RDoC and Psychopathology among Youth: Misplaced Assumptions and an Agenda for Future Research

ABSTRACT

Now over 10 years old, the Research Domain Criteria (RDoC) has gained impressive traction in the adult psychopathology literature, but enthusiasm among child and adolescent psychopathologists lags somewhat behind. We consider possible reasons why RDoC has not been embraced fully in the child and adolescent literatures. We emphasize common, interrelated, and sometimes outdated assumptions that impede scientific progress that RDoC could facilitate. Traditionally, child and adolescent psychopathologists have used behavioral syndromes as gold standards against which biological markers are validated, even though behavioral syndromes are often measured with less precision; sought to identify large main effects of single biological functions on single behavioral syndromes, thereby ignoring (even if implicitly) the overwhelming etiological complexity of psychopathology; expected 1:1 correspondencies between biological functions and behaviors, despite evidence that core biological systems subserving behavior are functionally interdependent (i.e., modulate one another); and failed to consider neurobiological mechanisms of homotypic and heterotypic comorbidity and continuity. Using examples from our work, we show how a developmental, RDoC-informed approach to externalizing behavior enriches our understanding of psychopathology. We also provide an agenda for future research, which includes calls to (1) adopt neural-systems-first approaches over disorder-first approaches when studying psychopathology, (2) eschew biological reductionism by integrating environmental risk mediators into our etiopathophysiological models, (3) integrate neural vulnerabilities into the empirical latent structure of psychopathology, and (4) replace null hypothesis significance testing with computational approaches that accommodate etiological complexity by evaluating functional dependencies among RDoC constructs, including positive valence systems (approach), negative valence systems (avoidance), and arousal/regulatory systems (self-regulation).

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Implications for and applications to eventual diagnostic systems maybe decades down the road (e.g., Cuthbert, 2015).

² We are not suggesting that RDoC is without limitations. Such limitations are well documented in this special issue and elsewhere; they need not be recapped here.

³ Such progression is specific to the hyperactive (HI) and combined (C) presentations of ADHD – not the purely inattentive (IN) presentation (e.g., Ahmad & S. P. Hinshaw, 2017). This observation, combined with differentiating factors at neural, behavioral, and treatment-response levels of analysis, suggests a separate etiopathophysiology for ADHD-I (see e.g., Adams et al., 2008; Fair et al., 2013; Stein et al., 2003).

⁴ In contrast, amygdala reactivity to fear-eliciting events is high in internalizing disorders but low in externalizing disorders, and is therefore differentiating (see Beauchaine & Constantino, 2017; Beauchaine & Tackett, 2020).

⁵ New studies call into question the psychometrics of many bifactor models (Burns et al., 2020; Eid et al., 2017). Nevertheless, such models are of considerable interest to psychopathologists, so we discuss them herein. Importantly, measurement issues in bifactor model-fitting do not invalidate the second-order factor structure of psychopathology (i.e., internalizing, externalizing, thought problems).

Additional information

Funding

This work was supported by the National Center for Advancing Translational Sciences [UL1TR002733]; National Institute of Dental and Craniofacial Research [UH2DE025980].