Coherence of EMG activity and single motor unit discharge patterns in human rhythmical force production

Abstract

The purpose of this study was to examine the modulation of the motor neuronal pool as a function of task dynamics. Specifically, we investigated the effects of task frequency on the single motor unit discharge pattern, electromyogram (EMG) activity and effector force output. Myoelectric activity and effector force were recorded while young adults isometrically abducted their first dorsal interosseus at five sinusoidal targets (0.5 Hz, 1 Hz, 2 Hz, 3 Hz and 4 Hz) and at two force levels (5% and 25% maximum voluntary contraction (MVC)). Individual motor unit spike trains were isolated from the EMG. Auto-spectral and coherence analyses were performed on the force output, EMG and motor unit spike trains. The frequency of maximal coherence between the EMG and force output closely corresponded to the target frequency in all conditions. There was a broadband distribution of power with multiple peaks in the EMG and motor unit spectrums in the 0.5 Hz and 1 Hz targets. However, the EMG and motor unit spectrums in the 2 Hz, 3 Hz and 4 Hz targets were characterized by an increasingly narrower band of activity with one dominant peak that closely corresponded to the target. There is high coherence between EMG output and target force frequency, but the relative contribution of the fast and slow neuromuscular bands are differentially influenced by the task frequency. The rhythmical organization of neuromuscular output in the 0.5 Hz task is relatively broadband and similar to that shown previously for constant level force output. The frequency structure of neuromuscular organization becomes increasingly more narrowband as the frequency of the target increases (2-4 Hz). The modulation of the motor neuronal pool is adaptive and depends on the relative contribution of feedback and feedforward control processes, which are driven by the task demands.