The Hypository: Registering Hypotheses for Cumulative Science

This is the authors' abstract. We don't add key points for this paper.

This article proposes the Hypository, a registration repository for sharing hypotheses resulting from unexpected findings, and hypotheses appearing to be so promising that researchers would like to publish them as soon as possible, whether they have the time and funding to study them or not. The goal of the Hypository is to enable researchers to get credit for their hypotheses when sharing them openly with the research community. The Hypository is expected to improve cumulative science by making sure that the hypothetico-deductive logic is not clouded by rogue intransparent exploratory approaches, while also enabling researchers to make the most of exploratory and/or unexpected findings.The Hypository contributes to solving the theory crisis by nudging researchers to better formalize their theories and hypotheses. It contributes to solving the replicability crisis by enabling researchers to find hypotheses waiting to be studied across labs, samples, and contexts. It contributes to new (or rediscovered) epistemological and methodological perspectives embracing both deductive and inductive steps of the research cycle while mitigating their respective limitations. We describe the possible usages, benefits and the possible features of this new registration platform for hypotheses, and provide a proof-of-concept shiny app and template illustrating the Hypository.