Journal editors, reviewers don’t show bias against novelty

Scientific journals are likely to accept papers that provide new findings compared with studies reporting conventional results, which is contrary to long-standing concerns about publication biases.

Using peer-reviewed data from 49 journals in the life and physical sciences, a new University of Michigan study found that evaluators are not biased against novelty, as scholars have suggested for decades.

For many years, concerns were noted that scientific institutions, which often rely on peer review to select the best projects, tend to select conservative ones, said study lead author Misha Teplitskiy , U-M assistant professor of information. This, in turn, could discourage scientists from pursuing new research, he said.

However, given the secrecy of the selection processes, it has been difficult to test whether the concerns are well-founded, and whether they apply to scientific evaluation generally or only to early-stage selections.

Focusing on later-stage selection, when the projects are largely complete, the researchers sought to determine whether less novel manuscripts were more likely to be accepted for publication. The data cover 20,538 manuscripts submitted between 2013 and 2018 to the journals Cell and Cell Reports, and 6,785 manuscripts submitted in 2018 to 47 journals published by the Institute of Physics Publishing.

When analyzing the selection decisions, the researchers did not find evidence of conservatism. Across journals, more novel manuscripts were more likely to be accepted, even when peer reviewers were similarly enthusiastic, the research showed.

"The findings suggest that peer review is not inherently conservative, and help explain why researchers continue to do novel work,” Teplitskiy said about the study, which appeared last month in the Proceedings of the National Academy of Sciences. "Novel ideas may still be disfavored at earlier stages, when there is much more uncertainty about whether the ideas will work.”

Teplitskiy and colleagues also investigated an additional type of bias that has been alleged to drive conservatism. If a paper or creative project is new, evaluators may disagree with each more often, and if being accepted in a journal requires consensus, this could create an anti-novelty bias at the organizational level rather than individual level, Teplitskiy said. Yet the study shows that reviewers did not disagree with each other more when assessing more novel manuscripts.

Overall, the findings suggest that journal peer review favors novel research that is well situated in the existing literature, incentivizing exploration in science and challenging the view that peer review is inherently anti-novelty, Teplitskiy said.

The study’s authors included Hao Peng, postdoctoral researcher at Northwestern University; Andrea Blasco, researcher with the European Commission; and Karim Lakhani, professor of business administration at Harvard University.