Men given testosterone are more likely to mirror opponents’ actions in strategic play

New research from China suggests that applying a 125 mg dose of testosterone gel makes men more likely to mimic the behaviors of their opponents in a variant of the Prisoner’s Dilemma game. Participants were more likely to defect (betray the other player) when their opponent defected and to cooperate when both players cooperated. They were also more likely to respond with angry emojis after receiving emojis indicating negative emotions from their counterparts. The research was published in Psychoneuroendocrinology.

Testosterone is the primary male sex hormone. It is produced mainly in the testes in men and in the ovaries and adrenal glands in women. Testosterone plays a crucial role in the development of male reproductive tissues, such as the testes and prostate. It is also essential for secondary sexual characteristics, including facial hair, a deeper voice, and increased muscle mass.

Beyond reproduction, testosterone contributes to bone density, red blood cell production, and overall physical energy. Levels of testosterone naturally rise during puberty and peak in early adulthood. As people age, testosterone levels typically decline, which can lead to symptoms such as reduced libido, muscle loss, or fatigue. In women, lower levels of testosterone help regulate ovarian function, bone health, and sexual desire. Abnormally low or high levels can cause health issues, requiring medical evaluation and, in some cases, hormone therapy.

Study author Renhui Fang and his colleagues aimed to explore the effects of testosterone administration on young men’s behavior and emotional expression in social games. Specifically, they wanted to understand how testosterone influences responses to the behavior of counterparts in such interactions.

The study involved 62 healthy young men with an average age of 21–22 years. Participants received 200 CNY (around $27) for their participation.

Free daily newsletter

Participants were randomly assigned to one of two groups. One group received testosterone gel applied to their left shoulder (150 mg dose), while the other group received a visually identical placebo gel applied in the same manner. Neither the participants nor the person applying the gel knew which treatment was administered—a double-blind design.

After one hour (the estimated time for testosterone to reach peak concentration in the body), participants played a modified version of the Prisoner’s Dilemma game. In this setup, two players (the participant and a computer opponent) decided whether to cooperate or defect. After each round, participants had the option to send one of four preset emojis to the other player.

Participants played up to 100 rounds of the game. The initial design included 80 rounds, but additional trials were conducted if some outcome types were underrepresented, ensuring all conditions were adequately sampled. The researchers ran two versions of the experiment with slight differences. For instance, in the second version, participants saw the emoji sent by their counterpart before responding.

The results indicated that testosterone administration increased the likelihood of cooperation when the counterpart cooperated and increased the likelihood of defection when the counterpart defected.

When analyzing emoji use, researchers found that participants in the testosterone group were more likely—compared to the placebo group—to respond with an angry emoji when they received a sad or angry emoji from their counterpart.

“We found that exogenous testosterone administration can affect the reciprocal tendency of men, making men more willing to cooperate in pursuit of mutual benefit and a win-win situation when facing opponents worthy of cooperation, but more unwilling to cooperate to maintain their self-interests when facing opponents who are not worthy of cooperation,” the study authors concluded.

The findings provide evidence that testosterone can influence social behavior and emotional expression in conflict-based interactions. However, it is important to note that this study involved young men in a low-stakes, simulated environment. Results may differ in other demographic groups or in real-world, high-stakes conflict situations.

The paper, “Testosterone and emotion expression shape social interactions in defection-cooperation conflicts,” was authored by Renhui Fang, Wenbin Pan, Mei Li, Zhihao Zhao, Kaiqiong Yuan, Ran Zhou, Benjamin Becker, and Hong Li.