“Watching Eyes” Do Not Strengthen the Behavioral Intention of Donating Blood: A High-Powered Pre-registered Replication Study

Abstract

Previous studies have demonstrated that eye-like images promote prosocial behavior and reduce antisocial behavior (watching-eyes effect). However, several recent reports of replication failure have called into question the replicability of the watching-eyes effect. It also remains unclear whether eye-like images promote prosocial tendencies or norm compliance. In this study, we attempted to replicate the watching-eyes effect on the behavioral intention of blood donation and examined the moderation effect of social norms in a high-powered pre-registered experiment. Participants were 1408 Japanese adults. A 2 × 2 (image: eye, control; norm: pro-donation, neutral) between-participants design was used. Participants first read an essay on either the importance of donating blood or a neutral topic. Then they reported their behavioral intention to donate blood while observing an eye-like or control image. The results showed non-significant effects of eye-like images and no interaction between eyes and norms. However, interestingly, our exploratory analysis suggests the possibility that the watching-eyes effect emerges only under specific contexts. We discuss the interpretations of the results and implications for future research.