Traces of EEG-fMRI coupling reveals neurovascular dynamics on sleep inertia

Figures

• 

  Figure 1. Illustration of experimental design and analytical indices. During sleep inertia, the neurovascular coupling may change along with this recovery period of consciousness. We probed the transitory changes of EEG-fMRI coupling, surrogates of neurovascular coupling, depending on the extended awakening time (A1 to A3).

• 

  Figure 2. EEG arousal features across four imaging sessions. (A) Averaged theta/beta ratio per imaging session per each subject was used to compare across four imaging sessions (Friedman test p = 0.09). Theta/beta ratio in A3 is higher than A2 but without statistical significance (Wilcoxon sign-rank test p = 0.06, FDR-corrected). (B) Spectral slope was found to fluctuate significantly across imaging sessions (Friedman test p = 0.025), but no pairs of imaging sessions were found to be significantly different. (C) No significance was found for alpha-vigilance across imaging sessions (Friedman test p > 0.1).

• 

  Figure 3. Time lag of peak positive correlation among sessions across sleep inertia. (A) Time lags where maximum positive correlation between thalamus and alpha-vigilance occurs across a [− 6, 14] s range were plotted for each subject. (B) There is a significant effect by imaging session (p = 0.02), where A1 and A2 were found with peak-correlation timings that are significantly later than A3 (**p = 0.01, FDR-corrected). (C) Time lags where maximum positive correlation between thalamus and EEG spectral slope occurs across a [− 6, 14] s range. (D) There is a significant effect by imaging session (p < 0.01), where A2 is found to be significantly earlier than A3 (*p = 0.03, FDR-corrected).