Testosterone doesn’t do what you think it does, according to surprising new research

Two independently conducted studies—one in the United States and one in Europe—set out to test whether a single dose of testosterone could influence men’s willingness to compete, confidence, or risk-taking. The results, published in the journal Hormones and Behavior, suggest that testosterone had no significant effect on these behaviors, raising questions about the popular idea that the hormone fuels dominance and risk-taking.

Researchers have long speculated that testosterone may play a role in decision-making, especially in situations where status, competition, or risk is involved. Animal studies and some human data suggest testosterone rises in anticipation of competitive challenges, potentially driving behaviors that help individuals gain or defend social status. This line of thinking, known as the Challenge Hypothesis, has inspired a large body of research on how testosterone might influence confidence and decision-making in both men and women.

But findings have been mixed. Some earlier studies found links between testosterone levels and competitive or risky behavior, while others reported weak or no effects. Many of those studies were small or used methods that made it difficult to draw firm conclusions. To address these issues, the current research team conducted two large, randomized, placebo-controlled trials—considered the gold standard for testing causality.

The first experiment was conducted in Germany and included 91 healthy young men. Participants were randomly assigned to receive either a testosterone gel or a placebo. They returned to the lab 24 hours later to complete a series of economic tasks. These tasks measured willingness to compete, confidence in their own performance, and risk-taking behavior. For example, in one task, participants could choose to be paid based on their own performance or enter a tournament in which they would compete against another participant. In another, they chose between certain and risky monetary rewards.

The second experiment took place in California and included 242 men. This study followed a similar structure, but used a slightly different timeline and additional economic tasks. Participants received either testosterone or a placebo gel, then completed a range of decision-making tasks throughout the day. The researchers again measured their willingness to enter competitions, confidence in their abilities, and risk preferences.

Both studies used different methods to measure testosterone levels—blood samples in the German study, and saliva samples in the American study. In both cases, the testosterone levels of participants who received the hormone were elevated compared to those who received a placebo, confirming that the treatment worked as intended.

Despite the effective delivery of testosterone, neither study found significant effects on the main behavioral outcomes. Men who received testosterone were no more likely than those who got a placebo to enter competitions, express confidence in their abilities, or take financial risks. Any observed differences between the groups were small and statistically indistinguishable from zero.

The researchers also used a method called equivalence testing. This statistical approach helps determine whether an effect is not just insignificant, but also small enough to rule out any meaningful impact. The results suggested that, if testosterone has any effect at all on these behaviors, it is likely very small—smaller than what these studies were designed to detect.

These findings stand in contrast to some earlier research suggesting testosterone boosts risk-taking or competitiveness. The authors note that while prior studies may have found positive effects, many were limited by small sample sizes, exploratory methods, or publication bias. The new results cast doubt on the idea that a single boost in testosterone meaningfully changes how men make decisions in competitive or risky environments.

There are several possible explanations for these null findings. One is that testosterone’s effects may depend on specific situations or personal traits that were not captured in these studies. For example, past research has suggested that testosterone’s behavioral effects might be stronger in men with low cortisol levels, or in those with certain genetic traits or personality profiles. While the researchers explored some of these possibilities, they found no consistent evidence for interactions with cortisol or other factors.

Another possibility is that testosterone may influence behavior in more subtle or long-term ways that were not measured here. For example, it might affect how people learn from experience, regulate emotion, or behave in social situations that involve real stakes, like romantic competition or public performance. The lab-based tasks used in these studies, while controlled and precise, may not fully capture these real-world dynamics.

The researchers also caution that their findings apply only to single-dose testosterone administration in young, healthy men. It remains possible that longer-term testosterone changes, different delivery methods, or studies involving women or older populations might yield different results.

Ultimately, the two studies offer a rigorous test of a widely held belief—that testosterone drives competitive, confident, and risk-prone behavior. At least under the conditions tested, that belief does not hold up.

Future research may need to take a broader view, exploring how testosterone interacts with other biological and psychological factors over time. For now, these results suggest that a quick hormonal boost is unlikely to change how men behave in economic games, even when the stakes involve competition, confidence, or risk.

The study, “Does a single dose of testosterone increase willingness to compete, confidence, and risk-taking in men? Evidence from two randomised placebo-controlled experiments,” was authored by Amos Nadler, Matthias Wibral, Thomas Dohmen, Armin Falk, Alessandro Previtero, Bernd Weber, Colin Camerer, Anna Dreber, and Gideon Nave.