Fig. 2: Adult-born hippocampal neurons are coupled to temporal network dynamics and increase population sparsity.

a, Top: raw wide-band CA1 LFP trace (black) showing example population vectors of PC spikes nested in individual theta cycles (red, theta-filtered signal). Bottom: (color-coded) raster-plot of spike trains from CA1 (orange), CA3 (red) and DG (blue) PCs and abDGCs (purple) (one cell per row). Scale bars, 200 ms (horizontal), 200 µV (vertical). b, Mean spike probability of abDGCs and other DG, CA3 and CA1 PCs as a function of CA1 theta phase (dashed line, with two cycles for clarity). Spike probabilities normalized to the cell’s baseline spike rate. c, Polar plots showing theta preferred phase and modulation depth (mean resultant length, MRL) of individual abDGCs and other PCs in DG, CA3 and CA1 (dots), with group means (black lines). d, Corresponding estimation plot showing the effect size for differences in MRL (w.r.t. DG PCs). Upper and lower plots as in Fig. 1f. e, Spike discharge probability of DG, CA3 and CA1 PCs (top row) and interneurons (bottom row) w.r.t. spontaneous abDGC spike times. For each population: the left panel reports (laser-off) spiking probability referenced to spontaneous abDGC spikes (black histograms; with red line: the corresponding shuffled spike distribution with the original theta phase of each spike preserved); the right panel shows the observed minus the shuffled distribution. f, Change in instantaneous firing rate for three example interneurons in DG, CA3 and CA1 w.r.t. group mean optogenetic activation of 4–7-week-old abDGCs. g, Estimation plot showing sparsity of hippocampal PC population vectors before (gray) versus immediately following (purple) optogenetic activation of 4–7-week-old abDGCs. Each data point represents the mean sparsity for one recording session (n = 52 sessions in 8 mice). Upper and lower plots as in Fig. 1f. ***P < 0.001, *P < 0.05.