Few attempts have been made to calculate the Joint torques produced by Individual muscles during dynamic contractions because no suitable technique of completely estimating the muscle forces has been documented* The purpose of this study was to develop and evaluate an EMG technique to account for the individual contributions of motor unit activation of three elbow Joint muscles to the resultant dynamic joint torque produced by the combined activity of these muscles* Horizontal flexion and extension movements of the forearm were studied*

A model was developed which considered the activity of the biceps, brachioradialis and triceps muscles in producing joint torque. The output of the model was the resultant Joint torque produced by the three muscles. The input was the theoretical maximum isometric force that each muscle could develop based on literature values of its cross-sectional area and force per unit area;​ the moment arm length of each muscle as a function of the changing joint angle calculated using geometrical techniques;​ the instantaneous proportion of maximum motor unit activity calculated digitally from EMG patterns evoked from each muscle during the motion.

Three experiments, using five subjects, were conducted, The first two experiments were devoted to establishing techniques of using the EMG as input to the model. The results of these studies*, based on isometric contractions, showed that the minimum time adequate for yielding a meaningful rectified integrated EMG was about 40 msec. In addition) isometric contractions at an elbow angle of 90 degrees and an intensity of 50 percent of maximum voluntary torque produced EMGs which were stable enough to be used as a normalization factor.

Assessment of the technique of motor unit activity representation in the calculation of individual muscle torque was made by comparing plots of the resultant joint torque, generated by the model, with the measured resultant dynamic joint torque. These results showed that the digital normalized EMG method of accounting for the contribution of motor unit activation of individual muscles to their respective joint torques during dynamic contractions was viable. The model predicted the torque output of the concentric phase of slow and moderate speed movements quite well. The eccentric phases of the contractions were not accounted for very well The reasons for discrepancies between the model and measured joint torques with implications for further refinement of the model were discussed. A sample calculation of the appreciable bone-on-bone forces created during the movements was demonstrated.