They Only Studied the Males
The fight-or-flight model of stress was built on male cats and mostly male volunteers. A rival account called tend-and-befriend, and a 2015 NIH rule requiring sex as a research variable, were both meant to fix that, and neither has worked as cleanly as advertised.
By Teona Abuladze · 7 min read
In 1914 Walter Bradford Cannon published a paper in the American Journal of Physiology under a title that reads like a verdict: “The Emergency Function of the Adrenal Medulla in Pain and the Major Emotions.” It ran in volume 33, pages 356 to 372. The following year he expanded it into a book, Bodily Changes in Pain, Hunger, Fear and Rage, published by D. Appleton in New York.
The animals in Cannon’s Harvard laboratory were cats. He watched a frightened cat’s stomach stop moving, then traced the chemistry behind it, the adrenal secretion and the blood pushed toward the muscles. The body, he argued, was readying itself for a physical emergency.
The compact phrase came later. Cannon’s 1915 text describes a body readied for struggle or for running away; the two-beat catchphrase hardened over subsequent editions and a century of textbooks. By 2004 the psychiatrist H. Stefan Bracha was arguing in CNS Spectrums that it was incomplete even on its own terms, since freezing and fainting are at least as common under acute threat.
So fight-or-flight was never a complete account of anything. It described what happens to a certain kind of body under a certain kind of threat, and the bodies were male.
Seventeen percent of participants, not seventeen percent of studies
In July 2000, six psychologists at UCLA published a paper in Psychological Review, volume 107, issue 3, pages 411 to 429. Shelley E. Taylor was the lead author, with Laura Cousino Klein, Brian P. Lewis, Tara L. Gruenewald, Regan A. R. Gurung and John A. Updegraff. The title was blunt: “Biobehavioral Responses to Stress in Females: Tend-and-Befriend, Not Fight-or-Flight.”
Their opening move was a head count. “Prior to 1995,” they wrote, “women constituted about 17% of participants in laboratory studies of physiological and neuroendocrine responses to stress.”
The figure has been repeated for twenty-five years, usually wrong. You will find it online as the claim that only 17 percent of stress studies included women. Taylor’s team did not say that. Seventeen percent was the share of human bodies in the room, not the share of experiments that admitted any woman at all. One study with 200 men and 40 women drags that figure down while excluding nobody.
The team also ran their own tally. They identified 200 studies of physiological and neuroendocrine responses to an acute experimental stressor conducted between 1985 and the time of writing, using 14,548 participants, 66 percent male and 34 percent female. The imbalance was real and large. It was also improving, and the paper says so.
Seventeen percent was the share of human bodies in the room, not the share of experiments that admitted any woman at all.
The estrous cycle excuse
Human volunteers were only half the problem. The animals were worse.
Taylor’s group stated the justification plainly as they found it in the literature: “because females have greater cyclical variation in neuroendocrine responses (due to the reproductive cycle), their data present a confusing and often uninterpretable pattern of results.”
Annaliese Beery and Irving Zucker put numbers on the result in 2011, in Neuroscience and Biobehavioral Reviews, volume 35, issue 3, pages 565 to 572. Surveying ten biological fields for the year 2009, they found male bias in eight of them. In neuroscience, single-sex studies ran 5.5 male for every one female. In pharmacology the ratio was five to one. In immunology, more than 60 percent of papers did not bother to say what sex the animals were.
The variability argument turned out to be false. In 2016, Jill B. Becker, Brian J. Prendergast and Jing W. Liang published a meta-analysis in Biology of Sex Differences, volume 7, article 34, drawing on 311 articles and more than 6,000 extracted data points. Their finding: female rats were not more variable than male rats on behavioural, electrophysiological, neurochemical or histological measures, and they were not more variable at any stage of the estrous cycle. “Even when female rats are used in neuroscience experiments without regard to the estrous cycle stage,” they wrote, “their data are not more variable than those of male rats.”
Generations of researchers left females out to protect their error bars. The error bars did not need protecting.
What the 2015 rule actually says
In May 2014, Janine Austin Clayton, director of the NIH Office of Research on Women’s Health, and Francis Collins, then NIH director, used a comment piece in Nature, volume 509, pages 282 to 283, to announce that this was going to change.
The policy itself arrived on 9 June 2015 as notice NOT-OD-15-102, “Consideration of Sex as a Biological Variable in NIH-funded Research.” It applied to applications due from 25 January 2016 onward. The operative sentence is short: “NIH expects that sex as a biological variable will be factored into research designs, analyses, and reporting in vertebrate animal and human studies.” Applications proposing to study only one sex must supply “strong justification from the scientific literature, preliminary data, or other relevant considerations.”
Note what that is. An expectation attached to grant review, with no statutory penalty behind it. And it asks for analysis, not just inclusion, which is the part that has caused all the trouble since.
Counting who actually ran the analysis
Three audits have now looked at whether the rule changed anything.
Nicole Woitowich, Annaliese Beery and Teresa Woodruff published a ten-year comparison in eLife in 2020, volume 9, article e56344. Across nine biological disciplines, studies using both sexes rose from 28 percent of articles in 2009 to 49 percent in 2019. Good news, and then the rest of the sentence: among the 356 studies that used both sexes in 2019, only 42 percent analysed the data by sex. In 2009 the equivalent figure had been 50 percent. Inclusion went up. Analysis went down.
Yesenia Garcia-Sifuentes and Donna L. Maney went further in eLife in 2021, volume 10, article e70817, reading 147 papers from 2019 that included both sexes. Of the articles claiming sex-specific effects, 70 percent presented no statistical evidence for the claim. Only 29 percent of factorial designs tested for an interaction between sex and treatment. Thirty-seven percent pooled male and female data, and just under half of those did it without first checking whether pooling was warranted. The authors’ warning is worth quoting exactly: the difference between “significant” and “not significant” is not itself statistically significant.
Their conclusion: sex-specific effects are probably over-reported, and a policy that pushes sex into every study without pushing statistical training could reduce rather than improve rigour.
The most recent audit, by J. H. Warden, M. Parangalan, L. J. Welty and colleagues, appeared in Communications Medicine on 27 April 2026, volume 6, article 208. Working from R01 awards made in fiscal years 2017 and 2018 and 574 resulting papers, they found 61 percent included both sexes and 44 percent of those dual-sex studies actually ran a sex-based analysis. Nearly twelve years after the Nature announcement, more than half of the studies that go to the trouble of including women and female animals still never compare them to anyone.
Inclusion went up. Analysis went down.
Where tend-and-befriend gets thin
Taylor’s replacement model says that under stress, females protect and quiet offspring and seek out other females, and that oxytocin, working with estrogen and endogenous opioid peptides, is doing it. Oxytocin is released under stress and its calming effects are enhanced by estrogen, while androgens appear to blunt them.
It is a good hypothesis. It is also, as Taylor’s own paper says in its final pages, not established. “Some of the points in the preceding argument remain conjectural,” the authors wrote. Oxytocin and endogenous opioids are “sufficient to be considered credible hypotheses but they are not definitively established.” Popular retellings drop that paragraph.
Four problems have grown since.
The oxytocin literature is in poor shape. Hasse Walum, Irwin Waldman and Larry Young reported in Biological Psychiatry in 2016, volume 79, issue 3, pages 251 to 257, that the median sample in intranasal oxytocin studies of healthy subjects was 49 people, and 26 in clinical trials. Average statistical power was 16 percent and 12 percent respectively. Of 33 papers they examined, 88 percent reported at least one positive finding and only three mentioned correcting for multiple comparisons, with an average of eleven tests per paper. Their estimate of positive predictive value was no higher than 24 percent. Young runs one of the field’s leading oxytocin laboratories, which makes the paper hard to dismiss.
Oxytocin also does not behave like a kindness drug. Carsten De Dreu and colleagues showed in PNAS in 2011, volume 108, issue 4, pages 1262 to 1266, that intranasal oxytocin promoted in-group favouritism and derogation of out-groups. Anne Campbell, reviewing the human evidence in Personality and Social Psychology Review in 2010, volume 14, issue 3, pages 281 to 295, argued that oxytocin’s effect depends on where a person already sits between trust and fear, and warned that extrapolation from rodents is unsafe because receptor distribution varies substantially across mammalian species.
Then there is the finding that the behaviour is not female-specific. Bernadette von Dawans, Urs Fischbacher, Clemens Kirschbaum, Ernst Fehr and Markus Heinrichs published in Psychological Science in 2012, volume 23, issue 6, pages 651 to 660, a study of 67 male students at the University of Zurich. Stressed men showed more trust and more sharing than unstressed controls, with no increase in antisocial behaviour. The authors describe this as evidence for the tend-and-befriend hypothesis. Read another way, it is evidence that the hypothesis was never about women.
And the essentialism objection is serious. A theory that says women are hormonally built to tend and affiliate lands in a culture that has spent centuries insisting women are naturally suited to caregiving, and the theory will be read in that light whatever its authors intend. Sarah S. Richardson, writing in Philosophy, Theory, and Practice in Biology in 2021, volume 13, article 9, argues that the NIH policy itself encodes sex as a single fixed binary variable across every research context, and that overemphasising sex difference “can lead to sex stereotyping in the clinic and in the laboratory,” while hiding variables such as age or body size that may be what actually explains the data.
Taylor received the National Medal of Science at the White House on 24 October 2023. Her model remains one of the most cited in health psychology, and the mechanism at its centre still rests on animal work and small human samples, exactly as she said in 2000 that it did.
What is not in dispute is the original complaint. For most of the twentieth century the emergency-response literature was assembled from male animals and mostly male undergraduates, and the results were reported as facts about human beings. That part is settled. The replacement is not.
If stress or anxiety is affecting your daily life, a general practitioner or a licensed therapist is the right place to start, and international directories of crisis lines are maintained by Befrienders Worldwide and the International Association for Suicide Prevention.
Teona Abuladze is the founder and editor of Shansi Magazine.
Sources: Walter B. Cannon, “The Emergency Function of the Adrenal Medulla in Pain and the Major Emotions,” American Journal of Physiology 33(2), 1914, 356–372 — https://journals.physiology.org/doi/abs/10.1152/ajplegacy.1914.33.2.356; Walter B. Cannon, “Bodily Changes in Pain, Hunger, Fear and Rage,” D. Appleton, 1915 — https://archive.org/details/bodilychangesinp00cannrich; H. S. Bracha, “Freeze, Flight, Fight, Fright, Faint: Adaptationist Perspectives on the Acute Stress Response Spectrum,” CNS Spectrums 9(9), 2004, 679–685 — https://web-archive.southampton.ac.uk/cogprints.org/5014/1/2004_C.N.S_Five_Fs_of_FEAR–Freeze_Flight_Fight_Fright_Faint.pdf; S. E. Taylor, L. C. Klein, B. P. Lewis, T. L. Gruenewald, R. A. R. Gurung and J. A. Updegraff, “Biobehavioral Responses to Stress in Females: Tend-and-Befriend, Not Fight-or-Flight,” Psychological Review 107(3), 2000, 411–429 — https://pubmed.ncbi.nlm.nih.gov/10941275/ and full text at https://taylorlab.psych.ucla.edu/wp-content/uploads/sites/5/2014/10/2000_Biobehavioral-responses-to-stress-in-females_tend-and-befriend.pdf; A. K. Beery and I. Zucker, “Sex Bias in Neuroscience and Biomedical Research,” Neuroscience and Biobehavioral Reviews 35(3), 2011, 565–572 — https://scholarworks.smith.edu/nsc_facpubs/12/; J. B. Becker, B. J. Prendergast and J. W. Liang, “Female rats are not more variable than male rats: a meta-analysis of neuroscience studies,” Biology of Sex Differences 7:34, 2016 — https://link.springer.com/article/10.1186/s13293-016-0087-5; J. A. Clayton and F. S. Collins, “Policy: NIH to balance sex in cell and animal studies,” Nature 509(7500), 2014, 282–283 — https://www.nature.com/articles/509282a; National Institutes of Health, “Consideration of Sex as a Biological Variable in NIH-funded Research,” notice NOT-OD-15-102, 9 June 2015 — https://grants.nih.gov/grants/guide/notice-files/not-od-15-102.html; N. C. Woitowich, A. Beery and T. Woodruff, “Meta-Research: A 10-year follow-up study of sex inclusion in the biological sciences,” eLife 9:e56344, 2020 — https://elifesciences.org/articles/56344; Y. Garcia-Sifuentes and D. L. Maney, “Reporting and misreporting of sex differences in the biological sciences,” eLife 10:e70817, 2021 — https://elifesciences.org/articles/70817; J. H. Warden, M. Parangalan, L. J. Welty et al., “Incorporation of the National Institute of Health (NIH) sex as a biological variable policy by R01 grant awardees,” Communications Medicine 6:208, 27 April 2026 — https://www.nature.com/articles/s43856-026-01547-0; H. Walum, I. D. Waldman and L. J. Young, “Statistical and Methodological Considerations for the Interpretation of Intranasal Oxytocin Studies,” Biological Psychiatry 79(3), 2016, 251–257 — https://pmc.ncbi.nlm.nih.gov/articles/PMC4690817/; C. K. W. De Dreu, L. L. Greer, G. A. Van Kleef, S. Shalvi and M. J. J. Handgraaf, “Oxytocin promotes human ethnocentrism,” PNAS 108(4), 2011, 1262–1266 — https://www.pnas.org/doi/abs/10.1073/pnas.1015316108; A. Campbell, “Oxytocin and Human Social Behavior,” Personality and Social Psychology Review 14(3), 2010, 281–295 — https://journals.sagepub.com/doi/10.1177/1088868310363594; B. von Dawans, U. Fischbacher, C. Kirschbaum, E. Fehr and M. Heinrichs, “The Social Dimension of Stress Reactivity: Acute Stress Increases Prosocial Behavior in Humans,” Psychological Science 23(6), 2012, 651–660 — https://journals.sagepub.com/doi/10.1177/0956797611431576; S. S. Richardson, “Sex Contextualism,” Philosophy, Theory, and Practice in Biology 13:9, 2021 — https://scholar.harvard.edu/files/srichard/files/richardson-sex-rc2.pdf; UCLA Newsroom, “UCLA’s Shelley Taylor awarded National Medal of Science at the White House,” 2023 — https://newsroom.ucla.edu/releases/ucla-shelley-taylor-awarded-national-medal-of-science
Выбрать восемь тем и проверить актуальность
Исследовать и написать четыре статьи: Soul Art, Mind Psychology, Mind Now, Mind Money
Исследовать и написать четыре статьи: Body Health, Fashion, Food, Sports
Отредактировать и проверить факты во всех восьми текстах
5
Сделать восемь обложек 1760×1100
6
Рисерч событий для женщин в Майами: сентябрь–ноябрь 2026body-sports-acl-en.mdbody-food-protein-en.mdbody-fashion-synthetic-clothes-en.mdbody-health-endometriosis-en.mdmind-money-retirement-gap-en.mdmind-now-pay-gap-2026-en.mdmind-psychology-stress-research-en.mdsoul-art-ruth-asawa-en.mdbody-sports-acl.pngbody-food-protein.pngbody-fashion-synthetic-clothes.pngbody-health-endometriosis.pngmind-money-retirement-gap.pngmind-now-pay-gap-2026.pngmind-psychology-stress-research.pngsoul-art-ruth-asawa.pngbody-sports-acl.mdbody-food-protein.mdbody-fashion-synthetic-clothes.mdbody-health-endometriosis.mdmind-money-retirement-gap.mdmind-now-pay-gap-2026.mdmind-psychology-stress-research.mdsoul-art-ruth-asawa.md
Shansi Magazine
Connectors
Web search
Downloaded mind-now-pay-gap-2026_1.png Show in Finder
