Fighting misinformation on social media: The roles of evidence type and presentation mode

An online experiment was conducted to examine the impact of evidence type (evidence type: statistical vs. non-statistical) and presentation mode (textual-only vs. pictorial-only vs. textual-plus-pictorial) on individuals’ responses to corrective information about COVID-19 on social media. The results indicated that corrective information backed by non-statistical evidence (in contrast to statistical evidence) enhanced message elaboration, which in turn led to greater misperception reduction, higher ratings of message believability, and stronger intention to engage in viral behaviors (e.g., sharing, liking, and commenting on the post). Compared to the textual-only modality and the textual-plus-pictorial modality, the pictorial-only modality induced a significantly lower level of message elaboration, which subsequently resulted in lower message believability and less viral behavioral intention. The theoretical and practical implications are discussed.