Symposia
Neuroscience
Stephanie N. DeCross, M.A. (she/her/hers)
Harvard University
Cambridge, Massachusetts
Sahana Kribakaran, MA (she/her/hers)
MD/PhD Candidate
Yale University
New Haven, Connecticut
Paola Odriozola, PhD
PhD student
Yale University
New Haven, Connecticut
Margaret Sheridan, PhD (she/her/hers)
Associate Professor
University of North Carolina at Chapel Hill
Chapel Hill, North Carolina
Nim Tottenham, PhD (she/her/hers)
Professor
Columbia University
New York, New York
Dylan Gee, Ph.D.
Assistant Professor
Yale University
New Haven, Connecticut
Katie McLaughlin, PhD (she/her/hers)
Professor
Harvard University
Cambridge, Massachusetts
Threat reversal learning is a potential mechanism linking childhood trauma (CT) to psychopathology. Threat reversal learning describes the ability to flexibly update associations of stimuli with threat and safety (a process relevant to both psychopathology etiology and treatment) and is vastly understudied, particularly in developmental populations. This study aims to be the first to characterize how CT may alter the neural underpinnings of threat reversal learning in youth in ways relevant to mental health.
A demographically diverse sample of 100 participants aged 9-19 was recruited by targeting schools, after-school and prevention programs, adoption programs, food banks, shelters, parenting programs, medical clinics, and the general community. Half the participants were exposed to CT, including physical abuse, sexual abuse, and exposure to domestic violence. Participants completed measures of clinical symptoms and then underwent a Pavlovian threat reversal learning task during an fMRI scan. First, one previously-neutral stimulus was paired with an aversive stimulus and thereby predicted threat (CS+) and one was not and predicted safety (CS-). Then, contingencies were reversed, such that the old safety cue became the new threat cue (New CS+) and the old threat cue became the new safety cue (New CS-). Whole-brain and regions-of-interest analyses were used and mixed effects models were conducted to examine patterns of neural activation, and nonparametric mediation models with 10,000 simulations were run to test whether these neural measures mediated the association between CT and transdiagnostic psychopathology.
In youth, canonical threat-related regions were activated during threat reversal learning, with youth with CT exposure displaying reduced activation in right amygdala, hippocampus, posterior parahippocampal gyrus (PHG), and ventromedial prefrontal cortex (vmPFC) during New CS+ >New CS-, reflecting blunted discrimination between cues with reversed contingencies. Critically, blunted discrimination in hippocampus and PHG mediated the link between CT and generalized anxiety and panic (ps≤.02), and blunted discrimination in right amygdala mediated the association between CT and panic (p=.02). The association between CT and PTSD was mediated by blunted discrimination in vmPFC (p=.046).
Taken together, alterations in threat reversal learning may reflect decreased ability to flexibly update threat-safety associations and may be a key mechanism underlying the link between CT and transdiagnostic psychopathology.