Abstract: This study aims to analyze the mechanisms of prebunking’s impact on two specific dimensions (i.e., individual perception and behavioral intention) and compare the efficacy of different prebunking strategies. The hypotheses were tested in a double-blind experimental design, wherein 182 participants were randomly assigned to a fact-based inoculation or a logic-based inoculation or a control group. Specifically, we use one-way ANOVA to test the mechanism of the effect of the pre-intervention and use multiple comparison analysis to test the effect of different interventions. The results indicate that prebunking have a significant impact on individual perceptions and behavioral intentions. In the perception dimension, prebunking substantially reduces individuals' perceived threat, perceived consensus, and perceived credibility of health misinformation. In the behavioral intention dimension, while prebunking effectively reduces individuals' sharing likelihood of health misinformation, it has little impact on their willingness to vaccinate against COVID-19. Furthermore, no significant difference was observed between the group of fact-based and logic-based inoculations. This study broadens the application scenarios and mechanisms of prebunking, offering evidence for assessing its feasibility and efficacy in the governance of health misinformation.

Key words: Health misinformation, Prebunking, Infodemics, Individual perceptions, Behavioral intention, Inoculation theory