Autism isn’t just one condition; it appears to have at least two biologically distinct forms that can be identified by looking at how different parts of the brain communicate with each other, according to a study published in Nature Nueroscience. The authors highlight that this could pave the way for more personalized care.
Two Different Patterns in the Brain
Using functional MRI (fMRI) brain scans, the team from the Italian Institute of Technology (IIT) in Rovereto, Italy identified two clear subtypes of autism, inclduing a hyperconnectivity subtype, where different brain regions communicate more than usual and a hypoconnectivity subtype, where communication between brain regions is reduced. Together, these two patterns described about 25% of the autistic individuals examined in the study, suggesting more subtypes likely remain to be discovered.
How They Did It
To make sense of these patterns, the researchers needed to connect what they saw on brain scans to what was happening at the biological level. They did this by studying 20 different mouse models of autism alongside brain scans from 940 children and young adults with autism and more than 1,000 neurotypical individuals.
The mouse studies acted as a kind of decoder. By combining brain imaging with genetic and biochemical analyses in mice, the team could see exactly which biological pathways produced which connectivity patterns. They then looked for those same patterns in the human brain scans.
“The mouse models gave us a biological ‘Rosetta Stone,'” said Dr. Adriana Di Martino, founding director of the Autism Center at the Child Mind Institute. “We could see which biological pathways drive which connectivity signatures, then search for those same patterns in humans.”
The human brain scan data came from the Autism Brain Imaging Data Exchange (ABIDE), a global neuroimaging project co-founded by Dr. Di Martino, along with data from the Child Mind Institute.
Different Patterns, Different Biology
The findings tied each connectivity pattern to a specific biological story. The hypoconnectivity subtype was linked to synaptic pathways, meaning the connections between brain cells. The hyperconnectivity subtype, on the other hand, was linked to immune-related systems in the brain.
When the team analyzed gene activity in the human brains, the results matched what they had seen in mice. Brain regions with reduced connectivity were rich in synaptic genes, while those with increased connectivity were rich in immune-related genes.
“For decades, we’ve observed tremendous variability in how autism manifests, but we lacked direct evidence that these differences reflected distinct underlying biology,” said Dr. Alessandro Gozzi, director of the Center for Neuroscience and Cognitive Systems at IIT, who co-led the study. “Our approach enabled us to isolate specific genetic and immune factors, then translate those signatures to human brain scans, showing that different connectivity patterns encode different mechanistic pathways underlying autism.”
Critically, the same two subtypes showed up consistently across many independent research sites. “Finding the same subtypes reproducible across dozens of independent research sites was critical validation,” Dr. Gozzi added.
What This Means for the Future
The two subtypes also showed slight differences in standard autism assessments, with the hyperconnectivity group scoring somewhat higher on measures of severity. But the biological differences ran deeper than behavior alone could reveal.
“Brain-based biological markers reveal distinctions that current behavioral assessments don’t fully capture,” Dr. Di Martino noted.
The researchers stress that these two subtypes are only part of the picture. Autism is a broad spectrum, and larger datasets and improved methods will likely uncover even more subtypes. But by tying brain patterns to specific biology, this work offers a real foundation for precision medicine, opening the door to support and care that’s tailored to each person’s underlying biology rather than a one-size-fits-all approach.
Pagani, M., Zerbi, V., Gini, S. et al. Autism subtypes identified using cross-species functional connectivity analyses. Nat Neurosci 29, 1476–1487 (2026). https://doi.org/10.1038/s41593-026-02287-z