A dose of the hormone makes human game-players behave more fairly.
The popular idea that testosterone always makes people more aggressive has been debunked by researchers. A team based in Switzerland has shown that the hormone can make people behave more fairly in an effort to defend their social status.
Ernst Fehr, an experimental economist at the University of Zurich, and his colleagues used the 'ultimatum bargaining' game to test how testosterone would affect behaviour in a group of 121 women. Counter-intuitively, women who were given testosterone bargained more fairly.
But the idea that testosterone causes aggression in humans, as it clearly does in rodents, is so firmly ingrained in the human psyche that women who believed they had been given testosterone — whether or not they had — bargained much less fairly.
Women, not men, were tested because they have less variable 'baseline' blood testosterone levels.
The study is published in Nature1. "It is a folk hypothesis that testosterone causes aggression," says Fehr. "But human society is more complex than this."
Several studies in humans have shown positive correlations between high blood testosterone levels and confrontational behaviour. But it has been hard to determine experimentally whether the aggression is caused by testosterone or is instead a consequence of a challenge to a person's social status.
The ultimatum game makes it possible to distinguish between these possibilities.
“It is a folk hypothesis that testosterone causes aggression. But human society is more complex than this. Ernst Fehr , University of Zurich”
In the game, two individuals must agree on the division of a sum of money. The proposer suggests a particular splitting of the sum and the responder must accept or reject the offer. The proposal is an ultimatum — the responder may not make a counter-offer. If the responder accepts the proposal, the money is duly allocated. If the responder rejects the proposal, neither the proposer nor the responder gets any money.
Responders normally reject very low offers as unfair — they would rather receive no money than see their partner carry off a disproportionate amount of cash.
Some proposers offer a 50-50 split because they are motivated by fairness, although most push to keep a bit more for themselves — but not so much more that they risk rejection and ending up with nothing.
Fehr's team reasoned that if testosterone caused aggression, it would cause proposers to make low offers. If, however, it promoted social-status-seeking behaviour, proposers would make higher offers to avoid the social affront of having their offers rejected.
The women were given either 0.5 mg testosterone or a placebo four hours before playing the ultimatum game for the sum of 10 money units. Before they played, they were asked to say whether they believed they had been given testosterone or placebo.
Women who received testosterone made significantly higher offers than those who received placebo — an average of 3.9 money units compared with the placebo group's average offer of 3.4 money units.
"In the socially complex human environment, pro-social behaviour, not aggression, secures status," says Michael Naef, an experimental economist at the Royal Holloway, University of London, who is a co-author on the paper.
The study has an additional, equally important message: those who believed they received testosterone, whether they had or not, made much lower offers — as low as 2 money units in some cases, or even nothing. "We think their belief that they had received testosterone, and that testosterone promotes aggression, gave them an up-front excuse to act more aggressively," says Fehr.
Responders remained as likely to reject a shabby offer when they were treated with testosterone as when they received placebo, showing that the hormone was not promoting altruistic behaviour.
Adam Goodie, a psychologist at the University of Georgia in Athens who works on decision-making, says: "The paper is a major blow to the popular wisdom that testosterone simply makes you more aggressive and less cooperative — the true picture is not nearly as negative."
"And it takes the field of neuroeconomics an important step further by showing that not only does biology affect economic behaviour — but so does belief," he adds.
The powerful impact of belief is a good lesson for neuroeconomists, adds Fehr: "Belief should always be controlled for in neuroeconomics studies, but often it is not."
Eisenegger, C. et al. Nature doi:10.1038/nature08711 (2009).