1. We studied the patterns of EMG activity in elbow muscles in three normal human subjects. The myoelectrical activity of 7-10 muscles that act across the human elbow joint was simultaneously recorded with intramuscular electrodes during isometric joint torques exerted over a range of directions. These directions included flexion, extension, varus (internal humeral rotation), valgus (external humeral rotation), and several intermediate directions. The forces developed at the wrist covered a range of 360°, all orthogonal to the long axis of the forearm. 2. The levels of EMG activity were observed to increase with increasing joint torque in an approximately linear manner. All muscles were active for ranges <360° and most were active for <180°. The EMG activity was observed to vary in a systematic manner with changes in torque direction and, when examined over the full angular range at a variety of torque levels, is simply scaled with increasing torque magnitude. 3. There were no torque directions or torque magnitudes for which a single muscle was observed to be active alone. In all cases, joint torque appeared to be produced by a combination of muscles. 4. The direction for which the EMG of a muscle reached a maximum value was observed to correspond to the direction of greatest mechanical advantage as predicted by a simple mechanical model of the elbow and relevant muscles. 5. Muscles were relatively inactive during varus torques. This implies that the muscles were not acting to stabilize the joint in this direction and could have been allowing ligaments to carry the load. 6. Plots of EMG activity in one muscle against EMG activity in another demonstrate some instances of pure synergies, but patterns of coactivation for most muscles are more complicated and vary with torque direction. The complexity of these patterns raises the possibility that synergies are determined by the task and may have no independent existence. 7. Activity in two heads of triceps brachii (medial head - a single-joint muscle and long head - a two-joint muscle) covaried closely for a range of torque magnitudes and directions, though shoulder torque and hence the forces experienced by the long head of the triceps undoubtedly varied. The similarity of activation patterns indicates that elbow torque was the principal determining factor. 8. The origins of muscle synergies are discussed. It is suggested that they are best understood on the basis of a model which encodes limb torque in premotor neurons. These interneurons would effectively contain a map of joint torque for a certain direction and would be accessed by appropriate descending pathways, when such torque directions were commanded.
ASJC Scopus subject areas