Fig. 1: Self-timing lever-pull task and the effects of weak inhibition of LCN and motor cortex on task performance.

‌a Schematic of the self-timing lever-pull task, which required mice to wait for 1–1.71 s before the lever pull. ‌b Lever trajectories in the fourth session (simple go-cue task) and twentieth session (self-timing task) in a representative mouse. Black lines show the mean lever trajectories and gray lines show all individual trajectories (the first 50 trials in each session). c Distribution of the pull latencies in the self-timing task in mice that were used for two-photon calcium imaging of thalamocortical axons (5069 trials from 8 mice). The latency was 1115.7 ± 6.2 ms (mean ± SEM). The number of non-pull trials (429, 8.46% of all trials) is shown on the right. d Schematic of inhibition of the LCN (roughly the area indicated by the oval) using muscimol. ACSF on the first day and muscimol on the second day were injected before the task performance. DN dentate nucleus, IN interposed nucleus, FN fastigial nucleus. e, f Pull latency (e) and pull rate (f) in the sessions in which ACSF or muscimol was administered to the LCN (n = 7 mice). Black lines indicate mean ± SEM. Pull latency, *P = 0.0156; pull rate, P = 0.156, Wilcoxon signed-rank test, two-sided. g Schematic of the inhibition of M2 or M1. ACSF on the first day and muscimol on the second day was injected into M2 before the task performance started, and the same procedure was repeated for the M1 injection on the third and fourth days. h Pull rate in the sessions in which ACSF or muscimol was administered to M2 (left) and M1 (right). Black lines indicate mean ± SEM. M2, *P = 0.0117, n = 9 mice; M1, P = 1, n = 7 mice, Wilcoxon signed-rank test, two-sided. Source data are provided as a Source Data file.