Small-World Propensity Reveals the Frequency Specificity of Resting State Networks

Abstract

Functional connectivity (FC) is an important indicator of the brain's state in different conditions, such as rest/task or health/pathology. Here we used high-density electroencephalography coupled to source reconstruction to assess frequency-specific changes of FC during resting state. Specifically, we computed the Small-World Propensity (SWP) index to characterize network small-world architecture across frequencies. Methods: We collected resting state data from healthy participants and built connectivity matrices maintaining the heterogeneity of connection strengths. For a subsample of participants, we also investigated whether the SWP captured FC changes after the execution of a working memory (WM) task. Results: We found that SWP demonstrated a selective increase in the alpha and low beta bands. Moreover, SWP was modulated by a cognitive task and showed increased values in the bands entrained by the WM task. Conclusions: SWP is a valid metric to characterize the frequency-specific behavior of resting state networks.

Investigating the spectral features of the brain meso-scale structure at rest

“…When presenting the results, we localized the wavelet carrier frequencies (i.e. 2, 4, 8, 16, 32 and 64 Hz) to the corresponding EEG spectral bands, as in our previous work (29), where the bands are defined as delta (δ, 1-4 Hz), theta (θ, 4-8 Hz), alpha (α, 8-13 Hz), beta (β, 13-30 Hz), and gamma (γ, 30-80 Hz).…”

“…We followed the same procedure as detailed in (Iandolo et al, 2020). Briefly, we used T1-weighted structural images in order to generate a realistic volume conductor model.…”

“…Among all, we applied spatial filtering during sources reconstruction, which mitigates the negative effects of volume conduction on FC (34). Indeed, we employed spatial filtering for the analysis in the time domain, whereas, for spectral analysis, we combined both spatial filtering and the method of power spectra orthogonalization (27), which has been largely accepted and validated (35) even by recent studies employing both low and high density montages (17,26,29,36). Moreover, time domain analysis is a valid tool to estimate the Kbest and for the further comparisons of the meso-scale structure in the frequency domain, because the time-courses of source-reconstructed hdEEG signals contain all the neuronal oscillations of interest, that are then band-pass filtered during spectral analysis.…”

“…Specifically, we tested whether the meso-scale structure is frequency-dependent, by examining if assortative, disassortative and core-periphery community motifs are tuned onto a specific frequency or if they are equally distributed over the frequency domain. To address these questions, we applied the WSBM to FC adjacency matrices estimated from source-localized hdEEG recordings of healthy participants (Iandolo et al, 2020;Liu et al, 2017;Samogin et al, 2019), in order to define the spatial cerebral distribution of modules at different number of partitions (K th ). Then, we described the assortative, disassortative and core-periphery community interactions across the frequency spectrum and we identified those brain regions exhibiting the highest degree of each community motif.…”

“…In other terms, we examined if assortative, disassortative and core-periphery modalities are tuned onto a specific frequency or they are equally distributed over the frequency domain. To address these questions, we applied the WSBM to FC adjacency matrices estimated from source-localized hdEEG recordings of healthy participants (17,18,29). We first selected the best number of communities to perform WSBM community detection with a data-driven approach.…”