Extended Data Fig. 7: Hippocampal event features in PBn-HC recordings of macaque monkeys. | Nature

Extended Data Fig. 7: Hippocampal event features in PBn-HC recordings of macaque monkeys.

From: Coupling of hippocampal theta and ripples with pontogeniculooccipital waves

Extended Data Fig. 7

a, Detected SWR (red arrows) and theta-like events in an exemplary hippocampal raw LFP trace (0.5-190 Hz) and its filtered version in the ripple frequency band (90-190 Hz). SWR are clearly marked by the occurrence of high-frequency oscillations as shown in the top insets. b, Theta- (left panel), gamma- (middle panel) and ripple-triggered (right panel) complex Morlet-wavelet power spectrograms. c, Averaged zero-lag spectral profiles in correspondence with each hippocampal event type. Shaded areas indicate SEM. d, Bivariate distribution of FWHM versus frequency peak (after Logothetis et al., 2012) (left panel). e, f, Frequency peak (e) and FWHM (f) box plots are also shown separately. Colours indicate the hippocampal event type shown in B. On each box, the top and bottom are the 25th and 75th percentiles of the samples, respectively; the middle dot of each box is the sample median; the dashed lines extending below and above each box are drawn to the extreme points; black crosses in the diagrams are the outliers of the samples. g, Theta- (top), gamma- (middle) and SWR-triggered (bottom) BLP signals population grand averages in and beyond the LFP frequency range: delta (1–4 Hz), theta (4–10 Hz), spindle (10–16 Hz), sigma (16–30 Hz), slow and high gamma (30–70 Hz; 70–90 Hz, respectively), ripple (90–190 Hz) and high-frequency oscillations (190–320 Hz) and MUA (801–2500 Hz). BLP signals were z-scored and averaged across all recording tips. All responses were thresholded for a significant deviation from zero at a statistical significance level P < 0.05 (permutation test) and Bonferroni-corrected across frequency bands. h, Time course population averages of BLP modulations in hippocampus over MUA (801–2500 Hz) for SWR (top), gamma (middle) and theta (bottom) episodes. Note the prevalent negative modulation associated with theta events, as opposed by a consistent increase of activity associated with ripple events. i, Point-process conditional intensities (cross-correlograms) accounting for the relationship between the occurrences of theta, gamma and SWR episodes. Conditional intensities were computed from pooled experiments across pons-hippocampus experimental sessions, at a resolution of 50 ms bins. Asterisks indicate a significant relationship (P < 0.01, Bonferroni-corrected) at and around the time of the occurrence of theta (left), gamma (middle) and SWR (right) events, according to an H0 distribution derived from cross-correlograms of randomized point processes of the same rate. The reference event for computing the conditional intensity is indicated by the first event of each pair in the top part of each plot. For panels bi, N = 24 experimental sessions contributed to the statistics; animals CM031, F12 and K13; virtually identical event profiles were obtained with sessions from animals A13, D11, i09 and e10).

Back to article page