This week’s new autism research, simplified: a large US study built with autistic people shows how much results depend on who gets recruited; a meta-analysis of more than 15,000 children links sensory over-responsivity to autism traits and anxiety; and a genetics study suggests Phelan-McDermid syndrome is far more common than once thought. Each finding is promising, and each has clear limits.
Key takeaways
- A 2026 paper in the journal Autism describes 870 autistic US adults recruited through health systems, disability services and the community, and finds the three groups differed markedly.
- A 2026 meta-analysis in the Journal of the American Academy of Child and Adolescent Psychiatry of 15,728 children found sensory over-responsivity was most closely tied to autism traits and anxiety, not ADHD.
- A 2026 study in Autism Research of nearly 180,000 autistic people who had genetic testing estimates Phelan-McDermid syndrome affects about 1 in 7,300 people.
- Only the first study clearly involved autistic people as research partners; news coverage of the other two does not say.
New to this series? Research, Simplified explains recent peer-reviewed studies in plain language every Monday. Browse past pieces in our research section.
Why does it matter who takes part in autism research?
It matters because results from one group of autistic people may not apply to another. A 2026 paper in Autism shows this with unusual clarity.
The question
Can autism services research reach a sample that looks like the real autistic adult population, including people who are usually left out, such as those who find speaking difficult, people of colour and people who use disability services?
The method and sample
The study comes from AASPIRE (the Academic Autism Spectrum Partnership in Research and Education), led by Christina Nicolaidis. It uses community-based participatory research, an approach in which autistic people and academics work as equal partners from the design stage onward. The team recruited 870 autistic adults through three routes: health care systems, government disability services and the wider community, and surveyed them three times. More than 80% responded at both follow-ups.
A companion preprint from the same team, which has not yet been peer reviewed, describes a sample aged 18 to 77, roughly half female, nearly half LGBTQ+ and about 30% Black, Indigenous or people of colour. About 30% said speaking was difficult or impossible for them, and most answered the surveys themselves, with or without support (medRxiv).
The findings
The three recruitment groups differed markedly in their characteristics. In plain terms: a study that only recruits online, or only through clinics, may describe a very different “autistic adult” from one that recruits through disability services.
Limitations
This paper describes the sample; outcome results come later. All participants live in the United States. The team also reports that too few caregivers took part to test the versions of the measures meant for proxy reporting.
Were autistic people involved?
Yes. Participatory research is the core of the AASPIRE model, and the author list includes autistic researchers and community partners.
Why it matters
Many headlines about “autistic adults” rest on samples that are mostly white, speaking and recruited online. This study is a reminder to ask who was counted, and a practical example of “Nothing about us without us” in research design.
Is sensory over-responsivity linked to autism or ADHD?
A large 2026 meta-analysis suggests sensory over-responsivity is most closely linked to autism traits and anxiety, not to ADHD or other conditions. Sensory over-responsivity means strong, often distressing reactions to everyday sensations such as noise, touch, light or clothing textures.
The question
Sensory differences are reported in many conditions. Is sensory over-responsivity specific to some of them, and does it show up in brain connectivity?
The method and sample
Rebecca Schwarzlose and colleagues at Washington University School of Medicine in St. Louis combined five community and clinical datasets covering 15,728 children aged 6 to 18. A subset of 9,197 children had resting-state brain scans (fMRI), which measure how brain regions work together at rest. The study was published in August 2026 in the Journal of the American Academy of Child and Adolescent Psychiatry (The Transmitter).
The findings
When the researchers looked at all psychiatric conditions together, sensory over-responsivity was uniquely associated with anxiety and autism traits. Its links with ADHD and other conditions were no longer significant. The brain scans showed weaker connectivity between a cingulo-parietal network and the caudate, a deep brain structure.
Limitations
Sensory over-responsivity was measured with a single yes-or-no questionnaire item, not a detailed assessment. Some brain effects were small and did not replicate in a smaller dataset. The design shows association, not cause, and does not rule out a connection with ADHD.
Were autistic people involved?
The datasets included many autistic children, but coverage of the study does not say whether autistic people helped design or interpret it.
Why it matters
The study adds weight to what many autistic people describe: sensory overload is real, common and tied to anxiety. That supports adjusting environments, not training children to tolerate distress. See our sensory processing and mental health coverage.
How common is Phelan-McDermid syndrome?
A 2026 study estimates that Phelan-McDermid syndrome affects about 1 in 7,300 people, or roughly 45,000 people in the United States, far more than have been diagnosed. Phelan-McDermid syndrome is a genetic condition caused by a deletion or change in the SHANK3 gene on chromosome 22; most people with it also meet criteria for autism.
The question
How many people really have Phelan-McDermid syndrome, given that many are never genetically tested?
The method and sample
Tess Levy, Joseph Buxbaum and colleagues at the Seaver Autism Center at Mount Sinai in New York pooled genetic data from nearly 180,000 autistic people across ten sources, including commercial labs and the SPARK research study. They then adjusted for people who are never tested. The paper appeared in Autism Research (DOI), published online in June 2026 and in print in September 2026.
The findings
The estimate is 13.7 cases per 100,000 people. According to Mount Sinai, the gap between known and estimated cases is driven largely by underdiagnosis, because many people with developmental disabilities are never offered genetic testing.
Limitations
This is a modelled estimate built on assumptions about who gets tested. The authors note insurance barriers and that some genetic tests do not examine SHANK3 well, which could affect the numbers in either direction.
Were autistic people involved?
The study and press materials do not describe involvement of autistic people or families in the research design.
Why it matters
A genetic diagnosis can open doors to specialist medical care and family support groups. The senior author recommends genetic testing for every autistic child. Some autistic advocates are wary of genetics research aimed at “treating” autism itself, so testing should stay an informed choice, explained clearly and never pushed.
How should we read new autism research?
Read new autism research by asking three questions: who was studied, how was it measured, and who shaped the question? Today’s studies show that the answers can change what a finding means. Diagnosis gaps are a recurring theme; see our piece on late autism diagnosis in women.
FAQ
What is participatory autism research?
Participatory autism research is research in which autistic people are partners, not only participants. They help choose questions, design methods and interpret results. AASPIRE in the United States is one of the longest-running examples.
Is sensory over-responsivity the same as autism?
No. Sensory over-responsivity can occur without autism, and not every autistic person experiences it. The 2026 meta-analysis suggests it is most strongly associated with autism traits and anxiety.
Should every autistic child have genetic testing?
Some researchers recommend it, and testing can find conditions such as Phelan-McDermid syndrome. It is a personal and medical decision; families and autistic adults should discuss benefits and limits with a qualified clinician. This article is not medical advice.
Does a higher Phelan-McDermid estimate mean more people have it now?
No. The estimate suggests many existing cases were never diagnosed, mainly because of limited genetic testing. It is about better counting, not a rise in cases.
Kaynaklar
- Nicolaidis C, et al. Addressing common problems in autism research: cohort development and sample description for a participatory, real-world, longitudinal study of outcomes in autistic adults. Autism, 22 August 2026. DOI: 10.1177/13623613261476381
- University of Portsmouth research portal — record and abstract of the AASPIRE cohort paper
- medRxiv preprint (not peer reviewed) — Measuring High-Priority Outcomes in Autistic Adults: the AASPIRE Measurement Toolkit
- The Transmitter — Sensory over-responsivity tied to autism, anxiety but not other conditions (10 September 2026)
- Levy T, et al. Phelan-McDermid syndrome prevalence. Autism Research, 2026. DOI: 10.1002/aur.70297
- Mount Sinai press release — Phelan-McDermid syndrome may affect 1 in 7,300 people (July 2026)
- ScienceDaily — “Rare” autism-linked genetic disorder may be far more common than thought
This post was prepared with AI assistance and reviewed by the OtizmTV editorial team.