Research Shifts Focus on Autism’s Brain Mechanisms

For a long time, studies on neurodevelopment, particularly concerning autism, have largely concentrated on the cerebral cortex—the area in our brains responsible for intellect and learning. Recent research, published in May 2026, indicates that the cerebellum might actually hold a more significant role in autism than previously recognized.

The researchers highlighted that perineuronal nets (PNNs), which are small supportive structures around cerebellar neurons, help facilitate proper communication in the networks involved in social interactions. When these structures were compromised, there was a noticeable shift in social behavior, offering fresh insights into the biological mechanisms associated with autism spectrum disorder (ASD).

Impact of Cerebellar Changes on Social Behavior

Conducted by Kanazawa University in Japan, this study published its findings in the journal Translational Psychiatry on May 13. The team utilized two distinct mouse models for their exploration. In one case, autism-like traits were induced in mice exposed to valproic acid before birth, while the other model involved mice with a mutation in the autism-linked gene CHD8. Despite these differing backgrounds, both models demonstrated a significant reduction in PNNs within the cerebellum.

To better understand the effects of these changes, researchers disrupted PNNs in healthy mice. The results were significant: those with damaged PNNs exhibited less social engagement and showed decreased interest in unfamiliar mice. These are behaviors normally linked to ASD.

Subsequent experiments shed light on the underlying reasons for this behavior. When healthy mice engaged in social interactions, neurons in the cerebellum activated, sending signals to other brain areas responsible for processing social information. Disrupting PNNs diminished this activity, hampering communication across these brain networks.

Additionally, the team identified a crucial molecular player known as ARNT2. When PNNs diminished, the levels of this protein rose, causing neurons to become less responsive. Interestingly, reducing ARNT2’s activity helped restore brain functionality and social behavior in the affected mice.

This research indicates that difficulties related to autism might not stem only from changes in the brain’s cognitive regions. Structural variations in the cerebellum and its networks seem to significantly contribute to these challenges. Still, more research is needed to determine if this same mechanism exists in humans. Nevertheless, these findings open up interesting avenues for understanding autism’s biological foundations.

New Insights into Autism Subtypes

Interestingly, this study coincided with another significant effort identifying two distinct subtypes of autism. Researchers from the Instituto Italiano di Tecnologia in Rovereto, Italy, and the Child Mind Institute in New York, in collaboration with the University of Trento, conducted this analysis, which was published in the journal Nature Neuroscience on May 15.

Here, the researchers analyzed brain scans from 20 various mouse models that exhibited autism-like traits and compared them with scans from 940 children and young adults diagnosed with autism, as well as 1,036 neurotypical individuals. They aimed to identify differences in communication across brain regions.

The analysis revealed two main patterns. One involved a hypoconnectivity group, where autism linked to diminished brain connectivity. This was tied to brain activity associated with genes responsible for synaptic junctions. The second, a hyperconnectivity group, was characterized by increased brain connectivity and showed links to immune system-related genes, often associated with more severe autism.

Because these patterns appeared consistently in both mice and humans across various datasets, researchers have reason to believe they may represent genuine biological subtypes of autism. If this hypothesis is confirmed and coupled with reliable diagnostic methods, it might lead to more tailored treatment approaches for individuals based on their specific subtype.

Historical Perspective on Autism Research

Interestingly, this current inquiry isn’t the first to propose categorizing autism into various types. A previous study from July 2025 by researchers at Princeton University and the Simons Foundation pinpointed four autism types based on behavior rather than brain imaging, analyzing over 230 characteristics in a group of 5,000 children.

Other studies also suggest that autism can manifest differently depending on its onset during various life stages, like early childhood, late childhood, adolescence, or even young adulthood. Combined, these efforts aim to enhance how autism is recognized and understood.

In the pursuit of identifying more autism subtypes, researchers believe that larger datasets and advanced analytical techniques will prove invaluable. The findings will contribute to a better understanding of autism and its biological nuances.