PROJECT SUMMARY Comorbidity in the anxiety disorders is common and strongly associated with poor outcomes, lower rates of remission, increased disability and higher rates of relapse. Yet despite its clinical relevance, clinicians are unsure how to treat comorbid anxiety patients, in part because it is unknown whether comorbid cases are marked by distinct pathophysiology, or whether they should simply be conceptualized as the 'sum of their parts' (i.e., multiple, separate clinical entities). Characterizing comorbid individuals as having multiple, independent, disease processes occludes understanding of synergistic, pathophysiological processes that may uniquely characterize highly comorbid patients. The current project focuses on a neural profile that corresponds to increased comorbidity load across the anxiety disorders and is routed in aberrant brain connectivity. This neural profile is comprised of increased alarm (heightened amygdala) and reduced motivated attention (attenuated late positive potential, LPP; an EEG measure of elaborated stimulus processing) to negative images, and is referred to as HARM-A (Heightened Alarm, Reduced Motivated Attention). Controlling for separate effects of alarm and motivated attention, preliminary data suggest that higher HARM-A is associated with (a) greater internalizing psychopathology; (b) higher rates of past comorbidity (controlling for current comorbidity) and (c) increased dysphoria 12-24 months later (controlling for baseline dysphoria). Those with higher HARM-A also showed aberrant connectivity between key nodes involved in threat detection and appraisal (amygdala-anterior cingulate cortex), as well as a stronger link between stress and negative affect. Therefore, HARM-A might underlie the worse outcome in comorbid anxiety cases, and may do so by increasing risk for negative affect generation following stressful events. The current project extends this preliminary work by examining negative emotion processing in 180 individuals, recruited to insure dimensionality on current and past internalizing symptoms. Participants will undergo 3 multi-level assessments (fMRI, EEG, clinical interview, self-report measures) over 24 months; at each assessment, participants will also complete 10 days of experience sampling assessments of stressful events and negative affect. The project will test a bidirectional model of HARM-A, in which HARM-A predicts increased future comorbidity load and higher scores on latent, transdiagnostic, internalizing psychopathology, which will in turn predict increased HARM-A (i.e., a mutually reinforcing “spiral”). It will also assess the neurocircuitry that supports higher HARM-A and will use experience sampling data to test whether HARM-A predicts stronger linkage between stress exposure and subsequent negative affect. Finding that HARM-A is prospectively and reciprocally associated with worse internalizing psychopathology, delineating its neurocircuitry and elucidating its role in th...