Redundancy Problems

Abstract

Humans and animals can repeatedly reach the same motor goal by combining different joint rotations (motor equivalency; Lashley 1951; Bernstein 1935), an important feature allowing behavioral flexibility in everyday life. Bernstein (1967) emphasized the necessity of solving the redundancy problem by answering the question of how the nervous system chooses a specific action from many possible actions each time the goal is reached. Bernstein formulated the problem in biomechanical terms, as the necessity to explain how the nervous system decides which mechanical degrees of freedom (DFs) should participate in a motor action. He assumed that, rather than controlled individually, DFs are combined into well-coordinated groups each of which is controlled as a coherent unit. He called such units synergies that, with some modifications, are now also called coordinative structures, primitives, modules, manifolds or weighed stored postures (see Sect. 5.8). Physiologically, this approach to the redundancy problem is justified—data indicate that DFs are usually controlled jointly—but it did not solve the redundancy problem itself: Bernstein simply re-defined the problem as the necessity to explain how the nervous system chooses a unique combination of synergies each time an action is performed. Actually, in the biomechanical framework, we encounter redundancy problems at every level of motor control, even in single-joint actions (see below).