Mapping macroscale structure-function relationships in the brain: a multimodal neuroimaging perspective

The structure of a system necessarily constrains its functionality, establishing more or less strict boundaries on the system's behavior. The brain is no exception and since the advent of the field of "connectomics", researchers have sought to understand the precise relationship between structural wiring and functional connectivity, aiming to uncover the fundamental laws governing the brain's multiplexed organization. What quickly became evident is that this relationship can be described not only as an "imperfect correspondence" but also as a local phenomenon: function increasingly decouples from structure as one moves from unimodal to transmodal cortices. This line of research is now focused on understanding why this occurs and which factors shape the regionality of structure-function relationships. This doctoral dissertation fits within this framework, beginning with the structural connectome mapped through Diffusion Weighted Imaging (DWI) tractography and examining how neural activity, measured via resting-state functional Magnetic Resonance Imaging (fMRI) and Magnetoencephalography (MEG), is locally and dynamically coupled or uncoupled from the underlying structure. The overall thesis comprises three separate experimental works aimed at exploring different aspects of the structure-function coupling problem. In the first study, we focus on linking structure-function coupling and its temporal variability to the concept of brain hierarchies. Furthermore, we combine univariate and multivariate approaches to leverage multimodal datasets, ranging from ex-vivo gene transcription to in-vivo fiber-length mapping, to infer the factors that shape time-resolved structure-function relationships. In the second study, we explore the local interplay between neurophysiological and structural connectivity and identify regional variations in cortical microarchitecture that correspond to variations in coupling. In the final study, we adopt a dynamic states perspective to challenge the prevailing assumption that the alignment between structure and function is stably higher in unimodal than transmodal cortices. Overall, we demonstrate that the tethering between structure and function is much more intricate and multiplexed than previously believed, and that integrating the temporal dimension is crucial for shedding light on this "imperfect correspondence".