6 Minutes
Imagine two children in the same classroom. One has an autism diagnosis, the other ADHD. On paper they are different. In the brain, they may share more than we thought.
Shared signals hidden in brain networks
A recent study led by researchers at the Child Mind Institute and published in Molecular Psychiatry reframes how we understand autism and attention-deficit/hyperactivity disorder. Instead of treating the labels as impermeable walls, the team looked for patterns that cut across diagnoses. What they found pushes psychiatry toward a dimensional view: symptom severity, especially autistic traits, maps onto distinct patterns of brain connectivity and the activity of genes tied to neural development.
Why does that matter? Because clinicians have long seen symptom overlap — children diagnosed with ADHD can show social or cognitive profiles that look a lot like autism, and vice versa. The new work asks a simple, probing question: are those behavioral similarities reflected in shared biology? The short answer: yes.
Methods and dataset
The group analyzed resting-state functional MRI from 166 verbal children aged 6 to 12, each diagnosed either with autism spectrum disorder or with ADHD (without autism). Resting-state imaging captures the ebb and flow of communication among brain regions while a person is at rest, revealing the architecture of large-scale networks such as the frontoparietal (FP) system and the default mode network (DMN).
Rather than comparing diagnostic bins, the team quantified symptom severity across the whole sample and related those measures to connectivity patterns. Then they took a step deeper: using spatial transcriptomic maps — large databases that tell us where genes are active in the human brain — they overlaid gene-expression profiles onto the connectivity patterns they observed. The result was a convergence of imaging and genetics that points to shared biological pathways.
Children with more pronounced autistic traits showed stronger functional connectivity between FP and DM networks. These systems govern executive control, attention, and social cognition. Normally, FP–DM connectivity declines as brains specialize during development. The heightened connectivity seen here suggests altered maturation trajectories in children with greater symptom severity.
This pattern links symptom severity to a shared brain–gene architecture across diagnoses.
What the findings reveal about diagnosis and biology
The most striking outcome is that the connectivity signature tied to autistic traits was not exclusive to children carrying an ASD diagnosis. A subset of children diagnosed with ADHD — who do not meet full criteria for autism — displayed the same neural pattern. In parallel, the brain regions implicated were those where genes involved in neural development are active. Some of those genes have prior associations with both autism and ADHD, suggesting overlapping genetic influences on emergent behavior.
Put simply: behavioral features that clinicians recognize as similar may arise from overlapping alterations in how large-scale brain networks mature, influenced by certain genetic programs. The study does not erase diagnostic categories. Instead it provides a richer map: diagnoses remain clinically useful, but they are only one lens through which to view a child’s neurobiology.
The approach used by the researchers — combining cutting-edge neuroimaging with in silico spatial transcriptomics — offers a template for future biomarker discovery. When imaging and gene maps converge, researchers can begin to identify biological signals that predict who is most likely to develop more severe autistic traits, regardless of their diagnostic label.
Implications for clinicians and researchers
What should a clinician take away? First, symptom dimensions matter. A binary diagnosis can obscure meaningful variation in how brain networks are organized. Second, personalized assessment that considers the neural profile — not just the diagnostic category — could lead to more targeted interventions. Third, research that embraces transdiagnostic, dimensional frameworks may accelerate the search for objective markers of risk and treatment response.
The study also highlights the value of large-scale data resources. Initiatives like the Healthy Brain Network, which provides no-cost assessments and open datasets, make it possible to link behavior, imaging, and genetics across thousands of children. Those datasets are the scaffolding for studies that hope to move psychiatry from symptom checklists to biology-informed decision making.
Expert Insight
“We see in the clinic that symptom profiles blur the lines between traditional diagnoses,” says Dr. Adriana Di Martino, who led the research. “By anchoring behavioral severity to brain–gene expression patterns, we are beginning to map the biological continuum that underlies neurodevelopmental diversity.”
Dr. Elena Morales, a fictional but representative developmental neuroscientist, adds: “This work is a reminder that brain networks develop according to dynamic trajectories. When maturation diverges, similar behavioral patterns can emerge from shared network and genetic influences. That knowledge should inform both assessment and intervention strategies.”
The study does not provide immediate new treatments. But it does chart a course. Focus less on the categorical label and more on the underlying biology — that shift could change how we detect vulnerability early, how we tailor therapies, and how we design trials that test interventions against specific neural signatures.
The path forward will require replication, larger samples, and longitudinal studies that trace how these network patterns change with age and with treatment. Still, the message is clear: the brain does not read diagnostic manuals. To find the mechanisms behind behavior, we must follow the biology where it leads.
Source: scitechdaily
Comments
bioNix
Wow this flips how I think about diagnoses, really cool to see genes and networks lining up. But curious, does this track over time? kinda hopeful ngl
Leave a Comment