DESCRIPTION (provided by applicant): Increased error-related brain activity, as indexed by the error-related negativity (ERN) component of the event-related brain potential, has been proposed as an inherited risk factor for anxiety. Both the ERN and anxiety symptoms increase dramatically during adolescence-our data suggest this is especially true between the ages of 11 and 14. The current study aims to examine whether increased ERN is a mechanism of risk for the development of increased anxiety, and whether the ERN is a modifiable biomarker of risk among a large (N=600) and prospective sample of adolescents recruited at Stony Brook University and San Diego State University. At each site, 300 adolescents aged 11 to 14 will be recruited. Based on our pilot data, we examine whether ERN can be altered using adaptive attentional bias modification (AABM). AABM is a computer-based approach to train individuals to attend away from threat in their environments, and our pilot data suggests AABM reduces the ERN. High-ERN participants (N=300) will be randomized to AABM (N=150) or a control condition (CC; N=150). We target increased ERN precisely because this is the proposed neural biomarker we wish to reduce with AABM. We will examine the impact of AABM on ERN, and the relationship between ERN and anxiety, both in the short term (i.e., after AABM/CC) and at two year follow-up. All participants (i.e., even low-ERN participants not randomized to AABM/CC) will return to the lab to have ERN and anxiety assessed at two year follow-up. Across the entire sample, we predict that baseline ERN will relate to anxiety; we expect anxiety to increase from baseline to two year follow-up; we predict that a larger ERN at baseline will predict increased anxiety prospectively, and that changes in anxiety will relate to changes in ERN (Aim 1). Crucially, we predict that AABM will reduce the ERN (Aim 2), and that the degree to which ERN is reduced will relate to AABM-related reductions in anxiety (Aim 3). Overall, we will test whether the ERN is a modifiable biomarker of risk. Although all adolescents randomized to AABM will be characterized by a relatively large ERN, we will also be able to examine whether a larger ERN at baseline will predict change in anxiety following AABM. Finally, we conduct exploratory analyses to examine the impact of gender on all hypotheses, and determine whether age moderates the strength of the proposed relationships tested in our primary aims-which would suggest critical information about the best timing to assess and modify ERN to inform and impact anxious trajectories of risk.