Autism Heterogeneity May Reflect Two Distinct Brain Network Patterns
Autism spectrum disorder is defined by wide variation in social behavior, communication, sensory processing, and day-to-day support needs. That diversity has long complicated brain research: some imaging studies have reported reduced functional connectivity, others increased connectivity, and others a mix that seemed hard to reconcile. A new study published in Nature Neuroscience argues that part of that inconsistency may reflect the existence of two distinct autism subtypes.
The work, led by researchers at the Italian Institute of Technology's Center for Neuroscience and Cognitive Systems and the Child Mind Institute in New York, identified one subtype characterized by hypoconnectivity and another by hyperconnectivity. Rather than treating conflicting imaging results as noise, the team proposes that they may capture real biological differences inside the broad autism diagnosis.
Why the Finding Matters
Autism research has struggled with a central problem: the condition is clinically heterogeneous, but many studies have still searched for a single underlying neural signature. That approach can flatten meaningful differences. If distinct groups of autistic people have different large-scale brain network organizations, averaging them together may obscure the very patterns researchers are trying to detect.
Senior author Alessandro Gozzi told Medical Xpress that the study grew out of frustration with how autism imaging results have often been interpreted. Reduced connectivity, increased connectivity, and more complex findings have all appeared in the literature. A subtype framework offers a possible explanation for why those results have not lined up cleanly.
From Contradiction to Stratification
The main advance here is conceptual as much as technical. Instead of asking whether autism is associated with too much or too little connectivity overall, the study asks whether different autistic subgroups show different network organizations. In that framework, conflicting past findings are not necessarily errors. They may be partial views of a more structured biological landscape.
According to the source text, the researchers linked those subtypes to distinct connectivity patterns seen on functional MRI. One group showed atypical hypoconnectivity, while the other showed hyperconnectivity. That is a notable claim because it suggests the diversity seen clinically may have measurable neurobiological correlates rather than being only a descriptive feature at the behavioral level.
What This Could Change
If the subtype model holds up, it could affect how autism studies are designed and how future interventions are evaluated. Clinical trials and biomarker work often depend on grouping participants in ways that capture meaningful differences. A broad category can be useful for diagnosis and services, but it may be too blunt for mechanistic neuroscience.
Stratification could improve reproducibility by ensuring researchers are not combining participants with fundamentally different brain connectivity profiles into one undifferentiated sample. It could also help explain why some findings replicate weakly across cohorts. If the balance of subtypes shifts from one sample to another, the average result could change even when each subtype is internally consistent.
What the Study Does Not Say
The source material does not support sweeping clinical claims, and the findings should be read carefully. The study does not suggest that autism can be reduced to just two lived experiences, nor does it imply that brain scans are about to replace behavioral diagnosis. Autism remains a complex neurodevelopmental condition, and the source text emphasizes that people on the spectrum vary widely in needs and abilities.
What the paper appears to offer is a more refined neurobiological model. That is important, but it is not the same as an immediate change in clinical practice. As with many imaging studies, the key next steps will involve replication, validation across populations, and testing whether the identified subtypes map onto meaningful differences in development, behavior, or treatment response.
A Larger Shift in Autism Neuroscience
The study also fits into a broader scientific move away from one-size-fits-all explanations in psychiatry and neurodevelopment. Many disorders are now understood as umbrella diagnoses that likely contain multiple biological pathways. Autism has always been a prime candidate for that kind of reframing because its heterogeneity is so visible in real life.
Finding coherent subtypes would not erase that complexity. It would give researchers a better starting structure for studying it. That alone could help move the field beyond arguments over whether earlier imaging inconsistencies were mainly technical artifacts or evidence of a deeper unresolved problem.
The Most Credible Takeaway
The strongest reading of the new paper is that it turns an old contradiction into a testable hypothesis. Some autistic individuals may show reduced functional connectivity, others increased connectivity, and both patterns may be real components of the condition rather than mutually exclusive findings.
For a field long weighed down by mixed imaging results, that is a meaningful step. It does not solve autism neuroscience in one paper, but it offers a clearer map for where the search should go next.
This article is based on reporting by Medical Xpress. Read the original article.
Originally published on medicalxpress.com
