VELOCITY-DEPENDENT TUNING OF MOTOR STRATEGY DURING 3D ARM MOVEMENT AND ITS RELATIONSHIP TO COMPOSITE COST FUNCTIONS

Abstract

The purpose of this study was to investigate the link between composite cost functions, within the framework of optimal control, and the velocity-dependent tuning of motor strategy observed during the control of unconstrained 3D arm movements. We considered an arm pointing task at three different speeds. Experimental results indicated a change of rotation axis for most subjects from the geometrical shoulder-elbow (SE) axis toward the minimum principal inertia (e3) axis as velocity increased. These findings were interpreted based on a numerical inverse optimal control approach, assuming a total cost composed of kinematic, energetic and dynamic elements. While the kinematic cost predominated at low speed, the contribution of energetic/dynamic costs was reinforced for speeded movements, likely to exploit the inertial properties of the arm.