P053 Modular structure-function support high-order interactions in the human brain
Jesus M Cortes1,2,3, Borja Camino-Pontes1,4, Antonio Jimenez-Marin1,4, Iñigo Tellaetxe-Elorriaga1,4, Izaro Fernandez-Iriondo1,4, Asier Erramuzpe1,2, Ibai Diez1,2,5, Paolo Bonifazi1,2, Marilyn Gatica6,7, Fernando Rosas8,9,10,11, Daniele Marinazzo12, Sebastiano Stramaglia13,14
*Email: jesus.m.cortes@gmail.com
1Computational Neuroimaging Lab, BioBizkaia Health Research Institute, Barakaldo, Spain
2IKERBASQUE: The Basque Foundation for Science, Bilbao, Spain
3Department of Cell Biology and Histology, Faculty of Medicine and Nursing, University of the Basque Country, Leioa, Spain
4Biomedical Research Doctorate Program, University of the Basque Country, Leioa, Spain
5Gordon Center for Medical Imaging, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USA
6NPLab, Network Science Institute, Northeastern University London, London, United Kingdom.
7Precision Imaging, School of Medicine, University of Nottingham, United Kingdom.
8Department of Informatics, University of Sussex, Brighton, United Kingdom.
9Sussex Centre for Consciousness Science and Sussex AI, University of Sussex, Brighton, United Kingdom.
10Center for Psychedelic Research and Centre for Complexity Science, Department of Brain Sciences, Imperial College London, London, UK.
11Center for Eudaimonia and Human Flourishing, University of Oxford, Oxford, United Kingdom.
12Department of Data Analysis, Ghent University, Ghent, Belgium.
13Università degli Studi di Bari Aldo Moro, Bari, Italy.
14INFN, Sezione di Bari, Italy.
Introduction
The brain exhibits a modular organization across structural (SC) and functional connectivity (FC), spanning multiple scales from microcircuits to large-scale networks. While SC and FC share similarities, FC fluctuates over shorter time scales. Structure-function coupling (SFC) examines statistical dependencies between SC and FC [1], often at the link-wise level. However, modular coupling offers a multi-scale approach to understanding SC-FC interactions [2-3]. This study integrates functional MRI and diffusion-weighted imaging to investigate modular SFC and the role of high-order interactions (HOI) in functional organization.
Methods
We analyzed SC and FC from multimodal neuroimaging data, using graph-based modular decomposition to assess brain network structure. To quantify HOI, we computed O-information [4], assessing redundancy and synergy among brain regions. HOI gradients were also derived to explore the organization of these interactions [5]. We then examined the coupling between modular SC and both redundancy and synergy, identifying statistical associations that reveal how structural networks relate to functional integration and segregation.
Results & Discussion
Our findings indicate that SC is linked to both redundant and synergistic functional interactions at the modular level. SC showed both positive and negative correlations with redundancy, suggesting that stronger structural connections within a module can either amplify or reduce functional redundancy. In contrast, synergy consistently exhibited a positive correlation with SC, indicating that increased SC density promotes synergistic interactions. These results refine our understanding of structure-function relationships, highlighting how SC modulates HOI in the brain’s modular architecture.
Acknowledgements
JMC acknowledges financial support from Ikerbasque: The Basque Foundation for Science, and from Spanish Ministry of Science (PID2023-148012OB-I00), Spanish Ministry of Health (PI22/01118), Basque Ministry of Health (2023111002 & 2022111031).
References
[1]https://doi.org/10.1038/s41583-024-00846-6
[2]https://doi.org/10.1038/srep10532
[3]https://doi.org/10.1002/hbm.24312
[4]https://doi.org/10.1103/PhysRevE.100.032305
[5]https://doi.org/10.1103/PhysRevResearch.5.013025
