Spatially structured heterogeneity shapes large-scale cortical dynamics in a model of the human cortex

Abstract

The human brain displays substantial spatial variability in molecular, anatomical, and physiological organization. Yet, how this heterogeneity shapes large-scale neuronal dynamics remains poorly understood. To address this question, we employed a biologically informed large-scale cortical model capable of generating distinct brain states, from awake-like to sleep-like dynamics. Our model was constrained by empirical human structural connectivity (SC) and regional cholinergic muscarinic receptor (CHRM) maps derived from transcriptomic data, together with complementary positron emission tomography (PET)-based receptor maps. These regional maps were implemented as modulators of adaptation-related excitability. We found that modulating excitability according to the spatial organization of CHRM maps significantly impacted large-scale cortical dynamics: It not only facilitated network synchronization but also enhanced information flow between cortical regions. Importantly, these effects were conserved across transcriptomic and PET-derived maps and could not be fully reproduced by multiple null models preserving generic forms of heterogeneity. Moreover, we addressed a particularly intricate dynamic regime characterized by the coexistence of localized sleep-like activity within otherwise awake-like states. We showed that the emergence of these sleep-like slow waves was a byproduct of both regional levels of neuronal adaptation and SC. In summary, our findings highlight the critical role of molecular and anatomical heterogeneity in shaping widespread cortical dynamics, suggesting broad avenues for linking microscale diversity to macroscale function.