A large study of brain development in more than 8,600 children has identified a structural pattern that appears to track both psychological wellbeing and vulnerability to psychiatric illness. The research, based on data from the Adolescent Brain Cognitive Development (ABCD) study, suggests that differences in the architecture of the developing brain are not tied to one diagnosis alone. Instead, they may form a broad neurodevelopmental continuum associated with cognitive performance, emotional and behavioral difficulties, psychiatric comorbidity and the likelihood of remaining healthy or developing persistent mental health problems.
The investigation included 8,672 children who were approximately 9 to 10 years old at the beginning of the study, including 4,412 males and 4,260 females. Researchers combined detailed magnetic resonance imaging measurements with a wide range of behavioral and psychological assessments. These assessments covered cognitive ability, motivation, impulse control, emotional states and behaviors that ranged from healthy functioning to symptoms associated with psychopathology. The children were assessed at baseline and followed for two years, allowing the scientists to examine not only the presence of psychiatric problems but also how those problems changed over time.
Rather than searching for a single brain region linked to a specific disorder, the researchers used a machine-learning framework based on canonical correlation analysis. This statistical approach is designed to identify relationships between two complex sets of variables. In this case, one set described brain morphology, while the other captured cognitive, psychological and behavioral characteristics. By examining how multiple brain measurements covaried with multiple dimensions of behavior, the method generated latent brain and behavioral variates—composite scores that summarize patterns distributed across many regions and psychological domains.
The analysis revealed a robust brain structural variate spanning several forms of morphology. These included cortical surface area, cortical volume, cortical thickness, subcortical volume and sulcal or gyral depth, which describe the folds and contours of the brain’s outer surface. Children with higher scores on this brain pattern generally displayed stronger cognitive performance and lower scores on psychological measures associated with greater psychopathology. The finding is important because it points to a shared structural signature across diagnostic categories, rather than a pattern that maps neatly onto only attention-deficit/hyperactivity disorder, anxiety, depression or another individual condition.
The morphology associated with higher scores was especially notable in the cerebral cortex, the brain’s outer layer responsible for complex functions such as perception, language, planning and decision-making. Larger cortical surface area and greater cortical volume were prominent features, particularly in the temporal gyri, regions involved in auditory processing, language, memory and social cognition. Cortical surface area reflects how much territory the cortex covers, while cortical volume combines surface area with thickness. These characteristics are shaped by highly complex developmental processes, including genetic influences, cellular organization and the formation of long-range neural connections.
The researchers also identified a spatial pattern in cortical thickness that followed a posterior-to-anterior gradient. Higher brain-variate scores were associated with greater thickness in occipital, parietal and temporal regions, while thickness was lower in parts of the cingulate and frontal cortex. Cortical thickness does not have a simple interpretation in children: a thicker cortex is not automatically better, and a thinner cortex is not automatically worse. During development, thickness can reflect the timing of maturation, synaptic remodeling and other biological processes. The study therefore describes a coordinated pattern across regions rather than claiming that thickness in any single area directly determines mental health.
The brain pattern was also related to the cumulative burden of psychiatric diagnoses. Children with lower scores on the structural variate tended to have a greater number of co-occurring diagnoses, both at the initial assessment and at the two-year follow-up. This dose-dependent relationship suggests that the brain pattern tracked overall psychiatric burden across conditions. In other words, the association became more pronounced as the number of diagnoses increased, supporting the idea that some aspects of brain development may be transdiagnostic—shared across multiple forms of mental illness—rather than specific to conventional diagnostic boundaries.
Longitudinal analyses added another layer to the findings. Lower baseline scores were associated with persistent psychiatric diagnoses, while higher baseline scores were associated with persistent healthy states. The researchers interpreted this pattern as evidence for a possible vulnerability–resilience continuum: a distributed brain profile may be related to the probability of maintaining psychological health or remaining vulnerable to continuing difficulties. The results do not show that brain structure causes psychiatric disorders, nor can the measurements predict an individual child’s future with certainty. Mental health is influenced by genetics, family relationships, stress, education, sleep, physical health and many environmental factors that cannot be reduced to an MRI-derived score.
The study’s scale and multimodal design make the findings potentially valuable for developmental neuroscience, but the authors’ conclusions should be understood as evidence of association rather than a ready-made clinical test. Machine-learning models can reveal subtle patterns that are difficult to detect with traditional region-by-region analyses, yet they must be tested in independent populations before they can support screening or intervention decisions. The children in the ABCD cohort also represent a particular developmental period, and brain patterns may change as participants move through adolescence, when cortical maturation, puberty and the emergence of psychiatric symptoms accelerate. Future research will need to determine whether the identified variate remains stable across later developmental stages and whether combining morphology with genetics, environmental exposure and repeated behavioral measurements improves prediction.
The findings nevertheless offer a compelling new view of childhood mental health. Instead of treating psychiatric disorders as entirely separate conditions with isolated biological signatures, the results suggest that a common dimension of brain development may help explain why cognitive strengths, psychological symptoms and diagnostic comorbidity often overlap. If replicated, morphology-informed approaches could eventually contribute to earlier identification of children who need support, while avoiding the assumption that a brain scan alone can define a diagnosis. For now, the study provides a large-scale map of how developing brain structure relates to a broad spectrum of human behavior—and raises the possibility that resilience and vulnerability emerge from the same continuously changing neurodevelopmental landscape.
Subject of Research: Brain morphology, cognitive function, psychological processes, behavioral traits and transdiagnostic psychiatric vulnerability and resilience in preadolescents.
Article Title: Brain morphological pattern is associated with the presence, severity and transition of transdiagnostic psychiatric disorders in preadolescents.
Article References: Kuang, N., Hammond, C.J., Salmeron, B.J. et al. Brain morphological pattern is associated with the presence, severity and transition of transdiagnostic psychiatric disorders in preadolescents. Nature Mental Health (2026). https://doi.org/10.1038/s44220-026-00704-7
Image Credits: AI Generated
DOI: https://doi.org/10.1038/s44220-026-00704-7
Keywords: adolescent brain development, brain morphology, cortical thickness, cortical surface area, psychiatric disorders, psychopathology, resilience, vulnerability, machine learning, canonical correlation analysis, ABCD study, cognitive function, preadolescents, transdiagnostic neuroscience
