This study examined interparental conflict as a linear and curvilinear predictor of subsequent changes in adolescents' negative emotional reactivity and cortisol functioning during family conflict and, in turn, their psychological difficulties. In addition, adolescents' negative emotional reactivity and cortisol functioning during family conflict were examined as subsequent predictors of their psychological difficulties. Participants included 258 adolescents (52% girls) and their parents and teachers who participated in 3 annual measurement occasions. Adolescents were 13 years old on average (standard deviation [SD] = 0.57) at the first measurement occasion, were generally from middle- and working-class backgrounds, and identified mostly as White (e.g., 74%). The results of latent-difference score analyses indicated that a multimethod and multiinformant assessment of interparental conflict linearly predicted subsequent changes in observational ratings of adolescent emotional reactivity and their overall cortisol output in response to family conflict over a 1-year period. These changes, in turn, predicted increases in multiinformant reports of adolescent psychological problems over a 2-year period. However, the linear association in the first link in the cascade was qualified by the quadratic effects of interparental conflict as a predictor. Consistent with risk-saturation models, the relatively strong associations among interparental conflict and youth emotional reactivity and cortisol output at mild and moderate exposure to conflict weakened as exposure to conflict reached higher levels.