Brain structure-function relationships across the human lifespan based on network eigenmodes

While brain function is considered to be tightly supported by the underlying structure, the connectome-based link estimated by current models is relatively moderate, leaving the structure-function relationship an ongoing challenge in neuroscience. Here, by proposing a novel mapping method based on network eigendecomposition, we present a concise and strong correspondence between structure and function. We show that the explanation of functional connectivity can be significantly improved by incorporating interactions between different structural eigenmodes, highlighting the potential importance of collective, higher-order coupling patterns between structure and function. We also demonstrate the pronounced advantage of the present mapping in capturing individual-specific information, and apply it to assess individual differences of structure-function coupling across the lifespan. We find that structure-function liberality weakens with age, which is driven by the decreases in functional components that are less constrained by anatomy, while the magnitude of structure-aligned components is preserved. Our results contribute to a more refined understanding of structure-function coupling and how it evolves with age.