Modeling relations between event-related potential factors and broader versus narrower dimensions of externalizing psychopathology.

Abstract

The organization of the Hierarchical Taxonomy of Psychopathology (HiTOP) model provides unique opportunities to evaluate whether neural risk measures operate as indicators of broader latent liabilities (e.g., externalizing proneness) or narrower expressions (e.g., antisociality and alcohol abuse). Following this approach, the current study recruited a sample of 182 participants (54% female) who completed measures of externalizing psychopathology (also internalizing) and associated traits. Participants also completed three tasks (Flanker-No Threat, Flanker-Threat, and Go/No-Go tasks) with event-related potential (ERP) measurement. Three variants of two research domain criteria (RDoC)-based neurophysiological indicators—P3 and error-related negativity (ERN)—were extracted from these tasks and used to model two latent ERP factors. Scores on these two ERP factors independently predicted externalizing factor scores when accounting for their covariance with sex—suggesting distinct neural processes contributing to the broad externalizing factor. No predictive relation with the broad internalizing factor was found for either ERP factor. Analyses at the finer-grained level revealed no unique predictive relations of either ERP factor with any specific externalizing symptom variable when accounting for the broad externalizing factor, indicating that ERN and P3 index general liability for problems in this spectrum. Overall, this study provides new insights about neural processes in externalizing psychopathology at broader and narrower levels of the HiTOP hierarchy.

To fully leverage fine-grained clinical phenomena, we have to think beyond DSM-based concepts and the presumption of diagnostic kinds.

“…However, this implies that mechanisms are nested within—and, therefore, relevant only to—the most fine-grained symptoms, such that higher-order dimensions are just convenient summaries for data reduction rather than potentially substantive constructs in and of themselves. Articles in this special section (e.g., Pasion et al, 2023; Rowe et al, 2023) demonstrate that psychopathological processes and mechanisms are themselves hierarchical. Complex systems theory (Olthof et al, 2023) takes this premise a step further in looking at dynamics theorized to represent general principles of clinical change, which suggests further avenues for hierarchical integration.…”

“…As noted, fine-grained indicators allow for the modeling of more nuanced higher levels and more detailed lower levels, thereby depicting the overall metastructure more clearly. This would, in turn, facilitate clearer links to other lenses through which to study psychopathology (e.g., neurobiological, interpersonal, sociopolitical) and allow for more sophisticated multimethod theories and models (as illustrated by Pasion et al, 2023). Compared to dimensions based on crude diagnostic indicators, dimensions based on symptoms are also considerably more informative for scale development—thus promoting the translation of findings regarding optimal levels of hierarchy for different purposes into clinical and research applications (Levin-Aspenson & Zimmerman, 2022).…”

“…This section is a remarkable effort led by three researchers who approach psychopathology in different, but complementary ways. Fittingly, the articles comprising this section provide distinct perspectives on the target topic, including genetics (Waldman & Poore, 2023), neuroanatomy (Rowe et al, 2023), psychophysiology (Pasion et al, 2023; Stange et al, 2023), epidemiology (Karlovich et al, 2023), and intervention (Olthof et al, 2023). I carry this stance of pluralistic collegiality into my commentary as well.…”

Studying fine-grained elements of psychopathology to advance mental health science.

“…Several respondents also noted a critical need to integrate various frameworks for psychopathology such as HiTOP and RDoC to ensure there were no redundancies between them (e.g., Michelini et al, 2021) and to allow for rigorous measurement and testing of constructs advanced by each system. Similarly, the importance of integrating the multimethod approaches to the measurement between frameworks was clear, although the small intercorrelations among various measurement modalities pose a challenge here—for example, self-report symptoms correlated more highly with other self-report symptoms than they do electroencephalogram (EEG), magnetic resonance imaging (MRI), or genetics, and vice versa (see Stange et al, 2023, and Pasion et al, 2023, for examples of such studies). The commentary by Joyner and Perkins (2023) discusses this topic in greater detail as well.…”