Efficient hubs in the intelligent brain: Nodal efficiency of hub regions in the salience network is associated with general intelligence

- We investigate intelligence-related differences in brain network organization.
- Graph analysis was used to study the intrinsic functional connectivity of the brain.
- Intelligence (IQ) was not related to global efficiency of network organization.
- High IQ was associated with high nodal efficiency in the brain's salience network.
- High IQ was associated with low nodal efficiency of the temporo-parietal junction.

Intelligence-related differences in the intrinsic functional organization of the brain were studied with a graph-theoretical approach, comparing effects on nodal measures of brain network efficiency (concerning specific nodes of the network) and global measures (concerning the overall brain network). Functional imaging data acquired for 54 healthy adult participants during wakeful rest were modeled as graphs representing individual functional brain networks. Nodal and global measures of efficient network organization (i.e., nodal efficiency and global efficiency) were correlated with intelligence scores (IQ from the Wechsler Abbreviate Scale of Intelligence, WASI). While global efficiency showed no significant association with intelligence, the nodal efficiency was significantly associated with intelligence in three brain regions. Participants with higher IQ scores showed higher nodal efficiency in right anterior insula (AI) and dorsal anterior cingulate cortex (dACC), two hub regions of a functional brain network previously described as salience network. Furthermore, higher IQ was associated with lower nodal efficiency in the left temporo-parietal junction area (TPJ). Distinct connectivity profiles were observed for brain regions showing a positive versus negative correlation between IQ and nodal efficiency. Our analyses suggest that intrinsic (i.e., task-independent) connectivity profiles of brain regions that have previously been associated with salience processing (AI and dACC) and the filtering of irrelevant information from higher-level processing (TPJ), play a role in explaining individual differences in intelligence. Based on these intelligence-related effects in resting-state fMRI data, we discuss the potential relevance of processing salient information for the explanation of differences in cognitive performance and intelligence.