Transient beta activity and cortico-muscular connectivity during sustained motor behaviour

Figures

• 

  Fig. 1. Trial-average beta activity and connectivity appear sustained during sustained motor behaviour (gripping). a ) Schematic of a single trial. Before each trial, participants held the gripping devices in both hands (resting grip). At time 0, two horizontal lines, indicating the gripping strength, appeared on the bars on the screen, prompting participants to grip. Participants began gripping until they reached a steady grip at the indicated strength, which they sustained for ~2 s. The drop of the horizontal lines to the bottom of the bars indicated the end of a trial and the return to resting grip. b) Average grip output across trials and across left and right gripping devices in a representative participant, expressed as a percentage of the peak force in each trial. Shading indicates 95% confidence intervals. c) Time-frequency spectrum of trial-average activity in selected motor MEG channels in a representative participant. Selected MEG channels correspond to those with maximal cortico-muscular coherence during sustained gripping (see methods) as also indicated in the topographical distribution in Fig. 1e (black dots). d) Time-frequency spectrum showing trial-average EMG activity…

• 

  Fig. 2. Beta activity is transient in single trials, despite sustained motor output. a) Gripper traces together with time-frequency spectra in brain and muscle in four example trials from the same participant whose trial-average data is shown in Fig. 1. b) Zoomed in view of the data in a, aligned to the occurrence of a beta event in the single trials above. Percentage of maximal grip output was defined relative to the whole epoch.

• 

  Fig. 3. Data aligned to beta ON and OFF periods in the brain reveal transient beta in brain, muscle, and connectivity despite sustained motor output. Beta ON events and OFF periods were identified using Empirical Mode Decomposition (EMD) in the selected MEG channels during the ~1–3 s period of interest. After ON and OFF period detection, data were aligned and averaged across all trials and all participants (N = 18). Columns from left to right represent our four signals of interest (a-d) aligned to the central point of a beta ON event (left), a beta OFF period (middle) and the difference between them (right). a) MEG time-frequency spectrum, b) EMG time-frequency spectrum, c) CMC time-frequency spectrum and d) gripper signal. Shading indicates 95% confidence intervals. Black outlines in the right time-frequency maps indicate significant clusters.

• 

  Fig. 4. Comparison of the mean duration of beta events during sustained gripping and rest. a) MEG time-frequency spectra aligned to all beta ON periods during sustained gripping (task) and during resting state. Topographies show beta power (13–30 Hz) at the time of the detected ON events during gripping and rest. b) Violin plot showing the distribution of mean beta event durations during gripping (top) and resting state (bottom). Grey lines represent individual participants. A paired-sample t-test revealed no significant differences in mean burst duration between periods of sustained motor contraction and periods of rest (t17 = 1.52, p = 0.15, d = 0.36).