Functional connections between and within brain subnetworks under resting-state

Abstract

functional connections between and within brain subnetworks under resting-state fabrizio parente * & Alfredo colosimo the focus of this paper is on the functional role of brain regions focusing on their modular architecture and individual variability. our main assumption is that the more variable anti-correlation patterns reflect random connections, while the more conserved ones play a functional role. Within this framework, we expanded on previous results using a different database and a different methodological approach. Aiming to identify the role of specific functional connections within a global network organization which includes subnetworks, we found that the fronto-parietal module acts as the main source of anti-correlations. In addition, the pre-frontal regions (namely: frontal middle, frontal middle orbital, frontal inferior triangular) and the parietal inferior region are highly conserved and, at the same time, act as highly connected nodes, thus confirming their importance in functional modulation. fMRI technology allows to explore brain regions' activities and, in the case of spontaneous activation, reveals an organized structure with peculiar segregation and integration properties 1-3. In addition to mutual activation among different regions (corresponding to positive correlations), another kind of interactions have been measured, namely anti-correlations (or anti-phase signal dependences) 4-8. A major role was assigned to these interactions in the reciprocal de-activation between the Default Mode Network (DMN) and the Task Positive Network (TPN) 9-11. Several papers associated anti-correlation to schizophrenia 8 , working memory 6 , aging 6 , NMDA receptor 4,12 as well as intracranial electrophysiological measurement 5,13 and magnetoencephalographic recordings 14. A basic methodological problem, however, was raised by Murphy 15 , underlining the role of the Global Signal Regression (GSR) method in the insertion of artifactual anti-correlations. Alternative methods were designed to face that problem 16-18 and, more recently, evidences were obtained indicating a reproducible anti-correlation pattern between the DMN and the Dorsal Attention Network (DAN), as well as the Salience Network (SN) 19,20. The list of unsolved problems, however, still includes:

Modelling a Multiplex Brain Network by Local Transfer Entropy

“…All positive interactions were validated on the whole data set, while the negative ones seem less accurate. This agrees with the higher variability of anti-correlations in 0-lag correlations estimates of brain connectivity 11 . A possible explanation is the different synchronization timing for the negative connectivity 14,48 .…”

“…In this regard, we hypothesize that the mutual deactivations are expressed by different hubs of TPN and DMN connecting different non-hub regions, in terms of crossing but not overlapping structures. This architecture would be in line with the model proposed by the network analysis of the (static) negative brain functional connectivity 11 .…”

“…It is quite interesting that the ActO rule appears associated to the pre-frontal brain regions which, according to previous evidence, are the main hubs of an anticorrelated network 11 . These regions, have been recently associated to a causal interaction suppressing the DMN activity 64 while previous studies identified, by a DCM approach 15 , the dorsal attention network (superior parietal lobe, FEF and inferior frontal gyrus) and the salience network (anterior cingulate, anterior insula and prefrontal cortex) as inhibitory nodes on the DMN, but not vice versa.…”

“…Notice that in the state of rest (null-activation), not only activation but also deactivation are possibly significant events. Thus, to account for the anti-correlated interactions 8,9,10,11 , we distinguish the null-activation (0) from the deactivation (-1).…”

“…Another relevant issue in the study of interacting brain regions is the assumption that functional connectivity values can be split into positive and negative 8,9,10,11 , corresponding to in-phase and anti-phase signal dependence, respectively. As for the negative functional connectivity, although the pre-processing methods (e.g.…”