Multi-Scale Parcellation of Dynamic Causal Models of the Brain

The hierarchical organization of the brain’s distributed network has received growing interest from the neuroscientific community, largely because of its potential to enhance our understanding of the human cognition and behavior, in health and disease. While most multiscale connectivity analyses focus on structural and functional networks, characterizing the effective connectome across multiple scales has been somewhat overlooked – mostly for computational reasons. The difficulty of estimating large cyclic causal models on the one hand, and the scarcity of theoretical frameworks for systematically moving between scales on the other hand, have hindered progress in this direction. This technical note introduces a top-down multi-scale parcellation scheme for dynamic causal models, with application to neuroimaging data. The method is based on Bayesian model comparison, as a generalization to the well-known Δ BIC method. To facilitate the computations, recent developments in linear dynamic causal modeling (DCM) and Bayesian model reduction (BMR) are deployed. Specifically, a naïve version of BMR is introduced, using which the parcellation scheme can be scaled up to hundreds or thousands of regions. Notably, the derivations reveal an analytical relationship between the model comparison score and the concept of cut size in graph theory. This duality puts the tools of graph theory at the service of model evidence optimization and significance testing. The proposed method was applied to simulated and empirical causal models, for face and construct validity. Notably, the large causal network (inferred from a neuroimaging dataset) showed scale-invariance in multiple measures. Future generalizations of this technique and its potential applications in systems and clinical neuroscience have been discussed.