6 - Control with muscle activations

Abstract

These approaches assume that the neural controller specifies patterns of muscle activation to produce desired movements. To do this, total presynaptic input into alpha-motoneuronal pools is specified. Reflex contributions to muscle activation may be very large. They are assumed to be predicted by the controller. The dual-strategy hypothesis assumes that the controller specifies the height and duration of rectangular excitation pulses into motoneuronal pools controlling the agonist and antagonist muscles. A speed-insensitive strategy is used to produce movements with changes only in excitation pulse duration, not height. A speed-sensitive strategy uses modulation of excitation pulse height. The dual-strategy hypothesis can account for certain salient features of the early components of the tri-phasic electromyographic burst during both quick movements and isometric force production. The pulse–step hypothesis assumes that an input to a motoneuronal pool can be represented as a sequence of a short-lasting pulse and a longer lasting step. It assumes separate plans for the trajectory (spatial or force) of an effector and its final state (position or force). Multi-muscle activation patterns can be described using the notions of muscle synergies or modes that assume parallel scaling of activation across subsets of muscles. Methods such as principal component analysis, factor analysis, and non-negative matrix factorization have been used to identify such muscle subsets.