Screen-based sedentary behaviors and emotional disorders are associated with one another in youth. Yet, the direction of the association is unclear, as is whether specific types of screen-based sedentary behaviors and emotional disorder symptoms are more closely linked. This study estimated the bi-directional associations between two types of screen-based sedentary behaviors and four types of self-reported emotional disorder symptoms, and tested whether physical activity buffered these associations in a Los Angeles high school student cohort (N = 2525, baseline Mage = 14.6 years). Participants completed baseline (9th Grade, 2013) and 12-month follow-up (10th grade, 2014) surveys reporting on: television viewing and computer/videogame use (≥4 h/day; yes/no), physical activity (≥60 min/day for ≥5 days/week), and Major Depressive Disorder (MDD), Generalized Anxiety Disorder (GAD), Panic Disorder (PD), and Social Phobia (SP) symptoms (meet/exceed [sub]clinical symptom threshold; yes/no). After adjusting for baseline screen-based sedentary behavior and covariates, students with (sub)clinical baseline MDD and GAD were at increased odds of high computer/videogame use one year later (OR = 1.36[95%CI, 1.07-1.73]; OR = 1.36[95%CI,1.09-1.71], respectively). Baseline SP was marginally related to increased computer/videogame use at follow-up (OR = 1.33[95%CI,1.04-1.69]). Greater baseline computer/videogame use was associated with increased odds of (sub)clinical GAD (OR = 1.54[95%CI,1.23-1.94]) and (sub)clinical SP (OR = 1.64[95%CI 1.27-2.12]) at follow-up; these associations were suppressed among baseline physically active students. Television viewing was unrelated to emotional disorder symptoms and PD was not associated with screen-based sedentary behaviors. Thus, only reciprocal associations between computer/videogame use, SP, and GAD during a one-year period of adolescence were observed. Interventions reducing computer/videogame use and increasing physical activity may improve adolescent emotional health.