Extended Data Fig. 15: Single unit response to silencing evoked DS2.

a, A coronal brain slice showing probe locations in secondary motor regions (M2). A red arrow points to the probe track revealed by DiI on the tips of probes. b, Channelrhodopsin (ChR2) expression in the DG inhibitory neurons. Upper and lower panels show examples of low and high ChR2 expression, respectively. Red arrows point to locations of optic fiber tips. c, Number of ChR2-positive cells in the DG for each mouse. Cells were counted from two consecutive 60-μm brain slices with the deepest optic fiber track. Mice with cell count lower than 20 were considered low ChR2 expression. d, Normalized firing rate of single units in DG and M2 aligned to sensory stimulus onsets. Units were separated into low and high ChR2 expression. Right and left panels show unit response to sensory stimulus with and without laser stimulation. e, Proportion of single units that significantly change their firing rates with laser stimulation. Trial indices were shuffled 1000 times to generate null distribution of normalized firing rate for each unit. Mean normalized firing rates were taken from 0 to 0.08 s after the onset of sensory stimuli. Units with mean normalized firing rate greater or smaller than 99th or 1st percentile of the null distribution (Monte Carlo’s p value less than 0.01) during the trials with laser were considered significantly increased or decreased by laser stimulation, respectively. Proportions of each type of units (i.e., no change, increase, and decrease) were significantly different between virus expression levels for each brain region (two-sided Chi-square test for DG: X2 (2, n = 122) = inf, for M2: X2 (2, n = 103) = 12.8). Data were collected in the authors’ lab.