Altered Visual Processing a Nonverbal Marker
of Autism?

Megan Brooks

August 28, 2019

Patients with autism may have altered visual processing, which could serve as an objective, nonverbal neural marker of the disorder, new research suggests.

Using electroencephalography (EEG), investigators found that adults with autism have differences in the part of the brain that processes visual information — and these differences can be used to predict which individuals have autism.

Caroline Robertson, PhD

"This signature of autism in vision seems to be a good objective neural marker of the condition in adults," coinvestigator Caroline Robertson, PhD, assistant professor of psychological and brain sciences at Dartmouth College, Hanover, New Hampshire, and director of the Dartmouth Autism Research Initiative, told Medscape Medical News

"Importantly, it does not require any verbal engagement from the individual, making it well-suited for future research with young preverbal kids and also nonverbal adults," Robertson said.

The findings were published online August 15 in Current Biology.

Altered "Binocular Rivalry"

"For a long time, we thought that autism was fundamentally about changes in regions of the brain that process social information. Sensory symptoms have only recently been recognized as core to autism," Robertson explained.

In earlier work, the investigators observed slower behaviorally reported rates of a basic visual phenomenon known as "binocular rivalry" in patients with autism.

During binocular rivalry, two images — one presented to each eye — vie for awareness, alternating back and forth in perception. This competition relies in part on the balance of excitation and inhibition in the visual cortex, which may be altered in autism. But until now, there was no direct neural evidence for this potential marker of excitation/inhibition (E/I) balance in autism, the researchers write.

Their latest study shows a "striking" alteration in the neural dynamics of binocular rivalry in individuals with autism, they add.

The study included 18 adults with autism and 19 age- and IQ-matched individuals to act as healthy controls. The investigators measured steady-state visually evoked potentials (VEP) via a single EEG electrode placed on a participant's head over the visual region of the brain.

They were then presented with one of two visual images on a computer screen: red checkerboards in the left eye and green checkerboards in the right eye that flickered back and forth at different rates.

Not Ready for Prime Time

The researchers replicated their prior findings of slower binocular rivalry and reduced perceptual suppression in the patients with autism compared with controls. The results also provide "direct neural evidence" for slower rivalry in autism compared with the control group, they write.

Using only the neural data from the visual test, they were able to predict autism symptom severity and correctly classify individuals' diagnostic status (autistic vs control) with 87% accuracy.

"These findings clearly implicate atypical visual processing in the neurobiology of autism," the researchers write.

However, Robertson cautioned that although the findings allowed them to predict which adults did vs did not have a diagnosis of autism, much work lies ahead.

"As a next step, we need to know if this difference in visual processing we identified in autism is specific to autism, as compared with other psychiatric conditions," she said.

To develop the measure as a screening tool for autism, it's also important to assess whether the results hold in younger, "ideally preverbal individuals," she added.

"In other words, can we predict which infants at risk for diagnoses of autism based on family history will go on to have a diagnosis based on early differences in visual processing? That's currently an open question, but we're hopeful," said Robertson.

Hard to Generalize

Commenting on the findings for Medscape Medical News, Alexander Kolevzon, MD, medical director at Seaver Autism Center for Research and Treatment, Icahn School of Medicine at Mount Sinai, New York City, called this is an interesting and well-done study.

"There is a lot of promise for using biomarkers through EEG to try to better understand the biology of autism and to better parse the phenotype," said Kolevzon, who was not involved with the research. "But as a diagnostic tool, I think it's very hard to generalize what we see in this small sample to autism as a whole."

He noted that this approach has to be tested in a separate patient population and in comparison with developmentally disabled controls, and in a much larger sample, in order to be confident in it as a diagnostic tool.

"The really interesting thing about using visually evoked potentials is that it does capture something about the underlying excitation/inhibition balance in autism, and we know that is an important contributor to at least some cases of autism. So being able to better understand that using a measure that is objective and easily obtained could be very valuable," he said.

In addition, VEP can be done very quickly, provides an immediate response, and can be used "in very low-functioning kids who don't follow instructions," said Kolevzon.

He noted that his personal belief about these kinds of tools is that they are going to be more informative in predicting or monitoring treatment response because it will give a direct reflection of excitatory and inhibitory activity in the brain.

It might also be possible to "stratify large groups of kids with autism on the basis of these kinds of VEP profiles in order to better select treatment," he said.

The fact that this group has shown a correlation between some of the changes on the VEP and the overall autism severity is also interesting, Kolevzon said.

He noted that "it would be even more interesting to dig a little deeper" into sensory sensitivities, particularly within the visual system.

"If responses on this particular paradigm were correlated with the degree to which patients experienced visual sensory reactivity symptoms," Kolevzon said, "then you've got kind of a potential target for treatment and a biomarker to measure it —and a clinical symptom that's really important to improve."

The study was supported by grants from the MIT-MGH Grand Challenge and the Simons Foundation Autism Research Initiative (SFARI). The investigators and Kolevzon have disclosed no relevant financial relationships.

Curr Biol. Published online August 15, 2019. Full text

For more Medscape Psychiatry news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: