The hierarchical taxonomy of psychopathology and the search for neurobiological substrates of mental illness: A systematic review and roadmap for future research

Abstract

Understanding the neurobiological mechanisms involved in psychopathology has been hindered by the limitations of categorical nosologies. The Hierarchical Taxonomy of Psychopathology (HiTOP) is an alternative dimensional system for characterizing psychopathology, derived from quantitative studies of covariation among diagnoses and symptoms. HiTOP provides more promising targets for clinical neuroscience than traditional psychiatric diagnoses and can facilitate cumulative integration of existing research. We systematically reviewed 164 human neuroimaging studies with sample sizes of 194 or greater that have investigated dimensions of psychopathology classified within HiTOP. Replicated results were identified for constructs at five different levels of the hierarchy, including the overarching p-factor, the externalizing superspectrum, the thought disorder and internalizing spectra, the distress subfactor, and the depression symptom dimension. Our review highlights the potential of dimensional clinical neuroscience research and the usefulness of HiTOP while also suggesting limitations of existing work in this relatively young field. We discuss how HiTOP can be integrated synergistically with neuroscience-oriented, transdiagnostic frameworks developed by the National Institutes of Health, including the Research Domain Criteria, Addictions Neuroclinical Assessment, and the National Institute on Drug Abuse's Phenotyping Assessment Battery, and how researchers can use HiTOP to accelerate clinical neuroscience research in humans and other species. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Figure 1.

The Hierarchical Taxonomy of Psychopathology (HiTOP). Recent efforts by an international consortium of researchers have produced this dimensional system for organizing psychiatric symptoms (Kotov et al., 2017, 2020, 2021, 2022; Krueger et al., 2021; Ringwald et al., 2021; Watson et al., 2022). Figure depicts a simplified version of the HiTOP working model (top) and its approximate correspondence to DSM constructs (bottom). Constructs toward the top of HiTOP are broader and more general, whereas those at lower levels are narrower and more specific. For specific constructs at the lower HiTOP levels see (Kotov et al., 2017, 2022). HiTOP is a work in progress and will be updated on the basis of new data. Dashed lines indicate provisional elements requiring more study. Categorical diagnoses from DSM are not HiTOP constructs but are included to allow mapping of existing nosologies onto HiTOP, and those with the most prominent cross-loadings are listed in multiple places. Minus sign indicates negative association between histrionic personality and the detachment spectrum.

Figure 2.

Flow chart of study selection process for review.

Figure 3.

Frequency of studies in review investigating different HiTOP constructs, separated within each spectrum by different levels of the hierarchy. The “specific constructs” category includes constructs from the levels “Empirical Syndromes” and “Symptom Components and Maladaptive Traits.” Studies of mania were double-counted in the subfactor category, for both internalizing and thought disorder. EXT = externalizing.

Figure 4.

Initial crosswalk between the Hierarchical Taxonomy of Psychopathology (HiTOP), Research Domain Criteria (RDoC), Addictions Neuroclinical Assessment (ANA), and the National Institute on Drug Addiction’s Phenotyping Battery (NIDA PhAB). Depicted links between HiTOP and RDoC are the strongest and most consistent associations according to a review of empirical studies (Michelini et al., 2021). Less prominent links are not shown. Due to paucity of relevant studies, it was not possible to link the recently introduced RDoC Sensorimotor domain to any HiTOP spectra, nor the HiTOP Somatoform spectrum to any RDoC domains. Negative associations are presented in red and positive associations in blue. Double arrows indicate that within the RDoC domain some constructs show positive links, whereas others show negative links to the HiTOP spectrum (for example, internalizing was associated positively with arousal and negatively with sleep constructs from the arousal/regulation domain). Associations between RDoC and ANA or NIDA PhAB domains are shown with symbols for approximate equality. Asterisk indicates a domain in NIDA PhAB but not ANA. (NIDA PhAB domains of metacognition and interoception have not been linked to RDoC and are not depicted.) Reprinted with permission from (Kotov et al., 2021).

Figure 5.

Using the Hierarchical Taxonomy of Psychopathology (HiTOP) in clinical neuroscience. Step 1 involves selecting clinical phenotypes from HiTOP to study (figure depicts a simplified model; for full list of constructs see Kotov et al., 2017, 2022). Optimal HiTOP measurement uses fully dimensional instruments without skip-outs and, if possible, with multiple assessment modalities (e.g., self- and informant reports and clinical interviews). Step 2 depicts a sampling design appropriate for HiTOP-based research, which involves sampling from transdiagnostic patient populations or the general population, rather than a case-control design. However, researchers may oversample participants manifesting or at high risk for the problems of interest. Step 3 depicts testing associations between HiTOP phenotypes and neurobiological variables, ideally examining nested constructs at multiple levels of the hierarchy and examining constructs from multiple spectra to assess discriminant validity. Adapted with permission from (Latzman et al., 2020).