The Medusa effect: a registered replication report of Will, Merritt, Jenkins and Kingstone (2021)

Will et al.'s (2021 Proc. Natl Acad. Sci. USA 118, e2106640118 (doi:10.1073/pnas.2106640118)) found the Medusa effect, which refers to the tendency that people evaluate a 'person in picture' more mindful than a 'person in picture of a picture'. The present study tried to directly replicate the Experiments 2 and 5 of Will et al.'s (2021), to examine the reliability, validity and generalization of the Medusa effect, as well as its effect on prosocial behaviour. We used the same stimuli and methodology as the original research, but recruited participants in Japan with a larger sample size (N = 1387 in total) as a registered report. In our two replication experiments, we again found that pictures with lower levels of abstraction (L1) were perceived to convey more mind and lead to higher levels of prosocial behaviour, successfully replicating the original findings. The results of the present study suggested the high reproducibility and generalizability of the Medusa effect. Pre-registered Stage 1 protocol: https://osf.io/xj46z (date of in-principle acceptance: 9 February 2023).