When evidence changes: Communicating uncertainty protects against a loss of trust

Abstract

Scientific findings can be overturned when new evidence arises. This study examines how communicating and explaining uncertainty around scientific findings affect trust in the communicator when findings change. In an online experiment ( N = 800, convenience sample), participants read a fictitious statement from a public health authority announcing that there was no link between a new COVID-19 vaccine and heart muscle inflammation. The authority communicated (1) no uncertainty, (2) uncertainty without giving a reason, (3) uncertainty due to imprecision, or (4) uncertainty due to incomplete accounting of patients. Participants were then informed that the authority’s statement was no longer correct as new data showed a link between the vaccine and heart muscle inflammation. Participants rated the authority’s trustworthiness before and after the evidence update. Our findings indicate that communicating uncertainty buffers against a loss of trust when evidence changes. Moreover, explaining uncertainty does not appear to harm trust.