Clarify the role of higher-order interactions in macroscopic brain dynamics

Determine the role and relative importance of higher-order (nonpairwise) interactions, compared to pairwise interactions, in shaping macroscopic brain dynamics by establishing whether and when higher-order interactions are necessary to explain neurophysiological time-series data such as EEG, iEEG, or fMRI recordings.

Background

The paper introduces the Taylor-based Hypergraph Inference using SINDy (THIS) algorithm to reconstruct hypergraphs and simplicial complexes from time-series data without assuming known node dynamics or coupling functions. Applying THIS to resting-state EEG data, the authors find that nonpairwise interactions account for a substantial fraction of macroscopic brain dynamics.

Despite these findings, the broader literature presents conflicting evidence: some studies suggest linear or pairwise models suffice for macroscopic brain activity, whereas others support significant higher-order dependencies. This discrepancy leads to a fundamental unresolved question about the necessity and magnitude of higher-order interactions in explaining brain dynamics across modalities and conditions.

References

Further studies are needed to clarify the role of higher-order interactions in the brain. We hope our inference method can provide a new tool and a novel perspective on this important open question in neuroscience.

— Hypergraph reconstruction from dynamics  (2402.00078 - Delabays et al., 2024) in Section: How important are higher-order interactions in shaping macroscopic brain dynamics? (end of section, preceding Discussion)

At the same time, this raises an important question: how does the brain achieve fast and efficient communication despite such high-order complexity?

— High-Order Triadic Functional Connectivity in the Brain and Beyond  (2609.03987 - Li et al., 3 Sep 2026) in Section “Challenges and Opportunities,” subsection “High-Order Complexity”