Balance of resting-state functional networks

Background and Aim: The brain is considered as a complex system whose functional connectivity could be modeled using the network. In the functional connectivity network, brain regions are nodes and Pearson's correlations between the regional activities are the links. The links are either positive or negative because of the essence of synchrony or anti-synchrony between their activities. This fact enables us to treat them as a signed network and estimate the brain network stability by exploring triadic relationships between the nodes. In this approach, the network stability can be measured using balance energy. Methods: f/MRI scans of ۵۷ healthy individuals (۱۸-۳۰ years) were selected from ABIDE dataset. After a standard preprocessing, time courses of the regional activities (parcellated by Schaefer's Local-Global cortical atlas) were extracted. RSFN was then calculated and their signed networks were computed. Then, the balance energies of the signed networks were statistically compared with correspondent random simulated signed networks using a Sign test. Topological effects of signed networks on the balance were also investigated by the use of a new global hubness measure named Tendency to Make Hub and estimation of negative degree distributions. Results: In this study, we investigated how changes in the arrangement of signed links affect the balance of resting-state functional network (RSFN). Conclusion: Our results showed that the balance energies of RSFNs are lower than correspondent random networks and RSFNs are more stable. Moreover, simulations showed that the tendency of the negative link to make hubs causes the balance. We also revealed that negative degrees of RSFNs are distributed in an exponential form and rate parameter of the distributions is negatively correlated with negative TMH. These findings demonstrate significant impact of negative links topology on the stability of the RSFNs. In this regard, we introduced a mechanism to describe the impact of topology on the stability of signed networks.

Keywords:

Resting-state functional network , Balance theory , Functional magnetic resonance imaging , Triadic associations , Frustration and stability