“PubPeer is okay, but …”: researchers' perceptions of post-publication reviews

ObjectiveWe study how researchers perceive comments pertaining to their papers on the debated post-publication peer review platform PubPeer.MethodsWe surveyed two groups: the commented group were researchers that received a critical comment on their article (10,603 sent, 80 respondents); and the citing group were researchers citing these articles, as a proxy for the wider community (84,325 sent, 720 respondents).ResultsAmong our respondents, the majority of commented authors knew of the PubPeer comment prior to the survey, compared to only a minority of citing authors. In both groups, the majority indicated that the issue raised on PubPeer did not affect the conclusions of their paper. Commented authors most frequently replied on PubPeer and/or contacted journal editors; citing authors largely did not act. Qualitative analysis revealed four types of defences used to minimise consequences: epistemic defence, honest error, contextual defence, and shifting responsibility. Respondents’ opinions about PubPeer varied between positive, negative and ambivalent, and anonymity was the main concern.ConclusionPubPeer's contribution to correcting the scientific record remains indirect, operating through retractions rather than through behavioural change in researchers. A diffusion of responsibility limits its correcting ambition, highlighting the need for complementary post-publication debate venues.