Network Path Convergence Shapes Low-Level Processing in the Visual Cortex

Bálint Varga; Bettina Soós; Balázs Jákli; Eszter Bálint; Zoltán Somogyvári; László Négyessy

doi:10.3389/fnsys.2021.645709

Network Path Convergence Shapes Low-Level Processing in the Visual Cortex

Front Syst Neurosci. 2021 May 24:15:645709. doi: 10.3389/fnsys.2021.645709. eCollection 2021.

Authors

Bálint Varga^{1

2}, Bettina Soós^{1

3}, Balázs Jákli⁴, Eszter Bálint⁵, Zoltán Somogyvári¹, László Négyessy¹

Affiliations

¹ Computational Neuroscience and Complex Systems Research Group, Department of Computational Sciences, Wigner Research Centre for Physics, Budapest, Hungary.
² János Szentágothai Doctoral School of Neurosciences, Semmelweis University, Budapest, Hungary.
³ Faculty of Science and Engineering, University of Groningen, Groningen, Netherlands.
⁴ Faculty of Information Technology and Bionics, Pázmány Péter Catholic University, Budapest, Hungary.
⁵ Department of Anatomy, Histology and Embryology, Semmelweis University, Budapest, Hungary.

Abstract

Hierarchical counterstream via feedforward and feedback interactions is a major organizing principle of the cerebral cortex. The counterstream, as a topological feature of the network of cortical areas, is captured by the convergence and divergence of paths through directed links. So defined, the convergence degree (CD) reveals the reciprocal nature of forward and backward connections, and also hierarchically relevant integrative properties of areas through their inward and outward connections. We asked if topology shapes large-scale cortical functioning by studying the role of CD in network resilience and Granger causal coupling in a model of hierarchical network dynamics. Our results indicate that topological synchronizability is highly vulnerable to attacking edges based on CD, while global network efficiency depends mostly on edge betweenness, a measure of the connectedness of a link. Furthermore, similar to anatomical hierarchy determined by the laminar distribution of connections, CD highly correlated with causal coupling in feedforward gamma, and feedback alpha-beta band synchronizations in a well-studied subnetwork, including low-level visual cortical areas. In contrast, causal coupling did not correlate with edge betweenness. Considering the entire network, the CD-based hierarchy correlated well with both the anatomical and functional hierarchy for low-level areas that are far apart in the hierarchy. Conversely, in a large part of the anatomical network where hierarchical distances are small between the areas, the correlations were not significant. These findings suggest that CD-based and functional hierarchies are interrelated in low-level processing in the visual cortex. Our results are consistent with the idea that the interplay of multiple hierarchical features forms the basis of flexible functional cortical interactions.

Keywords: anatomical hierarchy; granger causality; hierarchical counterstream; hierarchical dynamics; network resilience; network topology; oscillation; shortest path.