A learning feedback and feedforward neuromuscular control model for two degrees of freedom human arm movements

• Published in: Human movement science 1997-10, Vol.16 (5), p.621-651
• Main Author: Stroeve, Sybert
• Format: Article
• Language: English
• Subjects: Biological and medical sciences; Fundamental and applied biological sciences. Psychology; Human arm movements; Learning feedback; Learning feedforward; Motor control and motor pathways. Reflexes. Control centers of vegetative functions. Vestibular system and equilibration; Neuromuscular control model; Vertebrates: nervous system and sense organs
• ISSN: 0167-9457; 1872-7646
• DOI: 10.1016/S0167-9457(97)00009-2

Development of the understanding of human movement control requires models of neuromuscular control systems. Such models should be nonlinear, include feedforward and feedback control modes, account for delays in the control loop and be able to control musculoskeletal models with redundant actuators and multiple degrees of freedom (dofs). A model of the neuromuscular control system for human arm movements in the horizontal plane which satisfies those demands is discussed in this paper. The control process of the central nervous system (CNS) is represented by a neural network. This control system is trained by execution and evaluation of many movements. The arm, which is moved by four mono and two biarticular muscles, is represented by a 2-dofs system in which a wide range of movement directions and velocities can be controlled. In accordance with experimental data, typically triphasic activation patterns are found for the control of fast goal-directed movements and monophasic activation patterns for slow movements. In accordance with experimental results from the literature, it was found that the average torques and muscle forces during the initial phase of goal-directed movements as a function of the movement direction can basically be described by circular curves for both mono and biarticular muscles.