Putting Journal-based Peer Review in Its Place: Transparency, Post-publication Review, and Systems for Self-correction

Journal-based peer review is a widely used form of gatekeeping in science, and publication in a peer-reviewed journal is often treated as a reliable indicator of credible research. In this paper, we argue that this view of journal-based peer review is misguided. Currently, journal-based peer review rarely prioritises checking accuracy, and journals typically do not require submitted research to be reported transparently enough to do so. Moreover, vetting research requires much more time and expertise than can be feasibly achieved by an editor and a small number of peer-reviewers. With more investment in accuracy checking, journal-based peer review could serve as a reliable initial quality filter. However, comprehensive quality vetting requires an expanded pool of experts, which, we argue, can only be achieved via post-publication review.At present, there is little infrastructure to support post-publication review and few incentives for researchers to engage in it. We propose a fully transparent system in which all manuscripts submitted for publication are posted to a central repository alongside editorial decisions, peer review reports, data, code, and as they become available, post-publication reviews. Post-publication review efforts could be directed toward the most promising, influential, or high-stakes research. Evidence syntheses, such as meta-analyses, offer a natural focal point for post-publication review, provided they prioritise rigorous assessment of the primary literature.