Long-Range White-Matter Pathways Enable Efficient Spontaneous Neural Activity Propagation in the Human Brain

Abstract

Efficient brain-wide communication requires neural activity to traverse long anatomical distances rapidly. Here we examine how propagation timing is jointly associated with spatial geometry, functional network organization, and long-range white-matter pathways and their microstructural properties. And we ask whether the same rules govern epileptiform and physiological activity. Using stereo-EEG and diffusion spectrum imaging from 47 epilepsy patients (26 males and 21 females), we quantified interregional propagation with two complementary delay estimators: event-based interictal epileptiform discharge (IED) traveling waves and continuous lagged-correlation delays during IED-free periods. We found that IED propagation traversing gray and white matter formed reproducible spatiotemporal motifs that deviated from randomized null models, indicating structured routing rather than random spread. Epileptiform and physiological propagation delays increased over short ranges but saturated at longer distances, indicating that geometry alone cannot account for long-range fast propagation. Beyond geometry, stronger structural connectivity and higher functional connectivity were associated with shorter delays, and intrinsic functional modules facilitated efficient communication: within-network propagation was faster than between-network propagation. Crucially, diffusion-derived quantitative anisotropy (QA) revealed a microstructural mechanism for long-range fast propagation: long-range white-matter tracts showed higher QA, and QA was positively associated with apparent propagation velocity. Together, these results identify convergent, architecture-dependent constraints on propagation timing that generalize across epileptiform and normal activity, providing a principled bridge between macroscale connectome organization and fast intracranial spatiotemporal dynamics.