October 23, 2017

Perils in the Pursuit for Scientific Novelty

Print More

Scientific research isn’t perfect, far from it.

In fact, according to Richard Harris, correspondent at National Public Radio, the scientific process is in need of repair. Among the many issues, the limited ability to examine existing, mundane findings seems to be a consistent obstacle. At a lecture at Cornell on Oct. 16, Harris discussed his criticisms of the manner in which science research is published and presented.

Harris highlighted a critical issue affecting biomedical research in the nation and beyond: widespread reductions in funding across all sciences. Even without the cuts to research proposed in the current presidential administration’s budget, the National Institutes of Health lost 22 percent of its funding between 2003 and 2015. Coupled with a general downward trend in returns — for example, lower pharmaceutical drugs per billion dollars spent in research and development — scientists are under greater pressure than ever.

“People who happen to depend upon certain literature realize there are a lot of places with a lot of defects,” Harris said. “Especially in terms of this culture, the financial pressures confronting science. There has been a tremendous increase in the use of words like robust, novel and unprecedented.”

Harris said this funding pressure forces researchers to strive for novel breakthroughs. The competition for grants seems to funnel researchers into a “publish or perish” culture, giving them little wiggle room to test different hypotheses and causing them to read too much into results, or often, designing studies that cannot be replicated.

“Papers themselves don’t move science forward. If papers are the endpoint of science, then you are in trouble,” Harris said.

“We need to think differently about the incentives in biomedical research because if you create bad incentives, researchers are not going to come up with results that they don’t want. Instead we need a open science framework that encourages people to use transparency, talk about all their data, and encourages them to generate and present hypotheses, in short to excel in the way science is done.”

That scientific publishing is plagued by the pursuit of novelty over rigorous science is something Prof. Anthony Bretscher, molecular biology and genetics, agrees with.

“Previously, when you submitted a paper to any journal you didn’t put in a cover letter that said how wonderful your work was. You wrote a paper, the editor would get it and send it out to editors to check if it was scientifically solid and made some progress,” Bretscher said. “Now you try to say what you show in this paper is novel. What does that do to science?”

Furthermore, while Bretscher acknowledges that the scientific community needs to re-examine the structures that maintain the credibility of scientific research, he contests the notion that financial burdens are the cause of these issues. Instead, he finds greater fault in the decreasing scientific rigor of the reviewing process used by high profile journals. Novelty, he said, comes at the expense of necessary rigor.

“If I want to publish in these journals, I have to take the results I have and try to spin them in a way that makes the research sound more novel. I spun a good story but it may not stand the test of time,” Bretscher said.

He also contests Harris’ notion that publishing corrections to scientific research journals is not as widespread as it should be.

“Richard Harris brought up the issue of rebuttals to incorrect science in novelty-journal papers being published in obscure journals. That ought to be immoral. Negative results tell you stuff.”

One possible solution, Bretscher said, is to increase the involvement of faculty professors during the reviewing process. Open-access journals like eLife and PLoS ONE are already using such models, however increasing the number of professors who sign up for reviewer spots at journals across the world continues to be a serious challenge.