When Do Scientists Reach Their Peak?

In Organizing Genius, noted management thinker Warren Bennis chronicles the unifying characteristics of many of the finest teams mankind has ever known.  One particularly noticeable trait of many of these teams is their respective ages.

For instance, when Disney was blossoming in the late 1930s, Bennis notes that most of the employees were in their twenties, with the median age of just 26. Similar tales emerge from teams as diverse as Bill Clinton’s election team, the early Apple team, and even the scientists that converged on Los Alamos to create the atomic bomb.

Bennis reveals that in the vast majority of the great teams he studied, 35 was considered old. He suggests that the virtues of youth are that you have often boundless energy, combined with a lack of weariness in knowing what is ‘not’ possible. It makes for a heady cocktail.

Breaking the mold

This concept was also at the heart of This Idea Must Die, which builds on the scientific maxim that progress tends to advance “one death at a time”. In other words, scientific norms tend to quickly form and hang on until the people behind that idea pass away and the field is allowed to move on.

While I’ve written a number of times recently about the importance of not discounting or dismissing older people, in science at least, it seems youth is a clear virtue. This was reinforced by a recent study from Ohio State University, which showed that scientists were generally most creative in the first flushes of their careers. Indeed, the study found that the impact of one’s work seems to drop by up to two-thirds over the course of our career.

Despite this finding, the authors are at pains to point out that they’re not advocating dumping older researchers onto the scrapheap, but instead urge us to continue supporting scientists as they enter the second half of their careers.

Scientific impact

The researchers examined around 5.6 million biomedical papers that had been published between 1980 and 2009. The dataset included information on the authors as well as things like the citation record for each paper. This citation record allowed them to gauge the innovativeness of each paper by measuring the number of times other researchers referenced the paper. In other words, the more often a paper was cited, the more innovative it was perceived to be.

The data allowed the researchers to analyze the impact researchers had at various points in their career. The data supported the observation made by Bennis, with scientists making the most substantial breakthroughs earlier in their career. Indeed, the researchers found that the early phases of one’s career are high risk, with the chances of dropping out high. Should one make it through this period, however, then it’s likely that research will continue to be produced 20-30 years later.

“Early in their careers, scientists show a wide range of innovativeness. But over time, we see selective attrition of the people who are less innovative,” the researchers explain.

“So when you look at all biomedical scientists as a group, it doesn’t look like innovation is declining over time. But the fact that the least innovative researchers are dropping out when they are relatively young disguises the fact that, for any one person, innovativeness tends to decline over their career.”

Declining impact

This output declines considerably as one ages, however, as the data showed that the typical research would be cited up to two-thirds less often in the latter stages of their careers. The citation record was not the only evidence of declining impact, however, as the researchers also utilized a range of other metrics to try and capture the impact of any particular piece of work. These metrics included whether the latest ideas were employed, whether the latest research was cited, and whether multiple disciplines were drawn upon.

This analysis drew the same conclusion, that one’s innovativeness declined as the years passed. What’s more, this early peak was greatest among the most successful scientists, with less prestigious scientists suffering from a less vertiginous decline in fortunes.

“Young scientists tend to be at their peak of creativity, but there is also a big mix with some being much more innovative than others. You may not be supporting the very best researchers,” the researchers conclude. “With older, more experienced scientists, you are getting the ones who have stood the test of time, but who on average are not at their best anymore.”

Facebooktwitterredditpinterestlinkedinmail