Atypical Inter‐Network Deactivation Associated With the Posterior Default‐Mode Network in Autism Spectrum Disorder

Abstract

Previous studies have suggested that atypical deactivation of functional brain networks contributes to the complex cognitive and behavioral profile associated with autism spectrum disorder (ASD). However, these studies have not considered the temporal dynamics of deactivation mechanisms between the networks. In this study, we examined (a) mutual deactivation and (b) mutual activation‐deactivation (i.e., anticorrelated) time‐lag patterns between resting‐state networks (RSNs) in young adults with ASD (n = 20) and controls (n = 20) by applying the recently defined dynamic lag analysis (DLA) method, which measures time‐lag variations peak‐by‐peak between the networks. In order to achieve temporally accurate lag patterns, the brain imaging data was acquired with a fast functional magnetic resonance imaging (fMRI) sequence (TR = 100 ms). Group‐level independent component analysis was used to identify 16 RSNs for the DLA. We found altered mutual deactivation timings in ASD in (a) three of the deactivated and (b) two of the transiently anticorrelated (activated‐deactivated) RSN pairs, which survived the strict threshold for significance of surrogate data. Of the significant RSN pairs, 80% included the posterior default‐mode network (DMN). We propose that temporally altered deactivation mechanisms, including timings and directionality, between the posterior DMN and RSNs mediating processing of socially relevant information may contribute to the ASD phenotype. Lay Summary To understand autistic traits on a neural level, we examined temporal fluctuations in information flow between brain regions in young adults with autism spectrum disorder (ASD) and controls. We used a fast neuroimaging procedure to investigate deactivation mechanisms between brain regions. We found that timings and directionality of communication between certain brain regions were temporally altered in ASD, suggesting atypical deactivation mechanisms associated with the posterior default‐mode network.

Mesoscopic patterns of functional connectivity alterations in autism by contrast subgraphs

“…Increased coordination with nonessential regions may introduce low-level cross-talk and spoiling signal across primary network components (Belmonte et al, 2004). These results, there-fore, constitute a possible source of the widely observed decreased connectivity within the Default Mode Network (Lynch et al, 2013;Washington et al, 2014;Padmanabhan et al, 2017;Kotila et al, 2021), which may stem from the disturbing abnormal connectivity that one of its hubs (precuneus) has with different and unrelated regions (bilateral superior frontal gyrus).…”

Co‐activation pattern alterations in autism spectrum disorder–A volume‐wise hierarchical clustering fMRI study

“…Though FC and ICA RS metrics are not substantially affected by different TRs, faster imaging methods such as MREG with 10-20 Hz temporal resolution show "neural avalanches," which in traditional 0.5-1 Hz fMRI temporal resolutions are only seen as aliased images and could enable the study of higher cluster numbers and shorter CAPs (Huotari et al, 2019;Rajna et al, 2015), though the inherently slow hemodynamic response function may act as a limiting bottleneck (Bolton et al, 2020). Faster imaging and dynamic lag analysis (Kotila et al, 2020;Raatikainen et al, 2020) or causality analysis methods (Bernas et al, 2018;Bielczyk et al, 2019;Borchers et al, 2012;Chen et al, 2016;Deshpande & Hu, 2012;Kaminski et al, 2016;Li et al, 2020) may shed light on interactions between attention, visual, and other brain networks. MREG fMRI coupled with simultaneous EEG analysis (Hiltunen et al, 2014;Li et al, 2019;Ridley et al, 2017) could clarify the relationship between the neural avalanches and the brain's electrical activity in the future.…”