Fig. 5: Goal-associated decreases in interneuron activity occur in the CA1 during a choice task.
a, WT mice (n = 7) chose between left and right arms based on visual cues displayed on the centre walls of a virtual Y-maze. On most trials (top, delay only), the initial cue (the black dashed line marks start of initial cue presentation) indicated the rewarded arm (checkmark). On a subset of trials, an additional (update) cue was presented (teal dashed line), which indicated whether the rewarded arm stayed the same as the initial cue (stay, middle), or switched to the opposite arm (switch, bottom). The mice received a reward after entering the correct RZ at the end of the rewarded arm (pink dashed line). b, Electrophysiology analysis during Y-maze navigation. The mouse illustration was adapted from scidraw.io. c, Top, schematic indicating the rewarded areas. Left, the normalized residual firing rate of NS interneurons. Right, the cell-averaged percentage change in normalized residual firing as a function of the time to the RZ (pink) for NS interneurons. n = 485 cells. There was a significant effect of time to RZ (P < 2.2 × 10−16, LMM). The blue bar indicates bins that significantly differ from zero or the baseline (two-sided t-test followed by Bonferroni correction). d, The normalized firing rate and the percentage change in firing rate as in c, but for the time around the update cue (teal, n = 485 cells). There was a significant effect of time to the update cue (P = 5.19 × 10−13, LMM); however, no individual time bins were significantly different. For c,d, data are mean ± s.e.m. The update task and trial type41 (a,c,d), VR schematic41 (b), mouse head (E. Tyler and L. Kravitz; b) and the brain in mouse head (F. Claudi; b) diagrams were adapted under a CC BY 4.0 licence.
