Abstract
Overlap of brain changes across mental disorders has reinforced transdiagnostic models. However, the developmental basis for this overlap is unclear, as are neural differences among internalizing, externalizing, and thought disorders. We examined cortical thickness (CT) differences between healthy controls (n=4041) and patients with externalizing (n=1182), internalizing (n=1959), and thought (n=347) disorders in preadolescents (9-10 years) from the Adolescent Brain and Cognitive Development (ABCD) Study. We found that youth with externalizing or internalizing disorders exhibited significantly thicker cortex than controls. These disorders shared thicker CT in the left pars opercularis and caudal middle frontal gyrus, related to lower cognitive performance. Externalizing disorders uniquely affected somatosensory and primary auditory cortex, while internalizing disorders uniquely affected primary motor and higher-order visual areas. Only the externalizing group demonstrated significant CT increase at 2-year follow-up and decelerated cortical thinning. Genetic analysis implicated microglia in both phenotypes, while dopaminergic/glutamatergic/GABAergic cells related only to externalizing-specific regions. These results suggest that distinct anatomical trajectories result from unique genetic and cell-type changes occurring against a background of shared morphological variance.
Introduction
Psychiatry is increasingly moving away from rigid, siloed diagnostic categories toward transdiagnostic models, such as the Hierarchical Taxonomy of Psychopathology (HiTOP). These models recognize that many mental disorders share symptoms and genetic risks. However, the "brain blueprint" of these shared versus specific traits remains poorly understood, particularly during the critical developmental window of preadolescence.
During this period, the brain typically undergoes "cortical thinning"—a pruning process that refines neural circuits. This study investigates whether disruptions in this thinning process serve as markers for internalizing (e.g., anxiety, depression), externalizing (e.g., ADHD, conduct issues), and thought disorders.
Methods
The researchers utilized data from the Adolescent Brain Cognitive Development (ABCD) Study (Release 3.0), involving 11,878 participants at baseline (aged 9-10) and 6,571 at a 2-year follow-up. The study employed a multi-modal approach:
- Case-Control Analysis: Linear Mixed Models (LMM) compared cortical thickness (CT) between healthy controls and "pure" (non-comorbid) diagnostic groups.
- Longitudinal Tracking: Analysis of follow-up data at age 12 to observe structural changes over time.
- Genomics: A Genome-Wide Association Study (GWAS) on 4,716 unrelated youth of European ancestry to identify genetic variants linked to brain changes.
- Cell Type Specificity Analysis (CTSA): Mapping genetic findings to specific brain cell types using single-cell RNA sequencing data.
Key Insights
The "Thicker Cortex" Phenomenon
While increased brain volume is often associated with better function, this study found that children with internalizing and externalizing disorders exhibited thicker cortices in specific regions compared to healthy controls. This suggests a failure or delay in the healthy synaptic pruning and myelination processes that normally thin the cortex during puberty.
Shared Cognitive Control Hubs
Both internalizing and externalizing disorders shared thicker CT in the left pars opercularis and caudal middle frontal gyrus. These regions are critical for inhibitory control and emotional regulation. Notably, thicker cortex in these areas correlated with lower cognitive performance across the cohort, reinforcing the idea that delayed thinning may impede neural efficiency.
Sensory vs. Motor Divergence
Despite shared traits, unique structural markers emerged:
- Externalizing Disorders: Uniquely affected primary sensory areas, including the auditory and somatosensory cortex.
- Internalizing Disorders: Uniquely affected the primary motor cortex and higher-order visual association areas.
Longitudinal Stability and Thinning
Only the externalizing group showed a persistent CT increase at the 2-year follow-up. Furthermore, this group demonstrated decelerated cortical thinning from ages 10 to 12, indicating a distinct developmental trajectory where the brain remains "thicker" for longer compared to typically developing peers.
Biological and Genetic Mechanisms
The genetic analysis identified eight regions with significant genetic associations. For instance, a locus near the IGF2 gene—previously linked to anxiety—was associated with thickness in the superior temporal sulcus.
At the cellular level, microglia (the brain's immune cells) were associated with CT changes in both internalizing and externalizing regions. Because microglia are responsible for synaptic pruning, this points to a shared biological mechanism for neurodevelopmental risk. Conversely, dopaminergic, glutamatergic, and GABAergic cells were uniquely associated with regions specific to externalizing disorders, suggesting that neurotransmitter-specific pathways drive the unique aspects of externalizing phenotypes.
Discussion and Limitations
These findings suggest that while internalizing and externalizing disorders share a background of morphological variance, they are distinguished by unique genetic and cellular influences.
Several limitations remain:
- Thought Disorders: The smaller sample size (n=347) likely limited the power to detect significant CT alterations in this group.
- Longitudinal Scope: A two-year window is brief relative to the full span of adolescence. Continued tracking is necessary to determine if these "thick" regions eventually undergo excessive thinning in adulthood.
- Ancestry Bias: The genetic analysis was restricted to European-ancestry youth to ensure statistical validity, meaning these specific genetic results may not yet be generalizable to all populations.
By identifying these structural and genetic markers, this research provides a biological foundation for transdiagnostic psychiatric taxonomy and highlights potential targets for early intervention.
Summary updated