This article develops and tests a theoretical cognitive-affective process model of the hostile media effect (HME). To explain the HME, scholars have mainly focused on cognitive involvement, that is, the extent to which an issue is of personal importance. In addition, we introduce the notion of affective involvement and hypothesize three distinct routes responsible for a HME: a cognitive, an affective, and a cognitive-affective route. Simultaneously collected representative survey data from the United States, Norway, and France employing country-invariant measures provide clear evidence that the three routes each and independently drive the HME. Theoretical and methodological implications of these findings are discussed.