Multiple paired forward-inverse models for human motor learning and control

Humans demonstrate a remarkable ability to generate accurate and appropriate motor behavior under many dierent and often uncer-tain environmental conditions. This paper describes a new modular approach to human motor learning and control, based on multiple pairs of inverse (controller) and forward (predictor) models. This architecture simultaneously learns the multiple inverse models nec-essary for control as well as how to select the inverse models appro-priate for a given environment. Simulations of object manipulation demonstrates the ability to learn multiple objects, appropriate gen-eralization to novel objects and the inappropriate activation of mo-tor programs based on visual cues, followed by on-line correction, seen in the \size-weight illusion".