Picture two morning runs. In the first, you race a friend to the end of the trail, checking your watch and determined to finish ahead. In the second, you run alongside a group, matching stride and keeping the pace comfortable for everyone. The distance is the same, and the heart rate may be nearly identical. Yet the hormonal signatures of these two runs can look quite different.

Exercise physiology has traditionally focused on intensity, duration, and type of movement. A growing body of research in behavioural endocrinology suggests that the social frame of activity matters too. Whether we are striving against someone or moving with them appears to shape the release of testosterone, cortisol, endorphins, and the neuropeptides linked to bonding.

Herbert Benson showed decades ago that mental context can switch on measurable physiological states, such as the relaxation response. Competition and cooperation appear to be two more such contexts. Each reflects a different social situation, and the body responds to that situation as much as to the effort itself. Understanding both offers a fuller picture of what physical activity does to us.

Competition often begins changing our chemistry before the first move is made. Researchers have observed anticipatory rises in testosterone in athletes preparing for matches, and even in chess players before tournament games. Sociologist Allan Mazur's biosocial model of status proposes that this rise prepares us for a status contest. It may support focus, assertiveness, and a willingness to take risks.

What happens afterwards depends partly on the outcome. In many studies, winners show stable or rising testosterone while losers show a decline, which is often called the winner-loser effect. The pattern is not universal. A 2017 meta-analysis by Geniole and colleagues found the effect to be modest overall, and stronger in real-world settings than in the laboratory. How decisive the victory felt, and whether it seemed earned, also appears to matter.

Cortisol tells a related story. Competition reliably raises cortisol, partly because of physical exertion and partly because of what is at stake psychologically. In a 2004 meta-analysis of more than 200 laboratory stressors, Dickerson and Kemeny found that the largest cortisol responses came from tasks combining uncontrollability and social-evaluative threat, meaning situations where we might be judged or lose standing. Competition often contains both ingredients.

These hormones also interact. The dual-hormone hypothesis, proposed by Mehta and Josephs, suggests that testosterone's link to dominance-related behaviour is strongest when cortisol is low. When cortisol is high, that link weakens or disappears. None of this makes competition inherently harmful. Acute cortisol mobilises energy and sharpens attention. The concern arises when the stress response fails to switch off after the contest ends.

Takeaway

Competition is as much a social contest as a physical one. The body responds to what is at stake for our standing, not only to the effort our muscles exert.

One of the most cited studies here comes from the Oxford rowing team. In 2010, Emma Cohen and colleagues had rowers train either alone or in synchronised crews at comparable effort. After group rowing, the athletes' pain thresholds rose significantly more than after solo sessions. Pain tolerance is commonly used as an indirect marker of endorphin release, which suggests that moving in synchrony amplifies the body's own opioid activity.

Anthropologist Robin Dunbar has argued that the endorphin system is a key mechanism behind human group cohesion. It may explain why laughter, communal singing, and dancing feel so bonding. Later work by Bronwyn Tarr and colleagues found that synchronised dancing raised pain thresholds and increased participants' sense of closeness to one another. Shared rhythm seems to blur the boundary between self and group.

Oxytocin is often described as the bonding hormone, and some studies report increases after group singing or cooperative tasks. Here, caution is warranted. Measuring oxytocin in blood or saliva is technically difficult, results vary between labs, and human oxytocin research has faced replication problems. The picture is suggestive rather than settled. Endorphin and cortisol findings currently rest on firmer ground.

Cooperative settings also tend to carry less social-evaluative threat, so cortisol responses are often more modest. Testosterone appears sensitive to who we are facing. In fieldwork in a Caribbean village, Mark Flinn and colleagues found that men's testosterone tended to rise after defeating rivals from another community, but not after beating their friends. Team sports blend both modes, with cooperation inside the group and competition directed outward.

Takeaway

Moving in time with other people may be one of humanity's oldest bonding technologies, and the shared rhythm itself seems to do biochemical work.

Single sessions are one thing, but habits shape health. In a 2018 study in The Lancet Psychiatry, Adam Chekroud and colleagues analysed data from about 1.2 million American adults. Every type of exercise was associated with fewer days of poor mental health. Team sports showed the largest association. Because the data are observational, they cannot prove causation. People who join teams may already differ in personality, resources, or social networks.

Similar patterns appear in younger people. A 2022 study by Matt Hoffmann and colleagues, drawing on data from more than 11,000 American children, found that team sport participation was linked to fewer symptoms of anxiety and depression. Some individual sports were associated with greater difficulties. Researchers suspect that belonging, shared purpose, and supportive relationships mediate much of this benefit.

Social connection also changes how we handle stress. In a 2003 experiment, Markus Heinrichs and colleagues found that men who received social support before a public-speaking stress test showed lower cortisol responses. The lowest responses came when support was combined with intranasal oxytocin. By contrast, repeated competitive pressure without adequate recovery is associated with burnout and dysregulation of the hypothalamic-pituitary-adrenal axis in athletes.

None of this means competition should be avoided. It builds motivation, skill, and resilience, especially when we appraise it as a challenge rather than a threat. The practical lesson is balance. That might mean pairing competitive training with cooperative movement, nurturing team relationships, and noticing whether an activity leaves you energised or depleted hours later.

Takeaway

Over time, the most important question may be less about how hard you trained and more about who you trained with and what felt at stake.

The research points to a simple but often overlooked idea. The body does not experience exercise in a social vacuum. Competitive contexts tend to mobilise testosterone and cortisol, preparing us to contest status and perform under scrutiny.

Cooperative contexts, especially those involving synchrony, appear to engage endorphin systems and possibly oxytocin, which supports bonding and buffers stress. Each profile serves a purpose, and neither is simply good or bad.

In practical terms, the social frame of movement is a variable we can choose. Rotating between striving against others and moving with them may offer the benefits of both, the sharpening of competition and the connection of shared effort.