�鶹�ƽ��

“This is an important step toward understanding the neural mechanisms underlying communication in autism,” UVA neuroscientist , a coauthor of the study, said. “If we can identify reliable biological markers, they could eventually help researchers evaluate interventions more objectively and understand why communication abilities differ so widely across the autism spectrum.”

Researchers have long known that many autistic individuals experience challenges with language and communication, but the underlying brain mechanisms have remained difficult to measure. Most clinical assessments rely on behavioral observations, rather than biological indicators.

To investigate those mechanisms, the research team recorded brain activity from 306 participants aged 7 to 18, including 162 youths with autism and 144 typically developing peers. Participants wore high-density electroencephalography, or EEG, caps equipped with 128 sensors while listening to streams of spoken nonsense words designed to measure how the brain processes speech.

Jack Van Horn, left, and Kevin Pelphrey, professors at the University of Virginia, combine data science and neuroscience to reveal brain activity patterns that were previously difficult to detect. (Contributed photo)

Rather than focusing solely on traditional brain wave patterns, the researchers examined a newer measure of overall neural activity, known as the brain’s “aperiodic” signal. The signal reflects the balance between excitation and inhibition, two fundamental processes that help the brain distinguish meaningful information from background activity. 

The study found that autistic participants showed altered patterns in these signals, consistent with increased neural “noise,” suggesting the brain may process speech less efficiently.

More importantly, youths whose brain activity appeared noisier also tended to score lower on measures of everyday verbal communication. Those same brain signals were not associated with traditional language skills, such as vocabulary or grammar.

The researchers caution that the findings do not represent a diagnostic test for autism. Instead, they point to a promising biological marker that could eventually help researchers monitor changes in communication abilities over time, or measure whether therapies are affecting underlying brain function.

The work also highlights the growing role of advanced data science techniques in neuroscience, allowing researchers to uncover subtle patterns in complex brain data that were previously difficult to detect.

“The human brain generates an incredible amount of data every second,” said , a coauthor and professor in �鶹�ƽ�� School of Data Science and College of Arts & Sciences. “The challenge isn’t collecting it anymore; it’s making sense of it. Advances in computational analysis are allowing us to separate meaningful signals from background activity in ways that weren’t possible just a few years ago.”

Although the study included one of the largest EEG datasets of its kind, researchers say additional work is needed before the findings could influence clinical care. Most participants had average or above-average verbal abilities, and future studies will need to determine whether the results extend to minimally verbal individuals with autism. 

The authors also note that EEG provides an indirect measure of brain activity and should ultimately be combined with other imaging techniques to better understand the underlying biology.

Still, the findings move scientists closer to a longstanding goal in autism research: developing objective biological measures that complement behavioral evaluations.