New ideas are struggling to emerge from the sea of science

A flood of scientific papers may be paradoxically preventing new ideas from advancing — and slowing the progress of science.

Why it matters: Global warming, emerging viruses and the growing global burden of chronic diseases all underscore the need for faster, meaningful scientific innovations to help solve complex and consequential problems.

The big picture: There is an ongoing debate about why the rate of scientific progress appears to be slowing down despite an increase in the number of scientists, amount of funding for their work, and the quantity of papers they publish.

• Some researchers believe the cause is fundamental: The low-hanging fruits of discovery have simply already been plucked, meaning scientists have to work harder and more money needs to be invested to get what remains.
• Others say as knowledge accumulates in ever greater amounts, researchers face a higher burden to learn about a field.
• But in science, new ideas can be recombined with old ones and begin to multiply, which makes it difficult to tease out the drivers of stagnation, says Dashun Wang, who directs the Center for Science of Science and Innovation at Northwestern University's Kellogg School of Management.

What they found: A new analysis suggests another possible source of stagnation: A deluge of new publications is leading to an "ossification of canon," Johan Chu of Northwestern University and James Evans of the University of Chicago write.

• Chu and Evans argue scientists are "cognitively overloaded" by the flood of papers to read, review and cite, so they look for how the work in front of them relates to well-established research. And the researchers predict if ideas arrive too fast in publications, they will compete for attention and wash each other out.
• They tested those predictions by analyzing more than 1.8 billion citations of about 90.6 million papers published between 1960 and 2014 in 10 large scientific fields.

They found a massive inequality in how papers in those fields are cited, they report in the Proceedings of the National Academies of Sciences.

• For example, when the field of electrical and electronic engineering published about 10,000 papers each year, the top 1% most-cited papers got almost 9% of total citations and the bottom 50% least-cited papers got 44% of citations. When the field grew to produce 100,000 papers per year, the top 1% received almost 17% of citations whereas the bottom 50% got about 20%. That large inequality wasn't seen in smaller fields.
• And "the most-cited papers maintain their number of citations year over year when fields are large, while all other papers' citation counts decay," they write. That crystallizes the canon.
• They also found that in smaller fields, papers slowly rise into the top bracket, whereas in larger fields, the papers that do make it into the canon get there fast. "It is predetermined socially rather than scholarly," says Chu, who led the work.
• In larger fields, most papers build on others — without disrupting the canon — but in smaller disciplines, papers are more likely to be disruptive.

The impact: Ideally, scientific publishing is a mechanism for ideas to be assessed, and then either advanced or abandoned.

• But, "there is so much new work that we just keep turning back to the old stuff and not overturning it," says Ethan Mollick, who studies innovation at the Wharton School of the University of Pennsylvania.

Yes, but: The authors acknowledge there could be other explanations for their findings, including a field growing and citations concentrating as a field ages, but they say the size of the field still has the strongest effect.

• There are also open questions about how to even define and measure concepts like novelty and fields that can bleed into one another, says Joshua Graff Zivin, an economist at the University of California, San Diego.

What's next: Perhaps ironically, more research is needed to try to tease out the mechanisms driving the findings.

• Chu and Evans argue changes in the incentives and rewards for scientists to produce more science could help to spur true advances.
• But, but, but: It's not clear what balance between radical and incremental science will best benefit society, Graff Zivin says.

The bottom line: As a scientist, "you get a lot of reward for doing something better. You get an immediate paper and a lot of appreciation," says Konrad Kording, a professor of neuroscience at the University of Pennsylvania.

• There is a deep addiction among scientists to publish paper after paper in order to be funded and advance their careers, he says.
• "You don't get a reward or appreciation for doing something different."