Computing movement geometry: a step in sensory-motor transformations

Abstract

The generation of goal-directed movements requires the solution of many difficult computational problems. In addition to generating the forces needed for movement there are a number of essentially geometric problems. Among these are transformations from extrinsic to intrinsic reference frames, removing under-specification due to excess degrees of freedom and path multiplicity, and error correction. There are no current motor control computational models that address these issues in the context of realistic arm movement with redundant degrees of freedom. In this chapter, we describe a geometric stage between sensory input and physical execution. The geometric stage determines movement paths without reference to forces. It is implemented with a gradient technique that can generate movement paths online. The model is demonstrated by simulating a seven degree of freedom arm that moves in three-dimensional space. Simulated orientation-matching movements generated by the model are compared with human experimental movement data to assess the validity of several of the model's behavioral predictions.