Abstract
Increasing data indicate that prevalent forms of psychopathology can be organized into second-order dimensions based on their correlations, including a general factor of psychopathology (the "p-factor") that explains common variance among all disorders, and specific second-order externalizing and internalizing factors. Despite this organization and high levels of comorbidity, most existing studies on the neural correlates of psychopathology employ case-control designs that treat diagnoses as independent categories. This study tested the hypothesis that white matter microstructure properties in major tracts are related to these second-order factors. Using a representative community twin sample of 410 young adults, researchers examined fractional anisotropy (FA), radial diffusivity (RD), and axial diffusivity (AD). Findings revealed that features of white matter within specific tracts are associated with broad transdiagnostic dimensions rather than individual diagnostic categories, including the general factor (FA in the body of the corpus callosum), specific internalizing (AD in the fornix), and specific externalizing (AD in multiple tracts including the splenium and internal capsule).
Introduction
In clinical neuroscience, the traditional "case-control" design—comparing individuals with a specific diagnosis to healthy controls—has long been the gold standard. However, this approach often fails to account for two pervasive realities in mental health: comorbidity (the tendency for individuals to meet criteria for multiple disorders) and heterogeneity (the vast differences in symptom presentation within a single diagnosis).
Recent psychological research suggests that mental health disorders are better understood through a hierarchical structure. At the apex sits a "general factor" of psychopathology, representing a broad liability to mental illness. Below this are specific factors: Internalizing (encompassing anxiety and mood disorders) and Externalizing (encompassing impulsivity and substance use disorders). This study investigates how these latent, transdiagnostic dimensions relate to the brain's white matter microstructure, moving beyond diagnostic silos to find the biological signatures of mental health's broad dimensions.
Methodology and Data
The Tennessee Twin Study (TTS)
The researchers utilized data from Wave 2 of the Tennessee Twin Study, focusing on a sample of 410 young adults (ages 23–31), including 187 twin pairs. To ensure a robust range of symptom severity, the study oversampled individuals at high risk for psychopathology. Clinical assessments were conducted using the Young Adult Diagnostic Interview for Children (YA-DISC), covering a wide array of DSM-IV categories.
Imaging and Statistical Modeling
The study employed Diffusion Weighted Imaging (DWI) to measure white matter integrity across three primary metrics:
- Fractional Anisotropy (FA): A measure of the directional organization of water diffusion, often used as a proxy for overall white matter "integrity."
- Axial Diffusivity (AD): The rate of diffusion parallel to the axon, often linked to axonal integrity.
- Radial Diffusivity (RD): The rate of diffusion perpendicular to the axon, often associated with myelin health.
To analyze the data, the team used a bifactor model within a Structural Equation Modeling (SEM) framework. This allowed them to statistically isolate the shared variance of the general factor from the unique variance attributable specifically to internalizing or externalizing symptoms. These factors were then correlated with brain tracts identified via the JHU ICBM-DTI atlas.
Key Findings
The study identified distinct white matter correlates for each of the second-order factors, suggesting that these statistical constructs have unique neurobiological foundations.
1. The General Factor (p-factor)
Higher scores on the general factor were positively correlated with FA in the body of the corpus callosum ($\beta = .25, p = .001$). Interestingly, while higher FA is typically associated with better white matter health, the authors suggest this may reflect an accelerated developmental trajectory. In this model, individuals with higher psychopathology might reach peak white matter development earlier, potentially leading to a steeper decline later in life.
2. Specific Internalizing Factor
The internalizing factor was uniquely associated with lower AD in the fornix ($\beta = -.18, p = .003$). The fornix is a critical tract for emotional memory and regulation, and this finding suggests that axonal-specific properties in this region are specifically tied to internalizing symptoms like anxiety and depression, independent of general psychopathology.
3. Specific Externalizing Factor
The externalizing factor showed the most widespread structural correlates. It was negatively associated with AD in several tracts, including:
- Splenium of the corpus callosum ($\beta = -.22$)
- Anterior corona radiata ($\beta = -.28$)
- Internal capsule ($\beta = -.23$)
- Sagittal stratum ($\beta = -.33$)
These findings suggest that broad disinhibition and impulsivity are linked to reduced axonal integrity across the brain's communication networks.
Discussion and Implications
A significant takeaway from this research is the sensitivity of Axial Diffusivity (AD). While many studies focus solely on FA, the majority of the specific internalizing and externalizing findings in this study were found in AD. This suggests that axonal-specific properties may be more precise markers for transdiagnostic dimensions than broader measures of anisotropy.
Furthermore, the results provide biological validation for the hierarchical model of psychopathology. By showing that the general, internalizing, and externalizing factors have distinct neural signatures, the study reinforces the idea that these dimensions are not merely statistical artifacts but represent separable biological patterns.
Limitations
Despite its strengths, the study has limitations. The scanning parameters resulted in limited coverage of the cerebellum, a region recently implicated in p-factor research. Additionally, the sample excluded individuals with psychotic disorders, meaning a "thought disorder" factor could not be examined. Finally, as a cross-sectional study, the results cannot determine whether these white matter differences are a cause or a consequence of psychopathology.
Conclusion
By shifting the focus from individual diagnoses to broad dimensions of mental health, this study reveals that the biological markers of psychopathology often cut across traditional diagnostic boundaries. These findings pave the way for a more nuanced, transdiagnostic approach to clinical neuroscience, emphasizing the importance of white matter microstructure in understanding the shared and unique roots of mental illness.
Summary updated