The Theoretical Capture of Modern Science

The United States of America spends more on medical research than any other existing nation on Earth. Our scientific institutions sit atop the global hierarchy of research, receive billions in taxpayer funding, and generate an endless stream of publications. Yet, America remains one of the sickest populations in the developed world across a multitude of measures, including chronic disease, life expectancy, and mental illness.

The first thing I’d point to are the many treatments and guidelines that over the past century have been overturned. Hormone replacement therapy was promoted for decades as a way to prevent chronic disease until the Women’s Health Initiative (WHI) in 2002 found that it did not prevent heart disease and actually raised the risk of stroke, blood clots, and breast cancer. The damage was not abstract. In the WHI, combined hormone therapy was linked to about 8 extra cases of invasive breast cancer per 10,000 person-years, about 8 extra strokes per 10,000 person-years, and about 18 extra blood clots per 10,000 person-years.

Knee arthroscopy for osteoarthritis-related meniscal tears is another example: it was used for years before better evidence showed that physical therapy worked just as well for many patients. Even many recent interventions, like wearable fitness trackers for weight loss, have  failed to deliver the benefits earlier promised.

So, I would think it’s fair to ask why these failures continue to occur despite the resources and prestige of America’s “world-class” research institutions.

In a 2012 University of Virginia white paper, psychologists Brian Nosek, Jeffrey Spies, and Matt Motyl argued that scientific institutions increasingly reward novel findings over replication and verification. It started with a study led by Motyl and Nosek investigating embodiment of political extremism. They gave participants from across the political spectrum (left, right, center) and had them complete a perceptual task in which words were presented through different shades of gray. Participants had to click along a gradient representing grays from near black to near white to select a shade that matched the shade of the word. Their results showed that those who identified as “political moderates” were better at correctly perceiving the shades of gray. Before they were going to publish it, they decided to replicate their experiment because they “could not justify skipping replication on the grounds of feasibility or resource constraints.” In their replication, after testing 1,300 additional participants with enough statistical power to detect the original effect, the result completely disappeared. For Motyl, this was especially devastating and hurt his career advancement as he was about to enter the workforce.

Nosek used this as an example of a broader issue, arguing that “the incentives for publishable results can be at odds with the incentives for accurate results.” This has even more truth to it 14 years later, where the number of PhDs produced by American universities has grown to record numbers, exceeding 58,000 in 2024. At the same time, the path to traditional academic careers has narrowed, with a growing share of researchers cycling through postdoctoral and non-tenure-track positions. That means that for a grad student or even a junior professor, finding something new or surprising is more valuable for them professionally than publishing a well-run replication study. This is because universities rely on publication records as an indicator for success in order to get funding, which has led to measurable outputs like citation counts, impact factors, and H-index scores becoming increasingly more important in modern academia.

So, in 2026, even with open data practices and replication projects being more commonplace today as a result of the “great replication crisis” of the 2010s, the competitive pressures that created it in the first place still exist and are, in some cases, even stronger today than they were 14 years ago. A good way to put it is that we are more self-aware than self-correcting, since universities are still hiring, promoting, and rewarding researchers on the same basis of publications, citations, and academic visibility. These are the same universities in the U.S. and across the world which occupy the central position in the production of scientific knowledge itself. Until they put more value on verification, implementation, and patient results, America will continue to lead the world in scientific output without seeing any meaningful kind of progress in public health here at home.

-Alex Adamson

Link to Nosek/Motyl Case  

https://journals.sagepub.com/doi/10.1177/1745691612459058