A new neuroimaging study provides evidence of increased amygdala and face cortical network activation in individuals with autism spectrum disorder (ASD) in response to face-like (or pareidolic) stimuli. These findings support the hypothesis of an overly connected subcortical face-processing network in ASD, potentially resulting from an early imbalance between excitatory and inhibitory systems. The study was recently published in Cortex.
Autism is a neurodevelopmental condition characterized by atypical development of social communication, restricted and repetitive behaviors and interests, and sensory alterations. Individuals with autism may have difficulties with social interaction, communication, and exhibit repetitive patterns of behavior. Autism is considered a spectrum disorder, meaning that it varies widely in its presentation and severity among individuals.
The researchers were interested in studying the neural mechanisms underlying face perception in individuals with autism because there is evidence of an over-sensitivity of the subcortical system, particularly the amygdala, during face processing in autism. Previous research has shown increased amygdala activation in response to facial expressions of fear in autistic individuals compared to non-autistic controls.
The researchers proposed that this over-arousal to face stimuli in autism is a result of an over-connection between the amygdala and the rest of the face-processing system, which may be caused by an early imbalance of excitatory and inhibitory systems in the brain.
“I have been interested in face perception in autism for more than 20 years,” said study author Nouchine Hadjikhani, an associate professor at Harvard Medical School and director of Neurolimbic Research at the Martinos Center for Biomedical Imaging. “In 2004, I published a paper that demonstrated that, contrary to what had been said by others, autistic individual actually demonstrated activation similar to that of a control group in the fusiform face area (FFA), a occipitotemporal brain area specialized in face processing – as long as one made sure that people were actually looking at the faces.”
“This led me to try to understand face and gaze avoidance in autism, and I have published a number of papers around this topic. The problem is that when one uses ‘real’ faces, there is always a social component, together with the presence of eyes, that may confound the results if one thinks that there is something more fundamental to the gaze avoidance issue. Hence the idea of using pareidolic objects, that we had shown to activate the FFA with the same timing as real faces, back in 2009, using magnetoencephalography (MEG).”
“So the idea here was to test the hypothesis that there is a hyperconnectivity in autism within the subcortical system that is dedicated to face detection using stimuli that did not have eyes, or social content,” Hadjikhani explained.
The researchers conducted their study using functional magnetic resonance imaging (fMRI) to examine brain activation patterns in response to visual stimuli. They recruited a total of 71 participants, including individuals diagnosed with autism spectrum disorder (ASD) and non-autistic controls (CON).
The visual stimuli consisted of 42 face-like objects and 42 matched control objects without pareidolic properties. The stimuli were presented on a screen during the fMRI scan, and participants viewed them through a mirror attached to the coil. The stimuli were grayscale images presented for a duration of 1600ms, followed by a blank interval of 400ms. Participants performed a one-back task to indicate stimulus repeats and maintain attention.
The researchers aimed to compare the brain activation patterns in response to pareidolic faces and control objects, focusing on the subcortical system, particularly the amygdala. They hypothesized that individuals with ASD, who have been shown to exhibit over-sensitivity in the subcortical system during face perception, would also demonstrate heightened activation in response to pareidolic faces.
The researchers observed a significant increase in amygdala activation in individuals with ASD compared to controls when exposed to face-like stimuli. The extended face-processing network, which includes the primary visual cortex, the orbitofrontal cortex (OFC), the superior temporal sulcus (STS), and the temporal pole, also showed higher activation in individuals with ASD compared to controls when processing face-like stimuli.
“I was surprised to see how big of a difference there was between groups, I would have expected a bit more activation in the control group, given our previous MEG results,” Hadjikhani told PsyPost. “But on the other hand, fMRI does not have a very high temporal resolution, so it is possible that this is the reason why we did not see much in the controls, as it may have been a short-lived activation in that group.”
The researchers propose that the increased sensitivity of the amygdala in ASD for pareidolic objects may be evidence of over-connection between the amygdala and the rest of the face-processing system. This over-connection could result from an early imbalance between excitatory and inhibitory systems in autism.
The findings indicate “that the basis of eye contact avoidance in autism is the result of an abnormal sensitivity of the threat processing system,” Hadjikhani said. “From early on after birth, the subcortical face processing system is too sensitive to basic face configuration and gets over-connected. Since this system over time gets involved in gaze perception, this is experienced as a stressful thing.”
Understanding the neural mechanisms underlying face perception in autism is important because it can shed light on the underlying neural processes that contribute to the social communication difficulties experienced by individuals with autism. However, the study also had some limitations. The participants were approximately 20 years old on average. But ASD typically develops early in childhood, with symptoms often becoming apparent in the first two years of life.
“As long as we cannot show this in infants, all this remains speculative, but I think there is accumulating evidence to show that eye contact avoidance in autism results from a hypersensitive subcortical face processing system,” Hadjikhani said.
“Given these observations, and also given first person reports from autistic individuals, it is important to remember that faces, and in particular eye contact can be overwhelming for autistic people, and that forcing them to engage in it may not be such a great idea. However, being aware of it and trying to habituate them may help them having a better read on others.”
The study, “Overwhelmed by the man in the moon? Pareidolic objects provoke increased amygdala activation in autism“, was authored by Nouchine Hadjikhani and Jakob Asberg Johnels.
