Face processing in autistic children: what an EEG study found

In a case–control EEG study of 399 children, published in 2026 in Nature Mental Health, autistic children showed less distinct face- and identity-selective neural representations than neurotypical children, and those signals did not become more distinct with age the way they did in neurotypical children. If you are a parent looking for a therapist, or a clinician who works with autistic children, this is early research and not a guide to care, so use it to sharpen your questions at a first session. We checked both sources on November 8, 2026. They show a group-level difference in brain signals in children. They do not show how any one child will develop, and they do not test or recommend any therapy.

Key takeaways
- In a study of 399 children, autistic children showed less distinct face-related neural representations than neurotypical children.
- Across ages 6–7, 8–9 and 10–11, neurotypical children's face signals became more distinct with age; this was not observed in autistic children.
- The researchers say the findings are not yet ready for clinical use.
- Neither source says anything about therapist fit, or about any therapy improving face recognition or emotion reading.

What did the study actually measure?

It measured brain activity with EEG while children looked at faces, and compared autistic and neurotypical children. The abstract reports that autistic children showed less distinct face- and identity-selective neural representations than neurotypical children.

The study also compared three age groups: 6–7, 8–9 and 10–11. Neurotypical children's face-related representations became more distinct across those groups. According to the abstract, that pattern was not observed in autistic children.

One limit matters here. The Nature Mental Health page is paywalled, so we saw only the abstract. We did not see the methods or effect sizes beyond it.

What does "earlier than thought" mean here?

It refers to timing within a single viewing of a face, not to a child's age. A Yale School of Medicine summary, posted on ScienceDaily on 27 October 2026, says the researchers analyzed brain activity over a wider period. They found evidence that differences in face processing begin before the familiar N170 signal, which appears about 170 ms after a person sees a face.

So the differences appear at earlier stages of visual processing after a face is seen. The page does not say the differences begin earlier in a child's life.

Does this explain why some autistic people find faces hard to read?

Possibly in part, and the researchers hedge it. The summary says the discovery "could help explain" why some autistic people have difficulty recognizing faces and reading emotions. That is a possible partial explanation, not an established cause.

The study is also a cross-sectional comparison of age groups. It did not follow the same children over time. It tells us about groups at different ages, not about the path any individual child will take.

Does every autistic child need help with this?

No. The researchers stress that many neurodivergent people experience little or no impairment related to their autism and may neither need nor want additional support. Autism does not necessarily create difficulties that call for intervention.

If you are writing or reading about this study, keep that in view. A group-level difference in brain signals is not a reason to assume a particular child needs face-reading or emotion support.

What does this mean when you look for a therapist?

Not much directly, because the sources are silent on therapy. Neither page says anything about therapist fit, therapist experience, what "autism-informed" means, or how to verify it. Neither tests or recommends any therapy, and neither shows that any approach improves face recognition or emotion reading.

One point from the summary is usable. It notes that some clinicians historically required autistic individuals to maintain eye contact, and says research has since shown that forcing eye contact is not an effective solution. That makes a fair question for a first session: how does the therapist think about eye contact?

The questions below are our own framing, not findings from the study:

• What experience do you have working with autistic children?
• How do you decide what a child wants or needs support with?
• How do you think about eye contact in sessions?

For any therapist, take experience from the listing itself rather than assuming it. A directory can show you what a clinician chose to list and what a dated license receipt says. It cannot tell you from a study like this who is a good fit.

What does a clinician take from this?

The authors call the work a first step toward introducing a new biomarker, and say more research is needed before the findings can be used in clinical practice. For now, it gives clinicians a new way to think about how the autistic brain processes faces and how those responses change throughout childhood. It does not change what you can responsibly promise a family.

If you want to be found by families asking the questions above, a clear listing that describes your actual experience is what a parent will read.

Takeaway

In a study of 399 children, face-related brain signals became more distinct with age in neurotypical children and did not in autistic children. The researchers call it a first step, and they say it is not ready for clinical use. Use it to ask better questions, not to assume anything about a particular child.

If you are looking for a therapist, start at https://couchcommons.com/find. If you are a clinician who wants to be found, start at https://couchcommons.com/join.

Frequently asked questions

Does this study show what causes difficulty reading faces?

No. The researchers say the finding could help explain why some autistic people have difficulty recognizing faces and reading emotions. That is a hedged, possible partial explanation, not an established cause.

Does every autistic child need support with faces or emotions?

No. The researchers stress that many neurodivergent people experience little or no impairment related to their autism and may neither need nor want additional support.

Does the study say which therapy helps?

No. Neither source tests or recommends any therapy, and neither shows that any approach improves face recognition or emotion reading.

Does forcing eye contact work?

The ScienceDaily summary says some clinicians historically required eye contact, and that research has since shown forcing it is not an effective solution.