Modeling Musculoskeletal Movement Systems: Joint and Body Segmental Dynamics, Musculoskeletal Actuation, and Neuromuscular Control

8.1.1 The Modeling Challenge: Keeping the Goal in Mind It is doubtful that anyone would argue that biological motor control systems are less complex than robots. Given that modeling and designing robotic control systems that can walk or manipulate objects is quite challenging to engineers [e.g. (Lee, 1989)], is there any hope for those of us who wish to develop "adequate" models of biological motor control systems? The answer depends on the definition of "adequate". Whether a model is adequate or not depends on whether it helps, hopefully significantly, in fulfilling the scientific or engineering goal. For study of movement the underlying goal is usually to maximize one's insight into the system and its behavior for a given movement task or class of tasks. For instance, suppose the goal is to understand how muscles work together [i.e., act in synergy, see Zajac and Gordon (1989) for discussion of synergtsttc and agonist/antagonist muscle-group definitions] to control elbow flexion and extension where, say, the shoulder and trunk are to be kept motionless. One paradigm is to design a shoulder and trunk harness to keep the shoulder and trunk stationary, in which case a model with just one body-segment and elbow flexor and extensor muscles would probably be adequate. Another paradigm is to allow the shoulder and trunk to be free to move. The subject must then coordinate the elbow muscles with the shoulder and trunk muscles to perform two sub-goals to accomplish the overall task. The subject must a) move the elbow as before, and b) maintain the shoulder and trunk stationary. Thus a model is needed to understand how the shoulder and trunk muscles act with the elbow muscles to perform these two sub-goals and a complex multi-jointed segmental model controlled by shoulder, elbow, and trunk muscles must be formulated. This model cannot be subdivided into two models, one for elbow control and one for shoulder and trunk control, because of the dynamical interactions occurring in multijoint motor tasks. For example, muscles crossing one joint act to rotate the other joints and these multijoint effects must be considered (Gordon and Zajac, 1989; see below).