Fig. 5: The power-power cross-frequency coupling (CFC) between theta power and gamma power in TCD syndromes is stronger than that between alpha power and gamma power.

a Brain atlas heatmaps depict group-level differences (t-values) between CFC. Since CFC coefficients can represent either a direct or inverse relationship between frequencies, the higher the t values, the greater the difference in the magnitude of the coupling. Participants with FXS (n = 70) showed a significantly higher magnitude of inverse theta-gamma1 CFC compared to controls (n = 71), especially over the parietal and central regions. b Brain atlas plots with shaded areas representing 5% FDR-corrected significance of pairwise contrasts. c Bar plots of mean ± standard error of least-squared mean estimates (with superimposed subject-level data) of CFC by RSN. Asterisks signify significant mean differences between FXS (red) and control (gray) groups. An accompanying mean ± standard error model plot of group by CFC type across RSN demonstrates a prominent increase in inverse theta-gamma1 CFC and an increase in direct alpha2-gamma CFC in the FXS group. d Exemplar clinical correlations with EEG variables across all FXS subjects and a subgroup of full mutation, non-mosaic males (FM). Scatterplots in each quadrant depict subject-level bivariate and age-corrected partial Spearman’s correlations. Resting-state network abbreviations: DMN default mode network, DAN dorsal attention network, SAN salient affective network, VIS visual attention network, AUD auditory network. FDR-adjusted p values, *adj. p < 0.05; ***adj. p. < 1 × 10⁻⁵.