Many Labs 5: Registered Replication Report of Crosby, Monin & Richardson (2008)

Crosby, Monin and Richardson (2008) found that hearing an offensive remark caused participants (n=25) to look at a potentially offended person, but only if that person could themselves hear the remark. They thus argued that the computation of offense involves the coordinated processing of high level linguistic and interpersonal cues. Their key effect, however, was not replicated by Jonas and Skorinko (2015) as part of the Reproducibility Project: Psychology (Open Science Collaboration, 2015). Three labs from Europe and America (n=283) tested whether the size of that effect might be increased when the stimuli were modified to be more appropriate for a diverse range of participants, using a peer-reviewed and pre-registered protocol. We found that this manipulation of protocol did not affect the size of the social referencing effect but, interestingly, we did replicate the original effect reported by Crosby and colleagues, albeit with a much smaller effect size. We discuss these results in the context of ongoing debates about how replication attempts should treat statistical power and contextual sensitivity.