The standard story most people learn about stress is simple: when threatened, your body releases adrenaline and prepares to either fight or flee. Walter Cannon coined the phrase fight or flight in the 1920s, and it has structured a century of thinking about how humans handle danger.
But for half a century, the research on which that story was built had a quiet problem. Most of it was done on men.
In 2000, a team of psychologists led by Shelley Taylor at UCLA reviewed the human stress literature and noticed the imbalance. When they looked carefully at the studies that did include women, a different pattern emerged. Women under stress, on average, did not just fight or flee. They also reached out — to children, to friends, to family. They tended others. They sought social proximity.
Taylor and colleagues called this alternative response tend and befriend.
What the Research Actually Showed
Across a range of behavioral and biological data, Taylor's team identified a consistent pattern. Under acute stress:
- Women were more likely to engage in caregiving behavior toward offspring and dependents (tending)
- Women were more likely to seek social contact with others as a form of stress regulation (befriending)
- These patterns appeared in laboratory studies, observational data, and naturalistic life-event surveys
- The pattern was strong enough that prior fight-or-flight studies, conducted mostly on men, had been generalizing from a partial picture
This did not mean women never fight or flee, or that men never tend or befriend. Both responses exist in everyone. It meant the average pattern of stress response showed a meaningful sex difference that the older literature had missed.
The Biology Underneath
The proposed biological mechanism centered on oxytocin, a hormone involved in bonding, lactation, and social attachment. Several lines of evidence supported its role:
- Oxytocin is released under acute stress
- Oxytocin's effects on attachment and affiliation are amplified by estrogen, which means its bonding signal is stronger in women on average
- Androgens, more prevalent in men, partially attenuate oxytocin's calming effects
- Animal studies, particularly in prairie voles and rats, showed strong oxytocin involvement in maternal behavior under stress
The picture that emerged was not a wholesale rewrite of the stress system. It was a richer one. Sympathetic activation — the fight-or-flight machinery — still happens. But the parasympathetic and oxytocin-driven systems run alongside it, and in females their pull toward affiliation is, on average, stronger.
Why It Was Adaptive
The evolutionary case for tend and befriend is straightforward. Across most of human history, women were more often physically caring for vulnerable children. For a mother holding an infant, fleeing alone is not an option, and fighting a serious threat is rarely the best move. Protecting offspring and gathering allies was the more reliable survival strategy.
Tending the young, even under stress, kept them alive. Befriending — building and maintaining a network of other caretakers — provided the buffer that any one person could not provide alone.
Survival is not always about defeating the threat. Sometimes it is about being held, and holding, until the threat passes.
Where It Shows Up Today
The tend-and-befriend response is not a museum piece. It shapes a great deal of ordinary modern life:
- After a hard day, many women describe wanting to call a friend or be near family more strongly than wanting solitude
- During collective stress — a community tragedy, a national crisis — women, on average, mobilize caretaking networks more quickly
- Female-dominated professions like nursing, teaching, and social work involve sustained tending under chronic stress, and rely heavily on collegial networks for buffering
- Loneliness research consistently shows that social connection has particularly strong stress-buffering effects, with somewhat larger average benefits documented in women
This does not mean men do not benefit from social support. They do — substantially. But the pull toward seeking it under stress tends to be weaker on average in men, which has practical consequences for who reaches out and who isolates.
What It Means in Practice
Three takeaways are worth taking seriously.
For individuals. If your gut response under stress is to retreat into solitude, recognize that this is not the only valid stress strategy and may not be the most regulating one for you. If your gut response is to call someone, that is a legitimate biological strategy, not a sign of weakness or codependence.
For relationships. Misunderstanding between partners under stress often reflects mismatched stress strategies. One person reaches out; the other withdraws. Neither is wrong. They are running different responses, both real.
For workplaces and communities. Stress is rarely a purely individual phenomenon. Social infrastructure — networks of trusted others — is part of the buffering system, especially for those whose biology pulls them toward it. Strip out the network, and the entire system loses resilience.
What the Research Has Not Settled
Tend and befriend is a robust and well-supported model, but several open questions remain. The exact size of the sex difference depends heavily on the specific stressor and the cultural context. Some recent meta-analyses suggest the male/female gap is smaller than early framing implied, though the pattern of female-skewed affiliative response under stress holds. The biological story around oxytocin has also grown more complex; oxytocin is not a simple bonding hormone but a context-sensitive modulator that can amplify in-group affiliation while sharpening out-group wariness.
The core insight, though, has held up. There is more than one way for a body to respond to threat. Fight and flight are real, important, and shared across humans. So is the quieter, slower response of drawing closer to the people you love and the people who can help you survive what you cannot survive alone.
Naming that response gave a generation of researchers — and a great many ordinary people — language for something they had always done.
