Trial-by-trial analysis or averaging: implications for electromyographic models of rapid limb control

The control of human limb movement has been the focus of research for more than a century. A major issue to emerge from this work is the manner in which the central nervous system regulates electromyographic (EMG) activity to produce movements that differ in distance, velocity, and movement time. However the different methods of analysis often used to analyze EMG data could result in different kinematic-EMG relationships. In this study, participants performed an elbow flexion task to one of five distance goals (between 5 degrees and 50 degrees) using three movement speeds. EMG data from the right elbow flexors were compared using a trial-by-trial analysis and one based on averaged data. Averaging across trials underestimated biceps EMG amplitude at all movement distances and speeds compared to a trial-by-trial analysis. Averaging overestimated EMG burst duration compared to the trial-by-trial analysis. Peak agonist EMG amplitude was positively related to distance and inversely related to movement time. Agonist EMG duration was constant for movement distances less than 30 degrees but increased in the 50 degrees condition. The results support the view that peak EMG amplitude and duration can be controlled independently, but EMG duration changes only for longer distance movements when additional force is required.