Aberrant dynamics of cognitive control and motor circuits predict distinct restricted and repetitive behaviors in children with autism

Abstract

Restricted and repetitive behaviors (RRBs) are a defining clinical feature of autism spectrum disorders (ASD). RRBs are highly heterogeneous with variable expression of circumscribed interests (CI), insistence of sameness (IS) and repetitive motor actions (RM), which are major impediments to effective functioning in individuals with ASD; yet, the neurobiological basis of CI, IS and RM is unknown. Here we evaluate a unified functional brain circuit model of RRBs and test the hypothesis that CI and IS are associated with aberrant cognitive control circuit dynamics, whereas RM is associated with aberrant motor circuit dynamics. Using task-free fMRI data from 96 children, we first demonstrate that time-varying cross-network interactions in cognitive control circuit are significantly reduced and inflexible in children with ASD, and predict CI and IS symptoms, but not RM symptoms. Furthermore, we show that time-varying cross-network interactions in motor circuit are significantly greater in children with ASD, and predict RM symptoms, but not CI or IS symptoms. We confirmed these results using cross-validation analyses. Moreover, we show that brain-clinical symptom relations are not detected with time-averaged functional connectivity analysis. Our findings provide neurobiological support for the validity of RRB subtypes and identify dissociable brain circuit dynamics as a candidate biomarker for a key clinical feature of ASD.

Fig. 1. Overall approach to determine the temporal dynamics of cognitive control and motor circuits and its relationship with Restricted and Repetitive Behaviors (RRBs).

a Cognitive control circuit-based model of circumscribed interests (CI) and insistence on sameness (IS) symptoms. The model proposes that aberrant functional of organization of key fronto-parietal-opercular cognitive control circuit may contribute to the cognitive components of RRBs i.e., CI and IS symptoms in children with ASD. Specifically, this model posits a key role for the salience network (SN) in aberrant mapping of internal and external salient events leading to altered dynamic temporal interactions with the central executive network (CEN), and the default mode network (DMN), resulting in CI and IS. b Overall analysis pipeline for examining dynamic time-varying cross-network interactions within cognitive control circuit and their relationship with CI and IS symptoms. Time-varying cross-network interaction was measured using a dynamic functional connectivity approach. (1) We estimated dynamic functional interactions between SN, CEN, and DMN using a sliding-window approach. (2) To identify distinct group-specific states associated with dynamic functional connectivity, we applied a group-wise k-means clustering on the time-series of correlation matrices in each group separately. (3) Brain state-specific network interaction index (NII) was used to characterize cross-network interaction in each dynamic brain state. NII for each state k by averaging NII across sliding-windows labeled as state k. Cognitive NII (CNII) of a sliding window was computed as the difference in correlation between SN and CEN time series and correlation between SN and DMN. The correlation values were extracted from the covariance matrix associated with that sliding window. Mean and variability of time-varying CNII was calculated as average and standard deviation of CNII values across dynamics brain states respectively. (4) Linear regression analysis was used to examine the relation between dynamic time-varying cross-network interactions measure, including mean and variability of time-varying NII, and ADI-R derived RRB subtype symptom severity scores. c Motor circuit-based model of repetitive motor behavior (RM) symptoms. The model proposes that aberrant functional of organization of key cortical-subcortical motor circuit may contribute to the motoric components of RRBs i.e., RM symptoms in children with ASD. Specifically, this model posits a key role for altered dynamic temporal interactions between the cortical motor network (cMN), and the subcortical motor network (sMN), resulting in RM. d Overall analysis pipeline for examining dynamic time-varying cross-network interactions within motor circuit and their relationship with RM symptoms.

Fig. 2. Aberrant temporal dynamics of cognitive control circuit in children with ASD.

a Children with ASD showed four states and TD children showed two states. Color codes distinct states in each participant. States 1, 2, 3, and 4 in children with ASD are represented by dark orange, dark cyan, dark green, and dark magenta, respectively; states 1 and 2 in TD children are represented by lime green and dark blue, respectively. b Cognitive NII (CNII) of dynamic brain states showed intermittently reduced cross-network interaction in children with ASD compared to TD children. c The temporal mean of dynamic cross-network interaction in the cognitive control circuit, assessed using mean of dynamic CNIIs across states, was significantly lower in children with ASD, compared to TD children. The temporal variability of dynamic cross-network interaction in the cognitive control circuit, assessed using standard deviation of dynamic CNIIs across states, was significantly higher in children with ASD, compared to TD children. Two-sided two sample t-tests were used to compare mean and temporal variability of CNIIs between children with ASD and TD children. Error bar shows standard error of mean. ***p < 0.001.

Fig. 3. Aberrant temporal dynamics of motor circuit in children with ASD.

a Children with ASD and TD children showed two states. Color codes distinct states in each participant. States 1 and 2 in children with ASD are represented by dark orange and dark cyan, respectively; states 1 and 2 in TD children are represented by moderate violet and bright blue, respectively. b Motor NII (MNII) of dynamic brain states showed intermittently increased cross-network interaction in children with ASD compared to TD children. c The temporal mean of dynamic cross-network interaction in the motor circuit, assessed using mean of dynamic MNIIs across states, was significantly higher in children with ASD, compared to TD children. The temporal variability of dynamic cross-network interaction in the motor circuit, assessed using standard deviation of dynamic MNIIs across states, did not differ between children with ASD and TD children. Two-sided two sample t-tests were used to compare mean and temporal variability of MNIIs between children with ASD and TD children. Error bar shows standard error of mean. *p < 0.05.

Fig. 4. Aberrant brain circuit dynamics in children with ASD predict RRB subtypes severity.

Regression analysis revealed that temporal mean and variability of dynamic cross-network interactions in the cognitive control circuit predicted a CI and b IS symptoms, but not RM symptoms. c Regression analysis revealed that temporal mean of dynamic cross-network interactions in the motor circuit predicted RM symptoms, but not CI or IS symptoms. Error band represent 95% confidence interval for the regression estimate. Cross-validation analyses confirmed these results.