Functional Brain Networks in Autism Spectrum Disorder in Different Attentional States

Background:  

Interoception is our sensitivity to stimuli originating inside of the body. As humans, we perceive feelings from our bodies that relate our state of well-being, our energy and stress levels, our mood and disposition, all of which seem to be impaired in ASD. Anatomical and functional brain studies have converged to the hypothesis that ASD is associated with atypical connectivity, producing a system that is ineffective for integrating complex information at the neural and cognitive level. For instance, task positive (TPN) and task negative (TNN) functional brain networks of ASD and normal subjects have qualitatively different intrinsic organization, with abnormal connectivity in the TNN but not in the TPN in ASD. It remains however unexplored if the TPN differentiates whether the task involves directing attention internally (interoception) or externally.

Objectives:  

We reasoned that brain network organization during interoceptive states may show more pronounced differences between groups and hence constitute a precise physiological signature of ASD. To examine this hypothesis we measured functional connectivity in three different mental states varying the focus of attention of the subjects.

Methods:  

Participants included 12 individuals with high-functioning autism or Asperger’s Syndrome matched to a group of 12 typically developing individuals. Whole brain fMRI data were acquired with a GE HDx 3T scanner. Rest run : subjects lie eyes closed in the scanner. Introspective run : subjects had to count the number of their breathing cycles. Exogenous Run : subjects had to detect and count beeps (target) with a little higher pitch among others. Classic functional connectivity and network characterization using graph theory metrics were performed. Using graph theory metrics as characteristics of ASD and Control groups, we further performed a classification analysis, based on ROC curves.

Results:  

Network changes between groups in the interoceptive and exteroceptive states showed opposite effects, revealing that inferences about connectivity in ASD are state-dependent. ASD functional networks largely vary across conditions: in the exogenous run, the analysis of ASD brain network reveals sub-optimal metrics, suggesting that it is badly suited for this kind of task. As attention shifts to self, ASD brain networks improve their metrics – even surpassing those of Controls’- suggesting that ASD networks reaches its optimum capabilities in those cognitive states related to introspective tasks and body perception. While network measures decode whether a patient belongs to the ASD or normal group at a modest performance of 60%, comparing how network parameters change with state achieves very accurate decoding performance (90%).  

Conclusions:  

Connectivity analysis revealed consistent and organized differences in functional connectivity between ASD and control groups, suggesting that a distinct pattern of dynamical connectivity may be related to the physiopathology of ASD. ASD process interoceptive information just as they process external information, without adapting information processing to the fact that the signal comes from their own body, possibly caused by an altered capacity to self awareness. This results may have important behavioural consequences : ASD individuals may exaggeratedly focus on internal physical sensations that tend to automatically elicit anxiety and panic, leading to inappropriate emotional reactions.

See more of: Brain Imaging-Functional
See more of: Brain Imaging: fMRI-Social Cognition and Emotion Perception
See more of: Brain Structure & Function

| More