A Multimodal Characterization of Low-Dimensional Thalamocortical Structural Connectivity Patterns

Abstract

The human thalamus is a bilateral and heterogeneous grey matter structure that plays a crucial role in coordinating whole-brain activity. Investigations of its complex structural and functional internal organization revealed to a certain degree overlapping parcellations, however, a consensus on thalamic subnuclei boundaries remains absent. Recent work suggests that thalamic organization might additionally reflect continuous axes transcending nuclear boundaries. In this study, we used a multimodal approach to uncover how low-dimensional axes that describe thalamic connectivity patterns to the cortex are related to internal thalamic microstructural features, functional connectivity, and structural covariance. We computed a thalamocortical structural connectome via probabilistic tractography on diffusion MRI and derived two main axes of thalamic organization. The principal thalamic gradient, extending from medial to lateral and differentiating between transmodal and unimodal nuclei, was related to intrathalamic myelin profiles, and patterns of functional connectivity, while the secondary axis showed correspondence to core-matrix cell type distributions. Lastly, exploring multimodal thalamocortical associations on a global scale, we observed that the medial-to- lateral gradient consistently differentiated limbic, frontoparietal, and default mode network nodes from dorsal and ventral attention networks across modalities. However, the link with sensory modalities varied. In sum, we show the coherence between lower dimensional patterns of thalamocortical structural connectivity and various modalities, shedding light on multiscale thalamic organization.