Presumed media influence in health and risk messaging

Abstract

Health communication research has often focused on how features of persuasive health messages can directly influence the intended target audience of the messages. However, scholars examining presumed media influence on human behavior have underscored the need to think about how various audience’s health behavior can be unexpectedly influenced by their exposure to media messages. Two central theoretical frameworks have been used to guide research examining the unintended effects: the third-person effect and the influence of presumed media influence (IPMI). The theoretical explanations for presumed media influence is built on attribution bias, self-enhancement, perceived exposure, perceived relevance, and self-categorization. Even though both the third-person effect and the IPMI share some theoretical foundations, and are historically related, the IPMI has been argued to be better suited to explaining a broader variety of behavioral consequences. One major way that presumed media influence can affect an individual’s health behavior is through the shifting of various types of normative beliefs: descriptive, subjective, injunctive, and personal norms. These beliefs can manifest through normative pressure that is theoretically linked to behavioral intentions. In other words, media have the capability to create the perception that certain behaviors are prevalent, inculcating a normative belief that can lead to the uptake of, or restrain, health behaviors. Scholars examining presumed media influence have since provided empirical support in a number of specific media and behavioral health contexts. Existing findings provide a useful base for health communication practitioners to think about how presumed media influence can be integrated into health campaigns and message design. Despite the proliferation of research in this area, there remains a need for future research to examine these effects in a new media environment, to extend research into a greater number of health outcomes, to incorporate actual behavioral measures, and to ascertain the hypothesized causal chain of events in the model.