Modeling 3D object manipulation: synchronous single-axis joint rotations?

Abstract

In the present paper we introduce a movement planning model that is capable of predicting object manipulation movements in three dimensions. A basic assumption of this model is that the joint kinematics of the movement are optimized, which implies that joint rotations are synchronous. Synchronous joint rotations can be considered as a simplifying strategy to control arm movements, thus controlling the timing of several segments as a whole rather than for each joint separately. We will discuss evidence for synchronous joint rotations in 2D and explain why 3D synchrony is much more complex to substantiate. Different joint-angle representations and measures of asynchrony yield conflicting results. After showing that our model predicts realistic hand paths for various movement directions (the center-out task), we focus on a task that involves re-orientation of a hand-held cylinder, thus especially zooming in on those degrees of freedom not taken into account in 2D models. The more the cylinder needs to be rotated, the more curved the hand path is. With respect to 3D synchrony, a representation of shoulder and elbow rotations as single-axis rotations comes closest to synchronous joint rotations, which suggests that the brain plans a movement in joint space as a single postural transition.