Summary: A new study finds that the severity of autistic symptoms — rather than the formal diagnostic label — corresponds to common brain connectivity patterns across children diagnosed with either autism spectrum disorder (ASD) or attention-deficit/hyperactivity disorder (ADHD). Greater autistic traits were associated with increased functional connectivity between the frontoparietal and default-mode networks, brain systems central to social cognition and executive function.
Those connectivity differences also corresponded with spatial patterns of gene expression tied to neural development, pointing to shared biological mechanisms that transcend diagnostic categories. Taken together, the results favor dimensional, biology-informed frameworks for understanding neurodevelopmental conditions and for guiding future biomarker research.
Key Facts
- Symptom-driven brain patterns: Autism symptom severity predicted specific brain connectivity differences irrespective of whether children had an ASD or an ADHD diagnosis.
- Gene expression overlap: The connectivity pattern mapped onto genes involved in neural development, genes previously implicated in both autism and ADHD.
- Supports dimensional models: Findings bolster transdiagnostic, biology-driven approaches that consider symptom dimensions across traditional diagnostic boundaries.
Source: Child Mind Institute
Overview: A study published in Molecular Psychiatry by researchers at the Child Mind Institute and partner institutions examined how autistic traits relate to brain connectivity and gene expression in children diagnosed with ASD or ADHD. The research highlights that autistic symptom severity, rather than categorical diagnosis, corresponds to distinct functional brain network differences and associated gene-expression signatures.
The investigators evaluated resting-state functional MRI data from 166 verbal children aged 6–12 who had carefully established primary diagnoses of either autism or ADHD (without autism). Using a multivariate, connectome-based approach, they tested how individual differences in whole-brain intrinsic functional connectivity relate to clinician-observed autism symptoms and parent-reported ADHD symptoms.
Across the sample, more severe autistic symptoms were linked with stronger intrinsic functional connectivity between two specific network nodes: the middle frontal gyrus in the frontoparietal (FP) network and the posterior cingulate cortex in the default-mode (DM) network. These networks support executive control and social cognition, respectively. In typical development, connectivity between these networks tends to decrease with maturation, supporting increasing functional specialization; the observed increase in connectivity associated with greater autistic symptom severity suggests atypical maturation of internetwork communication in these children.
Importantly, this pattern appeared in children regardless of whether they carried an ASD label. In other words, some children with ADHD who did not meet full criteria for autism nonetheless showed similar brain connectivity changes when they exhibited higher levels of autistic symptoms.
To link these macroscale connectivity findings to molecular biology, the team applied in silico spatial transcriptomic analysis. This computational method compares observed connectivity maps to existing brain gene-expression databases. The connectivity map associated with autism severity overlapped with expression patterns of genes involved in neural development and neuronal projections—genes that show greater variability in both autism and ADHD. This overlap suggests a shared genetic contribution to the observed brain-clinical phenotype.
The study did not find statistically significant brain–behavior relationships tied to ADHD symptom measures in the same analyses, highlighting the specific association between internetwork connectivity and autistic symptom severity.
Key findings include:
- Autism symptom severity is associated with consistent patterns of brain connectivity across children diagnosed with ASD and in a subset of children with ADHD who show elevated autistic traits.
- These connectivity differences align with brain expression of genes implicated in neural development and neuronal projection processes.
- Shared clinical presentations of autistic traits and related connectivity patterns point to overlapping genetic mechanisms between autism and ADHD for this phenotype.
- Altered maturation of functional network interactions may contribute to the development of autistic symptoms in children across diagnostic categories.
- Results support the importance of integrating both dimensional (symptom-focused) and categorical (diagnostic) perspectives in neurodevelopmental research and clinical practice.
- The integrative imaging-to-transcriptomics approach used in this work may help guide future biomarker discovery and models of vulnerability for autistic symptom severity.
Implications for clinical practice and research:
These results argue for moving beyond strict diagnostic boundaries and toward assessment strategies that emphasize specific symptom dimensions and their biological correlates. Identifying neural and molecular signatures tied to symptom severity could support more precise, individualized approaches to intervention and monitoring. The study aligns with broader efforts in psychiatry to adopt dimensional, transdiagnostic, and data-driven models of mental health.
The Child Mind Institute’s Healthy Brain Network and similar large-scale initiatives that combine neuroimaging, phenotypic assessment, and open data access are well positioned to advance this line of research by enabling discovery and validation of biomarkers across diverse clinical presentations.
Key Questions Answered:
A: Autism symptom severity — rather than diagnostic label — mapped onto distinct brain connectivity patterns shared across children with ASD and those with ADHD who have elevated autistic traits.
A: Stronger connectivity between the frontoparietal network’s middle frontal gyrus and the default-mode network’s posterior cingulate cortex tracked with more severe autistic symptoms.
A: Yes. The connectivity pattern associated with autistic symptom severity overlapped with the brain expression of genes involved in neural development and neuronal projections—genes previously implicated in both autism and ADHD.
About this research news
Author: Media Office
Source: Child Mind Institute
Contact: Media Office – Child Mind Institute
Image: The image is credited to Neuroscience News
Original research (open access): “Connectome-based symptom mapping and in silico related gene expression in children with autism and/or attention-deficit/hyperactivity disorder” by Adriana Di Martino et al., published in Molecular Psychiatry.
Abstract (summary):
Clinical, neuroimaging, and genomic evidence indicate overlap between autism and ADHD. This study assessed how core symptom dimensions contribute to shared biology in 166 verbal children (6–12 years) with well-established primary diagnoses of autism or ADHD. The researchers examined associations between low-motion whole-brain intrinsic functional connectivity and dimensional measures of autism and ADHD symptoms, using clinician observation and parent interview data.
Multivariate distance matrix regression identified a transdiagnostic association between autism severity and intrinsic functional connectivity of two left-hemisphere nodes: the middle frontal gyrus (frontoparietal network) and the posterior cingulate cortex (default-mode network). Greater connectivity between these nodes tracked with more severe autistic symptoms across children, even after controlling for ADHD ratings. Secondary analyses supported the primary findings, emphasizing internetwork connectivity as a marker of autism symptom severity across diagnoses. No significant brain–behavior relationships were detected for ADHD symptoms in these analyses.
Gene enrichment analyses linked the autism-associated connectivity map to genes with higher variance in autism and ADHD and to genes involved in neuronal projections, suggesting shared genetic mechanisms underlying this brain–clinical phenotype. Overall, the results highlight the value of transdiagnostic, dimensional approaches for connecting clinical observations to shared circuit- and genomic-level presentations across diagnostic groups.