Experienced Scientists Produce Fewer Breakthroughs—but More Lasting Ones

Experienced scientists produce highly disruptive research less often than younger researchers, but when they do, their innovations are more likely to have lasting effects, according to a new study.¹ The research was published in Proceedings of the National Academy of Sciences (PNAS) on October 19.

Previous research suggests that experienced scientists are less likely to produce disruptive research.² Scientific discoveries that overturn established foundations, rather than gradually building on them, are considered more disruptive than discoveries made earlier in researchers’ careers. Sam Aerts, who studies science, technology and innovation at the University of Leuven in Antwerp, Belgium, said the latest findings challenge the idea that scientists become less innovative as they age.

Study analyzed more than 11 million scientific papers

The study analyzed more than 11 million papers published between 1800 and 2022 by approximately 206,000 scientists around the world who had more than 40 years of research experience. The authors used citation patterns as a proxy for measuring how much a paper disrupted previous research and how likely its findings were to be replaced by subsequent studies.

Although citation patterns are often used to measure changes in scientific innovation, some researchers question whether these metrics accurately capture the disruptiveness of scientific discoveries.³

How the CD Index measures disruptive science

The researchers used a measure called the CD Index to estimate how strongly scientific papers disrupted earlier research.⁴ The metric is calculated by analyzing how later papers cite the original study.

If a paper is cited without its accompanying bibliography, or if later studies cite only the original paper’s references rather than the paper itself, that can indicate that those earlier references are no longer relevant, says co-author An Zeng, who studies science at Beijing Normal University in China.

Persistent Disruption tracks lasting breakthroughs

The traditional CD Index measures how strongly a paper disrupts previous research, but it does not show whether that disruption persists. To address this limitation, the researchers also used a metric called Persistent Disruption, or PD.⁵

The PD metric identifies highly disruptive papers using the CD Index and then examines how much those papers are themselves disrupted by future research. According to the researchers, a persistently disruptive paper overturns previous research and is not subsequently replaced by later studies.

The researchers first identified a focal paper and then examined whether subsequent papers cited it directly or indirectly by referring to studies that had cited it. Co-author Nan Deng, a science researcher at the National Science Library of the Chinese Academy of Sciences in Beijing, said that when later studies cite both the original paper and the studies that cited it, this suggests that the original paper’s disruptive influence has persisted.

Experienced researchers produce fewer but more durable breakthroughs

The study found that the probability of a researcher publishing a highly disruptive paper decreased as their career progressed. However, the probability of publishing a persistently disruptive paper increased over time—from about 1% early in a career to about 2.2% after 40 years.

The authors also reported that scientists who produced persistently disruptive research tended to publish relatively more papers before publishing their disruptive work.

However, Kenneth Huang, who studies innovation and technology management at the National University of Singapore, cautioned that the findings should not be interpreted to mean that senior scientists are generally more innovative or that aging itself leads to more lasting breakthroughs. Research suggests that the longer a scientist has been in their career, the more likely they are to work on papers with lasting impact, Huang said.

Real-world impact