The experiments were designed to examine the effect of task constraints on the influence of kinematic information feedback to facilitate the acquisition of discrete arm movements. The findings of Experiments 1 and 3 revealed that when the criterion kinematic trajectory was an increasing acceleration function, the most effective control space representation for kinematic feedback (is. position-time; velocity-position) was the one that matched the error criterion t o be minimized. Furthermore, in Experiment 1 the velocity-position feedback condition led to greater performance error than the discrete knowledge of results of movement time or integrated position-time error. Experiment 2 showed that kinematic information feedback of the movement trajectory (position-time; velocity-position) did not facilitate acquisition of a constant velocity criterion, in contrast to knowledge of results of movement time or integrated velocity-position error. Collectively the findings suggest that the interaction of task and organismic constraints dictates the nature of the information feedback required to facilitate the acquisition of skill. The augmented information available must match the degrees of freedom requiring constraint in the movement sequence.
In a recent series of experiments we have contrasted the effect of kinematic information feedback with knowledge of results (KR) in the acquisition of discrete single biomechanical degree of freedom tasks (see , for reviews of this work). These experiments have clearly shown that kinematic information feedback can facilitate performance beyond the level achieved through traditional KR information (see also . On the basis of these experimental findings we have developed a framework for the application of augmented kinematic information feedback utilizing topological dynamics to describe both movement and the constraints imposed upon movement. The general principle advanced by , and explored in this paper experimentally, is that augmented kinematic information feedback must be presented to the subject in terms of the control space that is required to define unequivocally task constraints, together with the constraints imposed upon movement by optimization criteria.
Essential to the application of topological dynamics is an understanding of the kinematic constraints imposed by task criteria. Kinematic information relates to the space-time properties of movement, and examples include the displacement velocity and acceleration of body and limb motion. These criteria can be framed explicitly, in terms of the goal of the task, or implicitly, in terms of the various physical constraints inherent in the interaction of the organism and environment. For example, in a recent set of experiments on kinematic information feedback , it was shown that a continuous kinematic representation of the