Abstract: The Mediating Role of Perceived Conflict in Processing Conflicting Health Information

◆ Weijia Shi, University of Texas at Austin
◆ Rebekah Nagler, University of Minnesota

Exposure to conflicting health information has been found to produce undesirable affective and cognitive outcomes, including increased distress and frustration, heightened confusion and uncertainty about health issues, and lower perceived credibility of journalists and scientists (Chang, 2015; Nagler et al., 2019). To mitigate such adverse effects, effective communication and public health interventions are needed. Yet, little is known about how conflicting health information is processed; understanding such processing, particularly by identifying any intermediate variables between conflicting exposure and its subsequent effects, would yield valuable insights for intervention strategies. This study focused on one potential intermediate variable, perceived conflict, and examined whether affective (fear, anger, and surprise) and cognitive outcomes (confusion, ambivalence, backlash, and media skepticism) of exposure to conflicting health messages are mediated by perceived conflict. In Fall 2020, we conducted two between-subjects experimental studies in the context of coffee consumption and with a sample of the U.S. general population recruited through Prolific Academic. Informed by the manipulation-of-mediator approach (Pirlott & MacKinnon, 2015), we examined the independent variable–mediator relationship in Study 1 and mediator–dependent variable relationship in Study 2. Specifically, Study 1 (N = 649) tested whether people did indeed perceive conflict after exposure to conflicting health messages. Participants were randomly assigned to one of four message conditions: 1) one news story reporting the potential health benefits of drinking coffee, 2) one news story reporting the potential risks of drinking coffee, 3) one news story about general information about coffee, and 4) two distinct conflicting news stories that report potential benefits and risks of drinking coffee, respectively. Study 2 (N = 846) tested whether perceived conflict was associated with adverse affective and cognitive outcomes. Participants were randomly assigned to one of five message conditions—the same four conditions included in Study 1, plus one additional condition of a single news story highlighting contradictory findings about coffee consumption (in an attempt to produce heightened levels of perceived conflict). All stimuli were adapted from real news stories about coffee. As expected, in Study 1 participants in the conflicting health messages condition reported significantly greater perceived conflict than those in the other three comparison conditions. Study 2 results showed that perceived conflict induced by exposure to conflicting health messages was overall associated with greater confusion, ambivalence, and anger, but not backlash, media skepticism, fear, or surprise. Post-hoc mediation analyses using the PROCESS macro provided additional evidence that perceived conflict mediated the exposure to conflicting messages–outcome relationships, except for the outcomes of media skepticism (partial mediation) and fear (no mediation). To the extent that perceived conflict serves as a mediator in processing conflicting health information, interventions that can successfully help the public to resolve such conflict have the potential to offset adverse affective and cognitive effects. For example, perhaps journalists could help the public to recognize that conflicting study findings may be caused by different research designs, samples, and/or measures. Additionally, the challenges of using the manipulation-of-mediator approach to testing mediation in processing conflicting health information will be discussed.