The Biology of Stress

In brief

Confront a threat and your body runs a program older than thought. Within seconds the sympathetic nervous system floods you with adrenaline: the heart pounds, the airways open, glucose pours into the blood, digestion halts. Within minutes a second, slower system, the HPA axis, releases cortisol to sustain the mobilization. Walter Cannon (1871 to 1945) named the fast part the fight-or-flight response; Hans Selye (1907 to 1982) named the drawn-out pattern the general adaptation syndrome. Both were describing a machinery superbly built for a short emergency and poorly suited to a threat that never ends. The modern science of stress is largely the story of what that mismatch costs, and of the fact that the response is neither uniform across people nor untouched by how we interpret and manage the threat.

The full treatment

The problem it answers

An organism facing a predator has no time to deliberate. It needs energy delivered to the muscles at once, and every process that can wait (digestion, growth, reproduction, immune housekeeping) suspended. Cannon's insight, developed at Harvard and set out in The Wisdom of the Body (1932), was that the body defends a stable internal state, which he called homeostasis, and that under threat it departs from that state in a coordinated, purposeful way. The fight-or-flight response is that departure: a whole-body reallocation toward immediate survival, orchestrated too fast for conscious control.

How it works: two axes, two speeds

The response runs on two tracks. The fast track is the sympathetic-adrenal-medullary (SAM) axis. A threat signal reaches the hypothalamus, the sympathetic nervous system fires, and the adrenal medulla dumps adrenaline (epinephrine) and noradrenaline into the blood in seconds. Heart rate and blood pressure rise, pupils dilate, bronchi widen, and stored energy is released for use.

The slow track is the hypothalamic-pituitary-adrenal (HPA) axis, a hormonal relay. The hypothalamus secretes corticotropin-releasing hormone, which prompts the pituitary to release adrenocorticotropic hormone (ACTH), which tells the adrenal cortex to release cortisol, a glucocorticoid. Cortisol takes minutes rather than seconds but lasts far longer: it keeps blood glucose high, sharpens some forms of attention, and dampens inflammation and other systems the emergency can afford to defer. Crucially, cortisol feeds back on the hypothalamus and pituitary to shut the cascade down once the threat passes. That off-switch is the point. The system is designed to spike and then return to baseline.

What it claims: the general adaptation syndrome

Selye's contribution was to notice that wildly different insults (cold, injury, toxins, infection, psychological strain) produced the same underlying bodily reaction, which he called "nonspecific." In a one-page letter to Nature in 1936 and later in The Stress of Life (1956), he laid out three stages. First, alarm: the body's initial shock and the emergency mobilization just described. Second, resistance: if the stressor persists, the body adapts, running at an elevated set point that looks outwardly like coping. Third, exhaustion: if the stressor persists too long, the adaptive resources deplete and the body begins to break down, opening the door to what Selye called "diseases of adaptation." His central and unsettling claim was that the damage need not come from the stressor itself. It can come from the response, sustained past its usefulness.

Not one response: tend-and-befriend

For most of the twentieth century, fight-or-flight was treated as the human stress response, full stop. But the foundational studies were run overwhelmingly on males, human and animal. In a 2000 paper in Psychological Review, Shelley Taylor and her colleagues argued that females, who in evolutionary terms could not readily flee or fight while pregnant or caring for young, show an additional pattern they named tend-and-befriend: under threat, a pull toward protecting offspring (tending) and toward forming protective social alliances (befriending). They proposed that this pattern is supported partly by oxytocin, whose calming, affiliative effects are amplified by estrogen and blunted by androgens. The claim is not that women do not experience fight-or-flight, but that the stress repertoire is broader and partly sex-differentiated, and that the male-default model had quietly narrowed the science.

How we handle it: coping

Physiology is only half the story, because the same event is not equally stressful to everyone. Richard Lazarus (1922 to 2002) and Susan Folkman argued in Stress, Appraisal, and Coping (1984) that stress arises from a transaction between person and situation, mediated by appraisal. In primary appraisal we judge whether an event threatens us; in secondary appraisal we judge whether we can do anything about it. From that they drew a durable distinction between two coping strategies. Problem-focused coping targets the stressor itself: make a plan, gather resources, change the situation. Emotion-focused coping targets the distress the stressor causes: reframe it, seek comfort, accept what cannot be changed. Neither is simply better. Problem-focused coping serves controllable stressors; emotion-focused coping serves those that cannot be altered, where insisting on control only adds a second frustration to the first.

Lineage

The physiology of stress descends from Claude Bernard's nineteenth-century idea of the constant internal environment, which Cannon formalized as homeostasis and extended into his account of emergency arousal. Selye, who trained under the influence of Cannon's work, borrowed the very word "stress" from physics and engineering, where it named a load applied to a material, and applied it to the body's load-bearing response. Lazarus and Folkman added the cognitive layer that Cannon and Selye had left implicit, insisting that meaning, not just stimulus, drives the response. The line then runs forward to Bruce McEwen's concept of allostatic load and to Robert Sapolsky's popular synthesis, both of which recast Selye's "exhaustion" in modern terms.

The strongest case for it

The framework has earned its place because it is mechanistically real and clinically powerful. The HPA axis and the cortisol cascade are not metaphors; they are measurable, traceable pathways, and their chronic overactivation is tied to a documented list of harms: hypertension, suppressed immunity, impaired memory and hippocampal changes, metabolic disruption, and worsened cardiovascular disease. Sapolsky's framing in Why Zebras Don't Get Ulcers (1994) captures the core insight cleanly: a zebra escaping a lion turns the response on and, if it survives, off; a human can keep it on for months by worrying about a mortgage, so a system built for three minutes of emergency runs for three years and does the damage Selye predicted. The Lazarus and Folkman coping distinction, meanwhile, gave psychology a tractable, testable vocabulary that underpins much of stress-management practice, and Taylor's tend-and-befriend corrected a genuine and consequential bias in the evidence base.

The strongest case against it

The sharpest criticism came from within the field. Bruce McEwen (1938 to 2020) argued that Selye's model was too simple in two ways. First, "nonspecific" overstated the uniformity: different stressors do not produce identical responses, and the response is tuned to the challenge. Second, and more importantly, the notion of a fixed pool of "adaptive energy" that eventually runs out was never demonstrated. McEwen replaced it with allostasis (stability achieved through change) and allostatic load (the cumulative wear from repeated or poorly regulated stress responses, including responses that fail to shut off or fire when no threat is present). On this view the harm is not depletion but dysregulation, and the villain is often the response that will not turn off rather than one that has run dry.

Selye's own history invites suspicion too: much of his later career was funded by the tobacco industry, which found his diffuse notion of "stress" useful for deflecting attention from smoking as a specific cause of disease. The very nonspecificity that made his theory grand also made it convenient for muddying causation.

Taylor's tend-and-befriend has been criticized for leaning heavily on oxytocin, whose real effects are more context-dependent than early accounts implied (it can sharpen in-group loyalty and out-group wariness rather than simply soothe), and for risking the reinforcement of gender stereotypes it set out to correct. And the Lazarus and Folkman coping dichotomy, though useful, is contested: the two strategies interact rather than divide cleanly, and later researchers added categories (avoidant coping, meaning-focused coping) the original two-way split omitted. Measuring appraisal remains hard, and self-reported coping correlates only loosely with what people actually do.

Where it stands now

The core physiology is settled science and taught in every introductory course: the SAM and HPA axes, adrenaline and cortisol, the acute response and its costs when chronic. Selye's general adaptation syndrome survives as the historical scaffold and a useful teaching model, but the field now speaks in McEwen's language of allostatic load, which better fits the data on how chronic stress accumulates. Fight-or-flight is no longer treated as the whole human repertoire; tend-and-befriend is established enough to appear in textbooks, even as its mechanisms are still debated. Coping research has moved from clean typologies toward flexibility, the capacity to match strategy to situation, as the better predictor of who fares well. The through-line from a century of work is durable: stress is a real, physical, adaptive program, and most of the damage it does to modern humans comes not from the threats themselves but from a response that evolved for emergencies and rarely gets to end.

Test yourself

Bring to mind a stressor sitting on you right now. Ask first whether it is something you can actually change. If it is, notice whether you are spending your energy on solving it or on managing how it makes you feel, and whether that matches what the situation allows. Then ask the harder question: if this pressure has been with you for weeks, your body has been running its emergency program the whole time. The response was built to fire and then stop. What, for you, would count as letting it stop?