those of various task factors on dynamic postures are distinguished. Two statistical procedures are incorporated: a regression description of the relationship between the time-varying hand location and postural profiles to compress the movement data, and a series of analyses of variance to test the hypothesized task effects using instantaneous postures with prescribed hand locations as dependent measures. The use of this method is illustrated by an experiment which examines two generic task factors: 1) hand movement direction, and 2) motion completion time. The results suggest that the hand motion direction is a significant task factor and should be included as an important attribute when describing or modeling instantaneous postures. It was also found that the time to complete a motion under a self-paced mode was significantly different from a motivated mode, but the time difference did not significantly affect instantaneous postures. The concept of an instantaneous posture and its usage in dynamic studies of movements are discussed. Some understanding of human postural control as well as the implications for developing a general dynamic posture prediction model also are presented.