Guiding Movements without Redundancy Problems

Approaches to the problems of multi-muscle and joint redundancy have typically been based on the assumption that control levels of the nervous system directly deal with variables describing the motor output – electromyographic (EMG) signals, forces and kinematics. An alternative approach to these problems can be developed in the framework of the • model based on the empirical solution of another classical problem in the motor control that of the relationship between posture and movement. This solution implies that control levels guide movement by changing specific neurophysiological parameters and modify their pattern if the resulting action is in error. Specifically, these control parameters interfere with the transmission of afferent signals by spinal and supraspinal neurons to motoneurons. Some parameters reset the spatial coordinates at which a stable posture of the body or its segments can be reached. Other parameters deal with stability of posture and movement. This parametric control strategy releases higher control levels from the burden of solving redundancy problems at the level of output, i.e. mechanical and EMG variables. In response to changes in control parameters, appropriate values of mechanical and EMG variables and their transformations (e.g., from the hand kinematics to joint angles) emerge automatically, following the natural tendency of the neuromuscular system to reach an equilibrium state. This process results from the natural tendency to minimize the overall activity and the interactions between different components (neurons, muscles and joints) of the neuromuscular system in response to resetting of control parameters (the principle of minimal interaction). Based on these ideas, we outline noncomputational, dynamical solutions of the problems of multi-muscle and multi-joint redundancy. This approach does not reject the notion of synergies, primitives, or recently proposed classification of multijoint co-ordinations into a controlled and uncontrolled manifolds. Rather, it suggests that synergies or manifolds, like trajectories and forces, may be an emergent property of the neuromuscular behavior resulting from the response of the system to changes in control parameters in specific environmental conditions.