The most relevant human brain regions for functional connectivity: Evidence for a dynamical workspace of binding nodes from whole-brain computational modelling

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

  • Gustavo Deco, Center for Brain and Cognition, Computational Neuroscience Group, Department of Information and Communication Technologies, Universitat Pompeu Fabra, Roc Boronat 138, Barcelona, 08018, Spain; Institució Catalana de la Recerca i Estudis Avançats (ICREA), Universitat Pompeu Fabra, Passeig Lluís Companys 23, Barcelona, 08010, Spain.
  • ,
  • Tim J Van Hartevelt, Department of Psychiatry, University of Oxford, Oxford, UK; Center for Music in the Brain (MIB), Dept. of Clinical Medicine, Aarhus University, DK.
  • ,
  • Henrique M Fernandes
  • Angus Stevner
  • Morten L Kringelbach

In order to promote survival through flexible cognition and goal-directed behaviour, the brain has to optimize segregation and integration of information into coherent, distributed dynamical states. Certain organizational features of the brain have been proposed to be essential to facilitate cognitive flexibility, especially hub regions in the so-called rich club with shows dense interconnectivity. These structural hubs have been suggested to be vital for integration and segregation of information. Yet, this has not been evaluated in terms of resulting functional temporal dynamics. A complementary measure covering the temporal aspects of functional connectivity could thus bring new insights into a more complete picture of the integrative nature of brain networks. Here, we use causal whole-brain computational modelling to determine the functional dynamical significance of the rich club and compare this to a new measure of the most functionally relevant brain regions for binding information over time ("dynamical workspace of binding nodes"). We found that removal of the iteratively generated workspace of binding nodes impacts significantly more on measures of integration and encoding of information capability than the removal of the rich club regions. While the rich club procedure produced almost half of the binding nodes, the remaining nodes have low degree yet still play a significant role in the workspace essential for binding information over time and as such goes beyond a description of the structural backbone.

Original languageEnglish
JournalNeuroImage
Volume146
Pages (from-to)197-210
ISSN1053-8119
DOIs
Publication statusPublished - 5 Nov 2016

See relations at Aarhus University Citationformats

ID: 104559715