White matter microstructure correlates of general and specific second-order factors of psychopathology

Kendra E. Hinton, Benjamin B. Lahey, Victoria Villalta‐Gil, Francisco A. C. Meyer, Leah Burgess, Laura K. Chodes, Brooks Applegate, Carol A. Van Hulle, Bennett A. Landman, David H. Zald

TL;DR

This study moves beyond traditional "case-control" psychiatric research by using a bifactor model to identify white matter correlates of broad, transdiagnostic dimensions of psychopathology. The researchers found that general, internalizing, and externalizing factors of mental health are associated with distinct structural properties in major brain tracts, suggesting that biological markers of mental illness often cut across traditional diagnostic boundaries.

What Problem Does This Paper Address?

Traditionally, clinical neuroscience has relied on case-control designs—comparing "depressed" patients to "healthy" controls. However, this approach struggles with two major realities: comorbidity (people often have multiple diagnoses) and heterogeneity (two people with the same diagnosis can have very different symptoms).

Recent psychological research suggests that mental health disorders are better organized into a hierarchy. At the top is a "general factor" (or p-factor) representing a person's overall liability to any mental illness. Below that are specific factors like Internalizing (e.g., anxiety, depression) and Externalizing (e.g., impulsivity, substance use). This paper addresses the gap in understanding how these latent, transdiagnostic dimensions—rather than just individual categories—relate to the brain's white matter microstructure.

Methods

The researchers employed Diffusion Weighted Imaging (DWI) to examine the integrity of white matter tracts in 410 young adults. They focused on three key metrics:

  • Fractional Anisotropy (FA): A broad measure of white matter "integrity" or organization.
  • Axial Diffusivity (AD): Often associated with axonal integrity.
  • Radial Diffusivity (RD): Often associated with myelin health.

The core of their analysis was a bifactor model. This statistical approach allowed them to "disentangle" the shared variance (the general factor) from the unique variance of internalizing and externalizing symptoms. They used multiple regressions within structural equation models (SEM) to link these latent factors to specific brain tracts identified via the JHU ICBM-DTI atlas.

Datasets

The study utilized data from the Tennessee Twin Study (TTS), specifically "Wave 2," which focused on young adults (ages 23–31).

  • Sample Size: 410 participants (including 187 twin pairs).
  • Clinical Assessment: The Young Adult Diagnostic Interview for Children (YA-DISC) was used to assess symptoms across a wide range of DSM-IV categories.
  • Sampling Strategy: The researchers oversampled for individuals at high risk for psychopathology to ensure a robust range of symptom severity.

Key Insights and Ideas

  • Biological Separability: The most significant "aha" moment is that each second-order factor had distinct white matter correlates. This provides biological evidence that these latent dimensions are not just statistical constructs but represent separable neurobiological patterns.
  • The "p-factor" Paradox: Higher scores on the general factor were associated with higher FA in the body of the corpus callosum. While higher FA is usually seen as "better," the authors hypothesize this might reflect an accelerated developmental trajectory—where individuals with high psychopathology reach peak white matter development earlier, followed by a steeper decline later in life.
  • Axial Diffusivity is Key: Most findings for specific internalizing and externalizing factors were found in AD rather than FA or RD. This suggests that axonal-specific properties may be more sensitive markers for these dimensions than overall anisotropy.
  • Transdiagnostic Specificity: The externalizing factor showed the most widespread correlates (splenium, corona radiata, internal capsule), suggesting that broad impulsivity/disinhibition has a particularly strong signature in the brain's communication network.

Key Results

  • General Factor: Positively correlated with FA in the body of the corpus callosum ($\beta = .25, p = .001$).
  • Specific Internalizing: Negatively correlated with AD in the fornix ($\beta = -.18, p = .003$), a tract critical for emotional memory.
  • Specific Externalizing: Negatively correlated with AD in several tracts, including the splenium of the corpus callosum ($\beta = -.22$), anterior corona radiata ($\beta = -.28$), internal capsule ($\beta = -.23$), and sagittal stratum ($\beta = -.33$).
  • Robustness: These findings held up even after controlling for sex, head size (TICV), socioeconomic status, and current substance use.

Limitations and Future Work

  • Cerebellum Coverage: Due to scanning parameters, the study had limited coverage of the cerebellum, preventing them from replicating other recent studies that found "p-factor" correlates in that region.
  • Exclusion of Psychosis: The sample excluded individuals with psychotic disorders. Future work should include a "thought disorder" factor to see if it has unique white matter signatures.
  • Directionality: As a cross-sectional study, it cannot determine if white matter differences cause psychopathology or are a consequence of it (e.g., the long-term effects of substance use).
  • Developmental Context: The "accelerated development" hypothesis for the corpus callosum findings needs to be tested with longitudinal data spanning from childhood through middle age.

Summary updated

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 that explains the common variance among all disorders and specific second-order externalizing and internalizing factors. Nevertheless, most existing studies on the neural correlates of psychopathology employ case-control designs that treat diagnoses as independent categories, ignoring the highly correlated nature of psychopathology. Thus, for instance, although perturbations in white matter microstructure have been identified across a range of mental disorders, nearly all such studies used case-control designs, leaving it unclear whether observed relations reflect disorder-specific characteristics or transdiagnostic associations. Using a representative sample of 410 young adult twins oversampled for psychopathology risk, we tested the hypothesis that some previously observed relations between white matter microstructure properties in major tracts and specific disorders are related to second-order factors of psychopathology. We examined fractional anisotropy (FA), radial diffusivity (RD), and axial diffusivity (AD). White matter correlates of all second-order factors were identified after controlling for multiple statistical tests, including the general factor (FA in the body of the corpus callosum), specific internalizing (AD in the fornix), and specific externalizing (AD in the splenium of the corpus callosum, sagittal stratum, anterior corona radiata, and internal capsule). These findings suggest that some features of white matter within specific tracts may be transdiagnostically associated multiple forms of psychopathology through second-order factors of psychopathology rather with than individual mental disorders.
BibTeX
@inproceedings{hinton2019white,
  author = {Kendra E. Hinton and Benjamin B. Lahey and Victoria Villalta‐Gil and Francisco A. C. Meyer and Leah Burgess and Laura K. Chodes and Brooks Applegate and Carol A. Van Hulle and Bennett A. Landman and David H. Zald},
  title = {{White matter microstructure correlates of general and specific second-order factors of psychopathology}},
  booktitle = {NeuroImage},
  year = {2019},
  doi = {10.1016/j.nicl.2019.101705},
}

Loading related content...

Figure