The Myth
Call someone "testosterone-fueled" and everyone knows what you mean. Aggressive, reckless, selfish. The label carries so much cultural weight that it barely needs explanation. Decades of rodent studies seemed to confirm the connection: castrate a male rat and he calms down; inject testosterone and he fights. Scientists and journalists extrapolated freely from cage to boardroom, and the folk theory hardened into a fact so accepted that pharmaceutical companies now warn patients about "increased aggression" as a potential side effect of hormone therapy.
Christoph Eisenegger, a neuroscientist at the University of Zurich, suspected the story was wrong. In 2009, he and economist Ernst Fehr designed an experiment to separate what testosterone actually does in humans from what people believe it does. The results, published in Nature in January 2010, surprised even the researchers.
121 Women, One Pill, and a Pile of Money
The team recruited 121 women for a double-blind, placebo-controlled trial. Each received a sublingual dose of either 0.5 milligrams of testosterone, enough to raise circulating levels roughly fourfold, or an identical-looking placebo, and neither the participants nor the experimenters knew who received what until the data were analyzed.
Then the bargaining began. Each pair played an ultimatum game, one of behavioral economics' most studied paradigms: Player A proposes how to split a pot of real money, and Player B can either accept the split or reject it, in which case neither player receives anything. The rational move for Player A is to offer just enough to avoid rejection. The fair move is fifty-fifty.
If the folk theory held, women on testosterone should have made lower, greedier offers.
They did the opposite.
Fairer Under the Influence
Women who received testosterone made significantly higher offers than those who received placebo. On a 10-point monetary scale, the testosterone group's average offer was 0.6 units higher, a shift that sounds modest until you consider that ultimatum-game offers typically cluster between 3 and 5, and moving the distribution upward by more than half a unit substantially reduced the rate of rejections and made the bargaining process more efficient overall. The hormone associated with aggression had made people more cooperative.
No mood change, anxiety shift, or cortisol fluctuation explained the result. The molecule itself drove the behavioral change.
When Belief Overrides Biology
The most striking finding was not about the drug. It was about the story people tell themselves.
After the game, each participant guessed whether she had received testosterone or placebo, which created a two-by-two matrix of actual treatment crossed with believed treatment that revealed a pattern more dramatic than the drug effect itself. Women who received testosterone but believed they had received placebo made fair offers about 60 percent of the time. Women who received placebo but believed they had received testosterone made fair offers just 10 percent of the time. The belief effect pushed offers down by 0.9 units on average, which was larger in magnitude than the hormone's 0.6-unit boost in the other direction.
The cultural narrative about testosterone was powerful enough to overwhelm the molecule's biological effect. Women who thought they were under the hormone's influence behaved as though aggression were expected of them. "The preconception that testosterone only causes aggressive or egoistic behavior in humans is thus clearly refuted," Eisenegger wrote.
Why Fairness Can Be Status-Seeking
If testosterone does not promote aggression in humans, what does it promote? Eisenegger and Fehr offered a parsimonious answer in a 2011 review in Trends in Cognitive Sciences. Status. In species with simple social hierarchies, status comes from physical dominance, so testosterone facilitates aggression, but in humans, whose social systems are vastly more complex, status often comes from being respected, trusted, and perceived as fair, which means that making a generous offer in a bargaining game protects your reputation, reduces the chance your proposal is rejected, and serves the underlying drive for social standing.
A 2016 PNAS study by Jean-Claude Dreher and colleagues tested this framework directly in 140 men, and the results were unambiguous. Participants who received testosterone punished unfair opponents more harshly than the placebo group, but they also rewarded fair opponents more generously, amplifying both responses exactly as the status hypothesis predicted rather than pushing behavior in a single antisocial direction. Testosterone made men more responsive to social norms. Not uniformly aggressive.
The Strongest Counterargument
The most pointed critique came from the Neuroskeptic column at Discover Magazine shortly after publication. The belief analysis, the critic argued, is fundamentally correlational: participants guessed whether they received testosterone after playing the game, not before, which means that women who behaved selfishly for unrelated reasons may have then inferred they must have been on testosterone rather than the belief causing the selfish behavior. Eisenegger did not randomly assign belief independently of the drug. He measured it post-hoc. Without that randomization, the dramatic "belief reversal" narrative rests on a correlation that could run in either direction.
The primary finding is also not settled science. Zak and colleagues found in 2009 that testosterone made male proposers less generous, Zethraeus and colleagues in the same year found no behavioral effect at all in women receiving chronic testosterone, and Dreher's 2016 results, while supporting the status hypothesis in men, showed both prosocial and antisocial effects depending on whether opponents had been fair or unfair. The picture across the full literature is nuanced: testosterone's social effects appear highly sensitive to context, dosing protocol, and the specific bargaining game being played. One Nature paper does not close the book.
What We Didn't Prove
The study used exclusively female participants receiving a single acute dose of 0.5 milligrams. Whether men, who carry 10 to 20 times higher baseline testosterone, would show the same fairness boost is an open question; Dreher's 2016 study with men showed a more complex pattern of amplified responses to both fair and unfair treatment. Chronic testosterone exposure, as in clinical hormone replacement therapy, may differ from a single sublingual dose. The ultimatum game is one specific social situation; real-world negotiations involve repeated interactions, reputational stakes, and emotional histories that no lab protocol replicates. The cell sizes in the belief analysis were small enough that the dramatic percentages come with wide confidence intervals, and all participants were Swiss university students in a country where bargaining norms may differ substantially from other cultures.
The Bottom Line
A double-blind Nature trial found that testosterone made women fairer, not meaner. The far larger threat to cooperative behavior was the cultural belief that the hormone causes aggression, a belief potent enough to overwhelm the drug's biological effect. The evidence across multiple studies increasingly supports a simpler model: testosterone amplifies the drive for social status, and in humans, fairness often outcompetes aggression as a strategy for earning it.
What You Can Do
Be skeptical when someone reduces complex behavior to a single hormone. When a commentator invokes "testosterone-fueled aggression" to explain road rage or boardroom bullying, ask what trial design supports that claim. If you or someone you know is on testosterone therapy, understand that the evidence for increased aggression is far weaker than warning labels suggest. Clinical meta-analyses consistently fail to find a significant aggression signal at therapeutic doses. And if you study or report on endocrinology, treat animal-to-human extrapolation with the caution it deserves. What makes a rat aggressive and what makes a human unfair are separated by several hundred million years of social evolution, and the folk story about testosterone belongs to the rat.