Default Mode Network - Insight into Disruptive Behaviour in Autism Spectrum Disorder

Background: One third of children with autism spectrum disorder (ASD) exhibit comorbid disruptive-behaviour, anxiety, and attentional-deficit problems. Disruptive-behaviour in ASD may be linked to poor self-regulation, and the neural correlates have not yet been established. Moreover, poor self-regulation may compromise social functioning. The Default Mode Network (DMN) is a set of brain regions typically deactivated during cognitive tasks and may be related to self-regulation. DMN may provide a neural window to study disruptive behaviour and social functioning in children with ASD, and may help clarify how disruptive behaviours are linked with comorbid anxiety or attentional-deficit-problems at the brain level.

Objectives: To examine how disruptive behaviour in children with ASD is related DMN deactivation, to determine how much this relationship is mediated by comorbid anxiety and/or attentional deficit problems, and to explore the link between DMN deactivation and social functioning.

Methods: Participants included 31 children (all male) with ASD between the ages of 4 and 18 years. Age appropriate versions of Child/Adolescent Symptom Inventory-4 (CSI-4/ASI-4) was used to assess comorbid symptoms of anxiety, attentional deficit, and disruptive behaviour (Cronbach’s α= .75 to .91). Social Responsiveness Scale (SRS) was used to measure social functioning. All participants underwent functional magnetic resonance imaging (fMRI) whilst passively viewing alternating clips of biological (BIO) and scrambled point-light displays (12 blocks in total, 24 s each block) in a 3-Tesla scanner. The experiment began and ended with 20-s fixation interval, which were combined to create fixation (FIX) condition for all contrast analyses. A whole-brain group analysis was conducted using mixed-effects modelling (FSL’s FLAME1+2), voxel-level thresholding Z > 1.96, and cluster-level thresholding p < .05, while controlling for age.

Results: At the group level and as expected, DMN deactivations in medial prefrontal cortex (MPFC), posterior cingulate cortex (PCC), and precuneus (PC) were observed in BIO<FIX contrast (Fig 1-top). DMN deactivation was negatively correlated with disruptive behaviour in MPFC and PCC/PC (Fig 1-bottom), suggesting that disruptive behaviours reduced the DMN deactivation in these regions. Furthermore, anxiety partially mediated the disruptive behaviours’ effect of reduced DMN deactivation in a region within MPFC (Fig 2). There was no evidence supporting attention deficit problems mediated. DMN deactivation was also negatively correlated with SRS total raw score.

Conclusions: This study demonstrated that DMN deactivation provides a useful window to study the role of self-regulation in children with ASD. Reduced DMN deactivation in children with ASD who displayed disruptive behaviour suggests poorer self-regulation when engaged in a social perception task. The effect within MPFC is partially mediated by anxiety, suggesting interventions aimed to reduce disruptive behaviour via improving self-regulation may be more successful by simultaneously targeting comorbid anxiety in children with ASD. Reduced DMN deactivation is also linked to greater social impairment. To our knowledge, this study is the first to provide a neuromarker for disruptive behaviour in ASD, which describes a key brain-based dimension that may be used to facilitate the development of tailored treatment to reduce disruptive behaviour in ASD in the future.