Interacting with our environment: impedance control balances stability and metabolic cost

Abstract

Humans are able to stabilize motion in unstable dynamics by modifying arm impedance independently of force. This has led us to propose the existence of an impedance controller in the brain for learning and adaptation of external dynamics. We further investigated this adaptation to unstable dynamics to determine whether the CNS maintains a specific level of stability. Subjects performed reaching movements in force fields of varying levels of instability generated by a robotic manipulator. The force fields initially disturbed the subjects' movements, but in all cases successful movements were achieved with learning. After full adaptation, the endpoint stiffness of the limb was measured at the midpoint of the movement. The stiffness had been selectively modified in the direction of the instability while the stiffness in the stable direction remained relatively unchanged from that in a free environment. After adaptation, the net stiffness of the arm and environment in the direction of instability remained constant across all levels of instability at a value equivalent to that of the null force field. This suggests that the CNS attempts to expend the least energy in maintaining a specific margin of stability.