Random Effects Models for Understanding Variability and Association between Brain Functional and Structural Connectivity
🤖 AI Summary
This study addresses the inconsistency between network-level and edge-level conclusions in functional–structural connectivity (FC–SC) coupling analyses. We propose a novel random-effects model that, for the first time, systematically decomposes FC–SC coupling variability into three orthogonal components: subject-level, edge-level, and subject-by-edge interaction effects—enabling their statistical separation and quantitative assessment. Using multilevel modeling and variance decomposition on large-scale neuroimaging data, we find that network-level FC–SC correlations are predominantly driven by population-shared structural constraints, whereas edge-level correlations critically depend on subject-specific variability and FC–SC interaction effects. These results reveal fundamental differences in coupling mechanisms across analytical scales, challenging inferences drawn from single-scale analyses and correcting potential biases in interpreting brain connectome organization. The framework provides a new methodological foundation and theoretical perspective for investigating the hierarchical architecture of FC–SC coupling.
Application Category
📝 Abstract
The human brain is organized as a complex network, where connections between regions are characterized by both functional connectivity (FC) and structural connectivity (SC). While previous studies have primarily focused on network-level FC-SC correlations (i.e., the correlation between FC and SC across all edges within a predefined network), edge-level correlations (i.e., the correlation between FC and SC across subjects at each edge) has received comparatively little attention. In this study, we systematically analyze both network-level and edge-level FC-SC correlations, demonstrating that they lead to divergent conclusions about the strength of brain function-structure association. To explain these discrepancies, we introduce new random effects models that decompose FC and SC variability into different sources: subject effects, edge effects, and their interactions. Our results reveal that network-level and edge-level FC-SC correlations are influenced by different effects, each contributing differently to the total variability in FC and SC. This modeling framework provides the first statistical approach for disentangling and quantitatively assessing different sources of FC and SC variability and yields new insights into the relationship between functional and structural brain networks.
Problem
Research questions and friction points this paper is trying to address.
Analyzing edge-level vs network-level brain connectivity correlations
Explaining discrepancies in functional-structural association strength
Modeling variability sources in brain connectivity data
Innovation
Methods, ideas, or system contributions that make the work stand out.
Random effects models decompose FC-SC variability
Analyze network-level and edge-level FC-SC correlations
Quantify subject, edge, and interaction effects
🔎 Similar Papers
No similar papers found.
