Conversations about autism are often complicated by the fact that autism spectrum disorder (ASD) looks and feels different in different people. Because of this variability, we call autism a spectrum—a misleading term that suggests a linear progression from mild to severe.

That’s why a recent research paper published in Nature Genetics is so important: It advances our understanding by identifying four distinct subtypes of autism. Conducted by researchers at Princeton University and the Simons Foundation, the study linked each subtype to its own unique cluster of symptoms and genetic markers.

A Person-Centered Approach to Autism Research

Using a method called generative mixture modeling, researchers analyzed a large dataset of both behavioral and genetic information from more than 5,000 children in the SPARK autism cohort, funded by the Simons Foundation. This database included detailed tracking of each child’s individual symptoms.

The team called their approach “person-centered,” examining more than 230 traits in the children, including social behaviors, repetitive actions, and developmental milestones.

This approach flipped the script on autism research, which often looks for single genes linked to isolated symptoms. Instead, the researchers identified four distinct autism subtypes first, then examined the genetic patterns unique to each group.

4 Types of Autism

Autism is a difference in brain wiring that leads to a unique way of experiencing the world, but it is only diagnosed based on two parts of the autistic experience. Clinicians call these the “core autism traits,” which include social difficulties and restrictive or repetitive behaviors.

The study found that within those diagnosed with autism, there are four subtypes of autism with their own profiles beyond the core autism traits.

  • Social and behavioral challenges: Including about 37 percent of the study participants, these autistic individuals typically reach developmental milestones at a pace similar to children without autism. They also demonstrate core autism traits of social challenges and repetitive behaviors, and are more likely to have attention-deficit/hyperactivity disorder (ADHD), obsessive-compulsive disorder, anxiety, or depression, along with their autism.
  • Mixed ASD with developmental delay: This group of children included roughly 19 percent of the study participants, and usually showed developmental delays in areas like walking or talking. However, they did not usually have anxiety, depression, or disruptive behaviors. The study authors used the word mixed because the way the children’s core autism traits presented varied.
  • Moderate challenges: Representing about 34 percent of the study participants, the moderate challenges group generally reached developmental milestones at the same time as children without autism. They also demonstrated the core autism behaviors, but with less intensity than children in the other groups. This group typically did not have co-occurring psychiatric conditions.
  • Broadly affected: The broadly affected group of children was the smallest, including about 10 percent of the study participants. These children had more extensive and extreme challenges, including developmental delays, social and communication difficulties, and repetitive behaviors. They also had co-occurring psychiatric conditions like anxiety, depression, and mood dysregulation.

It would be natural for autistic individuals and those who love them to try to fit themselves into one of these groups, but the study authors caution that there may be more types of autism beyond these. “It means we now have a data-driven framework that shows there are at least four—and that they are meaningful in both the clinic and the genome,” Avila Litman, a Ph.D. student at Princeton and co-lead author, explained in a press release.

The Genetics Behind the Autism Subtypes

Autism has long been known to be genetic, but existing research looking for single genes has made it hard to find all the genes related to autism. Up until now, standard genetic testing has revealed genes to explain autism in only about 20 percent of patients. By using a large cohort and defining autism subtypes, this study was able to map each of the four subtypes to their own genetic pattern.

“What we’re seeing is not just one biological story of autism, but multiple distinct narratives,” said Natalie Sauerwald, associate research scientist at the Flatiron Institute and co-lead author, in the press release. “This helps explain why past genetic studies often fell short—it was like trying to solve a jigsaw puzzle without realizing we were actually looking at multiple different puzzles mixed together. We couldn’t see the full picture, the genetic patterns, until we first separated individuals into subtypes.”

Those genetic patterns, in turn, drive biological pathways and the patterns of child development that families observe. For example, in the social and behavioral challenges autism subtype, the study identified mutations in genes that turn on later in childhood. This makes sense in a group that shows no early developmental delays and a later autism diagnosis. Further, this autistic subtype may be the exception to the currently accepted wisdom that the genes that drive autism affect the brain before birth.

Meanwhile, the broadly affected group of autistic children had a high number of de novo mutations. These are new genetic mutations that happen out of nowhere and did not exist in the parents. They also tend to be more severe in their biological impact, which correlates with the severity of the symptoms observed in these children.

On the other hand, only the mixed ASD with developmental delay group was more likely to carry rare inherited genetic variants when compared to other subtypes. This suggests that while we might see the same behaviors from the outside, the genetic mechanisms driving those behaviors can be very different.

A Better Understanding of Autism Spectrum Disorder

Exploring autism subtypes also sheds light on long-standing questions. Why do some families have multiple relatives with high-functioning autistic traits, while others see a single child with more significant challenges? Why are some children diagnosed in early childhood, while others aren’t identified until later in life?

This paper marks a meaningful step forward in autism research. Clarifying autism subtypes and how they progress developmentally could help us offer more targeted support to autistic individuals and their families. For researchers, it introduces a clearer framework for studying autism. For clinicians, individuals, and families, it offers insight into why autism can present so differently from one person to another. As the model is refined through future research, it may help us better anticipate needs and tailor support accordingly.

Note on language: Many members of the autistic community prefer to be referred to with identity-first language, such as “autistic person” or “autistics,” because they see their autism as a valuable part of who they are. The language in this article honors that preference.