The Dependency of the Slowing of Velocity during Stimulation at 20 Hz upon Baseline Performance
(A) Initial velocity (average over 40–100 ms after response onset).
(B) Peak velocity. Gray lines mark gradients of 1. Note that the majority of points fall below the gray lines, indicating that stimulation slows velocity. In addition, linear regression lines (black) have gradients of < 1, suggesting that the slowing of velocity with stimulation at 20 Hz is greater in subjects making more rapid movements. See Results and Discussion for statistics.

The Velocity Profiles of Trials Performed with and without 20 Hz Stimulation
In (A) and (B), the mean velocities from 14 individuals have been realigned to response onset, and in (C) these have been further averaged across individuals. In (D) and (E), the velocity profiles in each trial have been realigned to peak velocity and then averaged for each of the 14 subjects, and in (F) these have been further averaged across subjects. The mean ± 2 SEM and the spread of individual % changes in velocity upon tACS at 20 Hz are shown to the right of (C) and (F). Green and blue arrows in (C) and (F) draw attention to periods in which mean velocity is slower and faster during stimulation. The bars of corresponding color along the x axes in (C) and (F) highlight periods of significant difference (serial two-tailed paired t tests p < 0.05).

Task
Upper panel shows a schematic of the trajectory of the target (red circle and arrows) on the monitor screen and 25 superimposed trials of movement of the cursor controlled by the joystick (blue traces). Only those trials performed during 20 Hz stimulation are shown. The coordinates are as measured on the screen, with 0 being the primary position in the center of the screen. A 5 cm movement of the cursor on the screen necessitated 4 degrees of movement of the joystick. Below the panel is the timeline of events. The principle behavioral event of interest was the rapid movement of the joystick to bring the cursor to the new position of the target after its jump (see yellow-shaded box in upper panel). Neither 20 Hz nor 5 Hz stimulation affected the reaction time or mean velocity of tracking of the continuous movement of the target later in the trial (Table S2). In stimulation trials, the initial target jump occurred 2 s after the onset of stimulation.

Time-Evolving Coherence between Scalp-Recorded Activity and Rectified EMG with and without 20 Hz Stimulation
Scalp activity was recorded from a bipolar pair of electrodes placed medial to the stimulating sponge electrode and EMG from the first dorsal interosseous muscle of the hand holding the joystick.
(A) Time-evolving coherence at 20 Hz during stimulation of the contralateral motor cortex at 20 Hz (red line) and without stimulation (blue line). Color-coded shaded areas represent ± standard error of the mean (SEM). There was a delayed increase in coherence during stimulation. Boxed area indicates the period over which coherence differed with and without stimulation (serial paired t tests over time, p < 0.05). Time is indicated with respect to jump in target spot. Smaller inset is time-evolving coherence between scalp-recorded activity and rectified EMG with time now indicated with respect to the onset of the movement made to catch up with the target jump.
(B) Time-evolving spectrum of coherence during stimulation of the contralateral motor cortex at 20 Hz showing that stimulation did not result in any other spectral change.
Arrow denoting timing of stimulation applies to (A) and (B). Data in (A) and (B) averaged over all subjects. Time-evolving coherence was determined with overlapping 1 s blocks centered on the timing used to denote each block.