Information flow from the sensorimotor cortex to muscle in humans

Objectives: To investigate the physiologic mechanism of human electroencephalogram–electromyogram (EEG-EMG) coherence, the directed transfer function (DTF) based on a multivariate autoregressive (MVAR) model was computed. Methods: Fifty-six channel EEG and EMG of the right abductor pollicis brevis muscle during a weak tonic contraction were recorded in 6 normal volunteers. The EEG over the left sensorimotor area and the rectified EMG were used to compute coherence and DTF. Results: EEG-EMG coherence was observed at the peak frequency of 15–29 Hz (mean 18.5 Hz). The peak frequency of DTF from EEG to EMG was 12–27 Hz (mean 17.8 Hz). DTF from EEG to EMG was significantly larger than that from EMG to EEG at 19–30 and 45–50 Hz (P<0.05). Conclusions: The present findings suggest that the EEG-EMG coupling mechanism for the 19 Hz or higher frequency might differ from that for the lower frequency. Directional information flow from EEG to EMG in the former frequency range likely reflects the motor control command. The finding of the directional information flow from EEG to EMG within the gamma band indicates that 40 Hz EEG-EMG coherence is not specific to the muscle Piper rhythm which is seen only with strong contraction.