Reputation without practice? A dynamic computational model of the unintended consequences of open scientist reputations

The push for open science has produced many initiatives that arguably improve the quality of science, such as preregistration of research methods, more general availability of scientific articles in the form of preprints and open access articles, and the availability of (meta)data and code. Open science can have direct benefits for the career prospects of the individual through higher quality of the work and increased chance of publication (Allen & Mehler, 2019). However, being an outspoken advocate of open science might also have indirect positive effects on individual scientific careers, in the form of status in a scientific community, decisions for tenure, and eligibility for certain kinds of funding. In light of these benefits, it may be profitable for individual labs to appear to engage in open science practices, without actually putting in the associated effort or doing only the bare minimum. In this article, we explore two types of academic behavior through a dynamic computational model (cf. Smaldino & McElreath, 2016) of an academic community that rewards open science: 1) actively practicing open science and/or 2) advocating open science. Crossing these two types of behavior leads to four different kinds of labs and we examine which of them thrive in this academic community. Implications of the model results are discussed.