Major depressive disorder (MDD) is a leading cause of disability worldwide and has a peak onset during
adolescence. While interventions are moderately effective for many adolescents, 40 to 70% will relapse within 5
years. Further, MDD relapse predicts academic difficulties, risky behaviors, and suicide. Thus, identifying
mechanisms of MDD relapse is critical to clarify intervention targets for this significant public health problem.
During adolescence, social processes and dynamics (especially with peers) are particularly significant,
although it is unclear which social processes are most critical to MDD relapse. The present study focuses on the
role of social communication, a set of mechanisms involving the receiving and delivery of socially relevant
information. Social communication is especially significant to adolescents, as maladaptive social communication
can negatively impact the establishment and maintenance of relationships, thus increasing risk for MDD relapse.
For the proposed study, we will employ an NIMH’s Research Domain Criteria (RDoC) lens and use multiple
measures (behavior, event-related potentials [ERPs], eye tracking) to compare adolescents (aged 14-17 years)
with remitted depression (remMDD; N=200) to healthy controls (HC; N=100) on deficits in several aspects of
social communication, including: (i) processing of nonverbal social information, (ii) processing of
socioemotional feedback, and (iii) digital communication. First, Aim 1 will test whether remMDD adolescents
abnormally process two types of nonverbal social information—facial expressions (as indexed by reduced
accuracy and abnormal ERPs [i.e., the N170]) and hand gesturing behaviors (assessed via eye-tracking).
Second, Aim 2 will test whether remMDD adolescents abnormally process socioemotional feedback (being
accepted versus rejected by same-aged peers), a well-established trigger of adolescent MDD. Specifically, Aim
2 will test whether remMDD adolescents exhibit a reduced Late Positive Potential [LPP]), an ERP indexing
emotional encoding, following positive social feedback from faux peers during a peer evaluation task. Third,
using an innovative smartphone app, Aim 3 will collect multiple indicators of digital communication regarding the
structure of adolescents’ digital social network (i.e., size of the network; frequency of communication) and
sentiment of the communication with their digital social network (i.e., coding sentiment from their texts, social
media posts); allowing us to test whether remMDD adolescents exhibit abnormal digital communication. Last,
we will follow adolescents for 1-year to determine whether processing of nonverbal information, social feedback,
and digital communication predict the escalation of depression symptoms and MDD relapse. Further, supervised
machine learning will explore which social communication deficit(s) predict symptoms or MDD relapse, and
whether these social processes predict outcomes independent of (or interaction with) other established
predictors of relapse (e.g., stressful life events). In summary, the project has the promise to identify social
process that contribute to recurrent depression, which, ultimately, will lead to innovative treatment approaches.