Coactivation pattern analysis reveals altered salience network dynamics in children with autism spectrum disorder

Abstract

Brain connectivity studies of autism spectrum disorder (ASD) have historically relied on static measures of functional connectivity. Recent work has focused on identifying transient configurations of brain activity, yet several open questions remain regarding the nature of specific brain network dynamics in ASD. We used a dynamic co-activation pattern (CAP) approach to investigate the salience/midcingulo-insular (M-CIN) network, a locus of dysfunction in ASD, in a large multi-site resting-state fMRI dataset collected from 172 children (ages 6-13 years; n = 75 ASD; n = 138 male). Following brain parcellation using independent component analysis, dynamic CAP analyses were conducted and k-means clustering was used to determine transient activation patterns of the M-CIN. The frequency of occurrence of different dynamic CAP brain states was then compared between children with ASD and typically developing (TD) children. Dynamic brain configurations characterized by co-activation of the M-CIN with central executive/lateral frontoparietal and default mode/medial frontoparietal networks appeared less frequently in children with ASD compared with TD children. This study highlights the utility of time-varying approaches for studying altered M-CIN function in prevalent neurodevelopmental disorders. We speculate that altered M-CIN dynamics in ASD may underlie the inflexible behaviors commonly observed in children with the disorder.

Distinct profiles of spatio-temporal brain dynamics along symptoms dimensions in autism

“…The altered fluctuations observed in the current study could be a sign of failing its function of switching between central executive and DM networks. Other studies in dynamic functional connectivity support this hypothesis by highlighting less frequent transitions between salience-central executive/lateral frontoparietal and DM/medial frontoparietal in children with ASD (Marshall et al 2020). These results encourage the idea of disruptions in brain mechanisms involved in cognitive flexibility in individuals with ASD.…”

“…the anterior cingulate cortex (Guo et al 2020)], hubs of the DM network [prefrontal and posterior cingulate cortices], the thalamus or some sensorimotor regions (Fu et al 2019; Guo et al 2020). The connectivity states were also characterized by less frequent transitions between salience-central executive and DM-medial frontoparietal in children with ASD (Marshall et al 2020) and by fewer transitions to an unstable states in adults with ASD, negatively correlated with the gold standard Autism Diagnosis Observation Schedule ADOS score (Watanabe and Rees 2017). Another study from Harlalka and al.…”

Altered patterns of brain dynamics linked with body mass index in youth with autism

“…The denoised data were then subjected to a group ICA using the GIFT toolbox (https://trendscenter.org/software/gift/). A high model order of 100 independent components (ICs) was used to parse the brain into functional networks as in previous work (Allen et al, 2014; Marshall et al, 2020). The infomax group ICA algorithm (Calhoun et al, 2001) was repeated 20 times using ICASSO (http://www.cis.hut.fi/projects/ica/icasso).…”

“…Using a squared Euclidean distance to cluster the patterns, a CAP analysis was conducted using the optimal value k = 5 on the group matrix. ASD and TD participants were included into one group for the CAP analysis as our previous work has shown there are no group differences between CAPs when clustered separately, nor systematic relationships between head motion and brain dynamics in each CAP (Kupis et al, 2020; Marshall et al, 2020).…”

“…In the case of ASD, dynamic methods have revealed novel insights into brain inflexibility, providing evidence for potential neural underpinnings of the inflexible behaviors commonly seen in the disorder (Uddin, 2020). Recent studies characterizing CAPs have found aberrations in how frequently the M‐CIN occurs during intrinsic states in children with ASD compared with TD children (Marshall et al, 2020), and evidence for altered network integration by the examination of dwell time (DT) and frequency of occurrence, metrics used to capture brain dynamics, among the M‐CIN, M‐FPN, and L‐FPN in children with ASD compared with TD children during task‐evoked states (Kaiser et al, 2019; Kupis et al, 2020). Other dynamic studies have discovered alterations in the number of transitions, another metric used to study dynamic relationships among CAPs, between brain states and aberrant between‐network integration in ASD (Uddin, 2020).…”