May 10, 2012

BSSR Lecture Series: Decline Effects/Other Reasons Why We Need an Open Repository

Speaker - Jonathan Schooler

Why do many published scientific effects appear to diminish with time? This so called "decline effect" has been observed both in individual labs (including my own) and in meta-analyses of findings across research in biology and medicine. Although some scientists dismiss the decline effect as simple statistical self-correction of initially exaggerated outcomes the truth is that we cannot be sure until we have better access to unpublished scientific work. In this talk I will review a variety of explanations for the decline effect including regression to the mean, publication bias, and gradual deterioration in experimental methods with replication. I will also consider the possibility that it may sometimes reflect genuine changes in scientific phenomena. In order to resolve the source of the decline effect I suggest the introduction of an open access repository for all research findings, which would enable researchers to fully log their hypotheses and methodologies ahead of time, and their results afterwards, regardless of outcome. Such a repository would not only shine light on the decline effect, but would also address a number of other difficulties stemming from the current lack of transparency in the scientific process.