Skip to main content
. 2020 Jul 13;9:e57011. doi: 10.7554/eLife.57011

Figure 4. Cortical-hippocampal communication is enhanced during long ripples.

Figure 4.

(A) Top: histogram of ripple durations pooled across all patients. Red vertical lines indicate the average cut-offs for short (mean ± SEM duration=0.046 ± 0.001 s) and long ripple events (mean ± SEM duration=0.059 ± 0.002 s), based on patient-specific tertiles (short ripples: first tertile; long ripples: third tertile). Bottom: Grand average rippled-locked time-frequency representations (TFRs) as well as example traces (raw and filtered between 80 and 120 Hz) exemplifying short and long ripples. Color bar represents percentage change to a pre-event baseline from −2 to −1.5 s. (B) Statistical maps (t-values vs. control events) for neocortical TFRs (NC) locked to short (left) or long (right) hippocampal ripples. Hot colors indicate power increases around SW-Rs, whereas cold colors indicate relative power decreases. White contours mark significant clusters obtained from a cluster-based permutation procedure (p<0.05, corrected). (C) Time course of partial directed coherence (PDC; z-transformed vs. control events) in the 12–16 Hz spindle range locked to short (gray) and long (black) ripples. Positive values signify a cortical to hippocampal information flow and vice versa for negative values. Blue lines mark standard significance thresholds (z = 1.96; p<0.05).