Social Science Research Council Research AMP Just Tech
Citation

People adhere to content warning labels even when they are wrong due to ecologically rational adaptations

Author:
Horne, Benjamin D.; Nevo, Dorit
Publication:
Scientific Reports
Year:
2025

In this paper, we build on the theory of ecologically rational heuristics to demonstrate the effect of erroneously placed warning labels on news headlines. Through three between-subjects experiments (n = 1313), we show that people rely on warning labels when choosing to trust news, even when those labels are wrong. We argue that this over-reliance on content labels is due to ecological rationality adaptations to current media environments, where warning labels are human-generated and mostly correct. Specifically, news consumers form heuristics based on past interactions with warning labels, and those heuristics can spill-over into new media environments where warning label errors are more frequent. The most important implication of these results is that it is more important to thoughtfully consider what information needs to be labeled than it is to attempt to label all false information. We discuss how this implication impacts our ability to scale warning label systems.