Resting State Functional Connectivity Alterations in Individuals with Autism Spectrum Disorders: A Systematic Review

Many studies investigate the alterations in resting state functional connectivity in autism. Most of these studies focus on different regions of the brain to find the connectivity differences between autism spectrum disorder and typically developing populations. The present review quantitatively extracts this data from all the seed based studies on autism spectrum disorder and uses it to build, for the first time, an Autism Altered Functional Connectome (AAFC) which summarizes the alterations in functional connectivity consistently reported in the literature. The data extracted from all the studies matching the inclusion criteria are presented at one place in human as well as a machine-readable format for further interpretation and analysis. Systematically reviewing 41 publications on 2818 subjects comprising 1459 typically developing and 1359 subjects with autism spectrum disorder, a total of 932 altered functional connectivity links were employed to construct an AAFC. The AAL atlas mapping of these links resulted in 71 replicated links of which 49 were consistent, and 574 unreplicated links that were reported just once. Out of 49, 38 were replicated across different non overlapping datasets. Majority of the studies analyzed the functional connectivity of the Default Mode Network (DMN) and its regions. Two important DMN regions, namely precuneus and posterior cingulate cortex were reported to exhibit different connectivity profiles with former majorly underconnected and later majorly overconnected consistently reported across multiple studies. After mapping the AAFLs to an atlas of brain networks, poor integration within DMN regions, and poor segregation of DMN regions with extra-DMN regions was observed.