The Topology of Limb Deceleration in Prehension Tasks

Whether a subject uses a constant tau-dot strategy to decelerate a reach to grasp an object is typically determined by performing a linear regression of tau onto real time to contact. This method has the disadvantage of requiring an assumption (on the part of the experimenter) concerning the endpoint targeted by the subject. In the present contribution, an alternative, topological method is proposed that makes use of a phase-plane representation of the movement. In two experiments, experimentally obtained movement trajectories were evaluated, using this new method, and the results were found to support the constant tau-dot strategy. At the same time, however, an alternative model, associated with dynamical systems theory, was demonstrated to fit the data even better. It is argued that the topology of the phase plane suggests a softening spring description for a large part of the decelerative phase, followed by a final adjustment. Deterioration of the functionality of the adjustment made in this latter phase was observed when visual information was not available during movement execution.