A new study published in Psychoneuroendocrinology suggests that while a person’s baseline testosterone does not simply make them more or less inclined to take risks, the hormone may shape how the body processes the outcomes of past decisions. Researchers measured sweat on the skin during a gambling game to show that testosterone can influence the body’s response to feedback and how that response colours subsequent choices.
The experiment recruited 55 young adults, comprising 27 women and 28 men, who completed 100 rounds of the Iowa Gambling Task while researchers monitored electrodermal activity. At the outset, participants provided saliva samples so their baseline testosterone could be assessed.
Grounded in the idea that decision-making can occur beneath conscious awareness, the researchers noted the backdrop of the Somatic Marker Hypothesis, which suggests bodily “gut feelings” help steer choices. They also acknowledged past studies linking testosterone to riskier choices on the task, though a 2026 meta-analysis cited in the study argued there is no robust direct link between the hormone and general risk-taking.
The team therefore asked whether testosterone might instead influence how people learn from reward and punishment, rather than how often they opt for risky decks. “This study is part of a larger research project investigating the relationship between hormones and decision-making,” said Félix Duplessis-Marcotte, a doctoral candidate at the Université du Québec à Montréal, describing the broader programme alongside Marie-France Marin.
As the researchers framed their investigation, they noted that recent meta-analyses question whether testosterone acts as a simple dial for risk, prompting a shift in focus toward its role in decision processes rather than static risk preferences. “As for testosterone, recent meta-analyses suggest that testosterone is not simply associated with people being more or less risk-taking overall,” Duplessis-Marcotte added. “That raised a different question: rather than asking whether testosterone relates to decision-making, what if it influences how people make decisions?”
The laboratory setup involved 55 participants completing 100 IGT trials while their baseline testosterone was measured and their physiological arousal tracked. Electrodermal activity was recorded to separate anticipatory responses—those occurring in the seconds before a card draw—from feedback responses that followed wins or losses.
In terms of general risk-taking, baseline testosterone levels did not correlate with how often participants chose the risky decks. The findings align with the view that testosterone is not simply a universal risk-acceleration factor.
Looking at anticipatory signals, the study found sex-specific patterns. In men, the physiological warning signals intensified before selecting a risky deck as the task progressed and were linked to upcoming choices. Women tended to show fewer differences in arousal before safe versus risky choices, maintaining a steadier level of risk-taking. Importantly, a participant’s testosterone level did not modify these anticipatory gut feelings or their link to subsequent choices.
The key discovery lay in how testosterone related to responses after outcomes. Among men, the combination of their testosterone level and their bodily reaction to feedback influenced their next move: those with lower testosterone showed that stronger arousal after an outcome tended to reduce risk-taking on the next turn, while those with higher testosterone displayed the opposite pattern, with stronger arousal after feedback predicting more risk-taking on the following draw. This particular interaction between testosterone, feedback-related sweat, and subsequent choices was not observed in the women studied.
Duplessis-Marcotte emphasised the broader interpretation: “The main takeaway is that testosterone may not make people generally more risk-taking. More broadly, these findings add to growing evidence that testosterone’s role in human behavior is more nuanced than the popular stereotype that it simply increases risk-taking.” He suggested testosterone might be linked to the affective and physiological processes through which people learn from past experiences when making decisions.
The study also identified a longer-running behavioural trend following losses. Across both sexes, those with lower testosterone levels tended to switch to a safer deck after a loss, whereas those with higher testosterone levels continued to take risks, a pattern resembling loss-chasing in gambling contexts.
The researchers note that their conclusions build on prior PsyPost reporting, which highlighted that testosterone does not drive general risk-taking but can influence responses to decision feedback. They also point out that their own work measured natural hormone levels, in contrast to some earlier studies that used artificial hormone doses.
As with all research, the authors caution that the effect sizes are modest and the study is correlational, not experimental. They emphasise that laboratory findings may differ in social settings or with different decision-making tasks, and that sweat measurements reflect arousal intensity rather than emotional valence. They also acknowledge the relatively small sample when split by sex and hormone levels, and that testosterone was measured only once at the start.
Looking ahead, Duplessis-Marcotte argued for a shift away from treating testosterone as a fixed trait and toward examining its real-time fluctuations in social and competitive contexts. “Future studies combining repeated hormone measurements with physiological signals and more ecologically valid decision-making tasks could help clarify when and how testosterone influences the decision process.”
The study, titled Testosterone is associated with affective feedback processing in ambiguous decision-making, was authored by Félix Duplessis-Marcotte, Valentin Magnon and Marie-France Marin.
