Proceedings of The Physiological Society

University of Oxford (2011) Proc Physiol Soc 23, C68

Oral Communications

Oscillatory corticomuscular coupling during isometric voluntary contraction influences reaction time in humans

J. Ushiyama1,2, R. Matsuya3, A. Kimura4, M. Liu1, S. N. Baker5, J. Ushiba1,3

1. Department of Rehabilitation Medicine, Keio University, Tokyo, Japan. 2. Graduate School of Science and Technology, Keio University, Kanagawa, Japan. 3. Department of Biosciences and Informatics, Keio University, Kanagawa, Japan. 4. Keio University Tsukigase Rehabilitation Center, Keio University, Shizuoka, Japan. 5. Institute of Neuroscience, Newcastle University, Newcastle, United Kingdom.


  • Figure 1. Example reaction time difference between GD- and GD+.

It has been well documented that the sensorimotor cortex activity shows coherence with muscle activity within the 15-35 Hz band (beta-band) during weak to moderate isometric contractions (e.g. 1-3). Recently, we reported that subjects with greater coherence between the electroencephalogram (EEG) over the sensorimotor cortex and electromyogram (EMG) of contracting muscle show prominent grouped discharge in the EMG signal within the beta-band (4). Although it is possible that such grouped discharge may affect motor performance, its functional role in the control of human movement still remains unclear. The present study aimed to examine how beta-band oscillations in EMG influences reaction time. We recruited seven healthy human subjects (four males and three females, 21-26 years) who showed significant EEG-EMG coherence during a sustained isometric contraction of the tibialis anterior muscle in a preliminary experiment. Subjects first performed a steady contraction at 30% of maximal effort for 5-7 s, and then reacted to a sound cue by performing a ballistic dorsiflexion as quickly as possible. They repeated this task for 100 trials. Reaction time was measured from the cue to the time when the force signal exceeded the mean + 10 SD of data of measured over 1 s before the cue. We found that within each subject, trials could be divided into two groups: those where EMG showed prominent grouped discharge within the beta-band before reaction (GD+), or those where it did not (GD-). Pooled EEG-EMG coherence before the reaction was calculated in both groups; coherence was greater in GD+ than in GD-. This implies that even within the same subject, the strength of corticomuscular coupling changed from moment to moment, resulting in the difference in the degree of grouped discharge in EMG during the initial contraction. Further, in order to examine the effects of such grouped discharge in EMG on reaction time, we assessed the differences in reaction time between GD+ and GD- within each subject by using an unpaired t-test. The reaction time in GD+ was longer than that in GD- for all seven subjects, and was significantly so in three subjects (Subject 1, GD+, 744 ± 49 ms, GD-, 722 ± 54 ms; Subject 4, GD+, 767 ± 44 ms, GD-, 748 ± 47 ms; Subject 6, GD+, 785 ± 59 ms, GD-, 757 ± 42 ms; P < 0.05). These findings suggest that when grouped discharge develops in EMG during a steady contraction, reaction time is delayed, entrained by the cycle of oscillatory corticomuscular coupling.

Where applicable, experiments conform with Society ethical requirements