Connectome modulates critical brain dynamics across local and global scales
Neuronal activity in the brain has been hypothesized to operate near criticality, a dynamical regime poised between order and disorder that maximizes information processing, adaptability, and dynamic range. While criticality has been extensively studied at local scales (within neuronal populations) and at global scales (across interacting brain regions), the interplay between these levels remains poorly understood. Here, we propose a multiscale computational framework that bridges and criticality within a single, mechanistic model. At the mesoscopic level, individual brain regions are represented by neural mass models tuned near the transition between asynchronous and synchronous regimes. These regions are then coupled via an empirically derived mouse connectome to investigate how structural connectivity shapes the emergence of large-scale coordination. We show that (1) local near-critical dynamics for an isolated brain region can be faithfully reproduced within a mean-field model framework, (2) local distance to criticality is modulated by long-range coupling, (3) whole-brain simulations reveal nonlinear gradients of timescales and heterogeneous shifts toward/away from local criticality, and (4) global criticality, manifested in scale-free avalanche distributions and optimal functional connectivity, emerges when local populations are locally tuned near criticality and coupled within an optimal range, a result that we confirm in whole-brain simulations of connectome-coupled spiking neurons. These results demonstrate that local and global criticality are dynamically intertwined but not directly aligned, and that their relationship depends on the underlying structural connectivity. Our multiscale modeling framework provides a tractable tool for generating testable hypotheses on how brain criticality co-arises across scales and how it may be modulated in health and disease.
Authors
- Pierpaolo Sorrentino (ORCID: https://orcid.org/0000-0002-9556-9800)
- Leonardo L. Gollo (ORCID: https://orcid.org/0000-0003-3505-9259)
- Damien Depannemaecker (ORCID: https://orcid.org/0000-0003-1936-4560)
- Giovanni Rabuffo (ORCID: https://orcid.org/0000-0003-3947-1662)
- Marco N. Pompili (ORCID: https://orcid.org/0000-0003-3567-6345)
- Tomoki Fukai (ORCID: https://orcid.org/0000-0001-6977-5638)
- Leonardo Dalla Porta (ORCID: https://orcid.org/0000-0003-3104-288X)
- Pietro Bozzo
- Bach Nguyen
Institutions
- Consejo Superior de Investigaciones Científicas (ES)
- Okinawa Institute of Science and Technology Graduate University (JP)
- University of Sassari (IT)
- Universitat Pompeu Fabra (ES)
- Institute for Cross-Disciplinary Physics and Complex Systems (ES)
- Institut de Neurosciences des Systèmes (FR)
- Consorci Institut D'Investigacions Biomediques August Pi I Sunyer (ES)
- Universitat de les Illes Balears (ES)
- Monash University (AU)
Publication Details
- Journal
- Physical Review Research
- Published
- 2026-11-17
- DOI
- https://doi.org/10.1103/n87d-5krk
- Citations
- 2
- Primary Topic
- Neural dynamics and brain function
- Type
- article
- Field-Weighted Citation Impact
- 3.68