Individuals with ASD and Co-Occurring Anxiety Show Increased Amygdala and Orbitofrontal Cortex Activity During Face Perception

Background:  

The nature of amygdala deficits in autism spectrum disorders has recently been the subject of considerable debate.  Recent data have shown both hyper- and hypo-activation of amygdala in response to social and emotional stimuli in ASD.  Almost no studies to date have examined whether these discrepant findings can be understood in terms of anxiety symptoms among individuals with ASD.  This gap in the ASD literature is surprising and problematic, as decades of research in typically developing populations have associated anxiety with abnormal amygdala function, and recent studies have documented very high levels of comorbid anxiety symptoms in ASD.

Objectives:  

Preliminary data are presented for an ongoing study testing the hypothesis that individuals with ASD and co-occurring anxiety symptoms would show increased activation in key emotion structures (namely amygdala) during the perception of negatively valenced stimuli (angry faces).

Methods:  

Individuals with ASD (N=11) and a control sample (N=8) completed a probe detection task that has been used widely among typically developing adults and children with anxiety disorders, but seldom in ASD.  Participants saw two side-by-side faces immediately followed by a probe appearing at the same location as one of the two faces.  Face pairs consisted either of two neutral faces, or one neutral and one angry face.  Participants were asked to indicate via button-press whether the probe appeared on the left or right side of the display.  Individuals with anxiety are typically faster to identify the probe when it is preceded by emotional rather than a neutral face.  Anxiety symptoms were measured via the parent report version of the SCARED – a DSM-IV-based inventory of anxiety symptoms in children.

Results:  

fMRI analysis examined the interaction between group (ASD and control) and stimulus valence, i.e trials with angry versus neutral faces.  The ASD group showed significantly greater activation in bilateral amygdala and orbitofrontal cortex when presented with an angry face.  Results for completed SCARED reports indicated that ASD participants (N=8) presented with significantly more anxiety, t(13) = 2.179, p < 0.05, and generalized anxiety disorder, t(13) = 2.278, p < 0.05 symptoms than control participants (N=7).  Post hoc analyses indicated that anxiety scores on the SCARED were significantly correlated with amygdala activation for angry faces among individuals in the ASD group.

Conclusions:  

These preliminary data support the hypothesis that the profile of amygdala function in ASD depends on the presence or absence of anxiety symptoms.  Individuals with ASD displayed significantly higher activation in two key emotion and emotion regulation structures – amygdala and orbitofrontal cortex.  This finding suggests that the existence of comorbid anxiety may help explain discrepant findings on amygdala function in ASD.  Further, the finding of increased activation in orbitofrontal cortex underscores the putative role of abnormal emotion regulation processes in the clinical picture of anxiety in ASD.