Autism: Discovery Could Revolutionize Diagnoses and Treatments

What if autism wasn't a single condition? Impressive research has analyzed the brains of nearly 2,000 people and discovered that there are at least two distinct biological patterns behind autism. This discovery could change how we understand the disorder and pave the way for much more personalized treatments in the future.

For many years, autism was treated as a single condition that manifests itself differently in each person. However, doctors, psychologists, and neuroscientists have always observed something intriguing: two people diagnosed with autism can be extremely different from each other.

Some have great difficulty communicating, while others have fluent language. Some have intellectual disabilities, while others have above-average intelligence.

This enormous diversity has led scientists to question whether autism is truly a single condition or whether it could represent several different biological subtypes grouped under the same diagnosis.

The difficulty in answering this question is that the genetics of autism are extremely complex. Hundreds of genes associated with the risk of the condition have been identified, but none of them explains the majority of cases. Furthermore, environmental factors, such as inflammation during pregnancy and changes in the immune system, also appear to contribute to the development of the disorder. This means that different people can arrive at the same diagnosis through completely distinct biological pathways.

To investigate this question, researchers adopted an innovative strategy. First, they studied 20 different models of mice genetically modified to exhibit alterations related to autism. Each group of animals possessed different biological alterations, allowing scientists to observe how these changes affected brain function.

To do this, they used functional magnetic resonance imaging (fMRI), a technique that measures brain activity by observing blood flow in different regions of the brain. The goal was to discover if there were consistent patterns of brain communication associated with different biological mechanisms.

By analyzing the data, the researchers realized that the mice's brains did not present a single pattern of alteration. Instead, two large groups emerged. In the first, different brain regions communicated less than expected, a phenomenon called hypoconnectivity.

In the second, the regions exhibited excessive communication, known as hyperconnectivity. Interestingly, these two patterns were linked to distinct biological mechanisms. Hypoconnectivity was more associated with problems in synapses, the connections that allow neurons to exchange information. Hyperconnectivity, on the other hand, appeared to be related to alterations involving the immune system and processes that regulate gene activity.

Functional Magnetic Resonance Imaging in Mice

The next step was to verify if the same thing happened in humans. To do this, scientists analyzed brain scans of more than 1,900 people, including 940 individuals with autism and 1,036 neurotypical participants. Using the same analysis methods employed in mice, they sought to identify similar patterns of brain connectivity.

The result was surprising: the same two major profiles appeared in the autistic participants. Some people showed a predominance of hypoconnectivity, while others showed hyperconnectivity. Furthermore, these groups also exhibited behavioral and cognitive differences, suggesting that it was not simply a random variation.

This discovery is important because it may help explain why treatments and interventions work very well for some people and have little effect on others. If different groups of autistic individuals have distinct biological mechanisms behind their symptoms, it is possible that they also need different therapeutic approaches.

In other words, what we call autism today may represent a set of related, but not necessarily identical, conditions.

Although further research is needed to confirm these findings and translate them into clinical applications, the study offers one of the strongest pieces of evidence to date that there are distinct biological subtypes within the autism spectrum.

Instead of viewing autism as a single condition, science is beginning to consider the possibility that it is composed of different brain, genetic, and biological profiles. This shift in perspective may pave the way for more accurate diagnoses and more personalized treatments in the future.

Researcher Alessandro Gozzi, PhD, director of the Center for Neuroscience and Cognitive Systems (CNCS) at IIT in Rovereto, Italy, who coordinated the study with Adriana Di Martino, MD, founding director of the Child Mind Institute's Autism Center in New York. Credit: IIT - Italian Institute of Technology

READ MORE:

Autism subtypes identified using cross-species functional connectivity analyses

Marco Pagani, Valerio Zerbi, Silvia Gini, Filomena Grazia Alvino, Abhishek Banerjee, Andrea Barberis, M. Albert Basson, Yuri Bozzi, Alberto Galbusera, Jacob Ellegood, Michela Fagiolini, Jason P. Lerch, Michela Matteoli, Caterina Montani, Davide Pozzi, Giovanni Provenzano, Maria Luisa Scattoni, Nicole Wenderoth, Ting Xu, Michael V. Lombardo, Michael P. Milham, Adriana Di Martino, and Alessandro Gozzi

Nature Neuroscience.  29, pages 1476-1487 (2026) 15 May 2026DOI: 10.1038/s41593-026-02287-z

Abstract: 

It is often assumed that phenotypic heterogeneity in autism reflects underlying pathobiological variation. However, direct evidence supporting this link is lacking. Leveraging cross-species functional neuroimaging, we show that brain dysconnectivity patterns in autism can be parsed into biologically dissociable subtypes. Specifically, we found that functional magnetic resonance imaging (fMRI) connectivity alterations in 20 distinct genetic mouse models of autism cluster into hypoconnectivity-dominant and hyperconnectivity-dominant subtypes. These subtypes are linked to distinct biological pathways, with hypoconnectivity being associated with synaptic dysfunction and hyperconnectivity reflecting transcriptional and immune-related alterations. Here we identified analogous hypoconnectivity and hyperconnectivity subtypes in a multicenter human fMRI dataset of n = 940 individuals with idiopathic autism and n = 1,036 neurotypical individuals. The human autism subtypes are highly replicable, are associated with distinct functional network architectures and behavioral profiles and recapitulate the synaptic and immune-related pathways identified in the rodent dataset. Our work provides a new empirical framework for targeted subtyping of the autism spectrum.