We’re closer to understanding how autistic brains process faces differently, thanks to artificial intelligence

New research could help us understand how the brains of people with autism have a harder time recognizing emotions in facial expressions.

Image via Pixabay.

Facial expressions are one of the most important ways people convey their emotions to those around them. Smiles are a good indicator of happiness; rolling eyes are a pretty reliable sign that someone is getting frustrated. People with autism, however, may have trouble picking up these screens.

We don’t really know why. New research focused on artificial intelligence could finally help us find out why.

Internal workings of the brain

As far as we know, there are two brain areas that may explain where the processing differences between typical and autistic brains lie. One of them is the inferior temporal cortex (IT), which handles facial recognition. Another is the amygdala, which collects information from the computational cortex and interprets the emotional content of the expressions it perceives.

In order to understand to what extent these two domains are involved in the differences in processing, Kohitij Kar, a researcher in the laboratory of Professor James DiCarlo at MIT, drew on previous research. One of the studies he investigated involved showing images of faces to autistic adults and neurotypical controls. These images were generated by software that gave them different levels of happiness or fear; participants were asked to judge whether each face expressed happiness. Compared to controls, adults with autism needed higher levels of happiness in faces to perceive it correctly.

The other study he relied on involved recording neural activity in the tonsils of people undergoing surgery for epilepsy, while they performed the facial task. This article reported that a patient’s neural activity could be used to predict their judgment of each face.

For the study itself, Kar created an artificial neural network, a computer system that mimics the architecture of our brains and is organized into multiple computational layers. He trained him to perform the same tasks. The behavior of the network on the emotion recognition task was very similar to that of neurotypical controls. Then Kar set about dissecting it to understand how it did its job and to find clues as to why adults with autism interpret emotion in facial expressions differently than neurotypical individuals.

First, he reports that network responses might most closely resemble those of autistic participants when his output was based on the last layer of the network. This layer most closely mimics the computational cortex and sits at the end of the visual processing pipeline in primates, he explains, citing previous research.

Second, Kar looked at the role of the amygdala. Working with previously recorded data and factoring it into the output of his network, in which the effect of the computational cortex had already been quantified. This showed that the amygdala has a very small effect on its own. Together, these two results indicate that the computational cortex is strongly implicated in the differences between neurotypical controls and autistic adults.

He further explains that his network could help select images that would be more effective in diagnosing autism.

“These are promising results,” says Kar. Better methods will surely be developed “but often in the clinic we don’t need to wait for the absolute best product”.

To validate the results, he trained separate neural networks to match the choices of neurotypical controls and autistic adults. For each, he quantified the strength of the connections between the final layers and the decision nodes; those of the “autistic network” were weaker than those of the network corresponding to neurotypical responses. This, he explains, indicates that the neural connections that interpret sensory data are “noisier” in autistic adults.

Such a view was further reinforced by Kar’s addition of various levels of fluctuation (“noise”) in the functioning of the last layer of the network modeling autistic adults. Within a certain range, this additional noise significantly increased the proximity of network responses to those of autistic adults. Adding him to the control network had a much weaker effect in aligning his responses with those of neurotypical adults.

Although based on how computers work, the findings strongly point us toward answers regarding the differences between data processing in neurotypical and autistic brains.

The article “A Computational Probe into Behavioral and Neural Markers of Atypical Facial Emotion Processing in Autism” was published in The Journal of Neuroscience.

#closer #understanding #autistic #brains #process #faces #differently #artificial #intelligence

Leave a Comment

Your email address will not be published. Required fields are marked *