Skip to Main content Skip to Navigation

Cerebral connectivity study by functional and diffusion MRI in intelligence

Abstract : The idea that intelligence is embedded not only in specific brain regions, but also in efficient brain networks has grown up. Indeed, human brain organization is believed to rely on complex and dynamic networks in which the communication between cerebral regions guarantees an efficient transfer of information. These recent concepts have led us to explore the neural bases of intelligence using both advanced MRI techniques in combination with graph analysis. On one hand, advanced MRI techniques, such as resting-state functional MRI (rs-fMRI) and diffusion MRI (dMRI) allow the exploration of respectively the functional and the structural brain connectivity while on the other hand, graph theory models allow the characterization of brain networks properties at different scales, thanks to global and local metrics. The aim of this thesis is to characterize the topology of functional and structural brain networks in children and in adults with an intelligence quotient higher (HIQ) than standard levels (SIQ). First, we focused our attention on a children population with different cognitive characteristics. Two HIQ profiles, namely homogeneous (Hom-HIQ) and heterogeneous HIQ (Het-HIQ), have been defined based on clinical observations and Intelligence Quotient (IQ) sub-tests. Using resting-state fMRI techniques, we examined the functional network topology changes, estimating the "hub disruption index", in these two HIQ profiles. We found significant topological differences in the integration and segregation properties of brain networks in HIQ compared to SIQ children, for the whole brain graph, for each hemispheric graph, and for the homotopic connectivity. These brain networks changes resulted to be more pronounced in Het-HIQ subgroup. Finally, we found significant correlations between the graph networks’ changes and the full-scale IQ, as well as some intelligence subscales. These results demonstrated for the first time, that different HIQ profiles are related to a different neural substrate organization. Then, the structural brain network connectivity, measured by dMRI in all HIQ children, were significantly different than in SIQ children. Also, we found strong correlations between the children brain networks density and their intelligence scores. Furthermore, several correlations were found between integration graph metrics suggesting that intelligence performances are probably related to a homogeneous network organization. These findings demonstrated that intelligence neural substrate is based on a strong white matter microarchitecture of the major fiber-bundles and a well-balanced network organization between local and global scales. This children population was finally studied using a memory-word task of fMRI. Significant changes were observed between both HIQ and SIQ groups. This study confirms our hypothesis that both HIQ profiles are characterized by a different brain activity, with stronger evidences in Het-HIQ children. Finally, we investigated both functional and structural connectivity in a population of adults HIQ. We found several correlations between graph metrics and intelligence sub-scores. As well as for the children population, high cognitive abilities of adults seem to be related brain structural and functional networks organization with a decreased modularity. In conclusion, the sensitivity of graph metrics based on advanced MRI techniques, such as rs-fMRI and dMRI, was demonstrated to be very helpful to provide a better characterization of children and adult HIQ, and further, to distinguish different intelligence profiles in children
Document type :
Complete list of metadatas
Contributor : Abes Star :  Contact
Submitted on : Thursday, March 5, 2020 - 2:11:10 PM
Last modification on : Monday, October 19, 2020 - 10:58:38 AM
Long-term archiving on: : Saturday, June 6, 2020 - 4:05:15 PM


Version validated by the jury (STAR)


  • HAL Id : tel-02499621, version 1


Ilaria Suprano. Cerebral connectivity study by functional and diffusion MRI in intelligence. Medical Imaging. Université de Lyon, 2019. English. ⟨NNT : 2019LYSE1282⟩. ⟨tel-02499621⟩



Record views


Files downloads