THRIVING after the FALL
Book Two PAIRING after the FALL

PAIRING · Part I · Overlapping Bells

The numbers and sources

Chapter 3: The Endocrine Accent

The narrated edition tells this chapter with a handful of figures. Here is everything behind it: each figure with its study, its research document and its row in the log, the case for and the case against, where the sources disagree, and what is still open.

  • 79 open
  • 16 to check first

Read about the chapter · Its 79 rows in the log

Numbers to carry

The figures the chapter is built on, each with its research document and its row in the log.

  1. Testosterone reference ranges: men 10.4 to 34.7 nanomoles per litre, women 0.5 to 2.4, no overlap (BI-02). Row C03-15. Source: laboratory reference intervals, BI-02 names no laboratory, check Endocrine Society or Mayo reference intervals.
  2. Baseline testosterone and aggression: r = 0.054, confidence interval 0.028 to 0.080, 45 studies, 9,760 people, Geniole et al. 2020 (BI-03, AN-08). One report gives 0.14 (BI-02). Row C03-32, cross-reference Preface row P-22. Source: Geniole et al. 2020, Hormones and Behavior 123:104644.
  3. Fatherhood in Cebu: median testosterone down 26 per cent waking and 34 per cent evening, or 20.5 per cent in one report, over 4.5 years, sample given as 394 to 683, Gettler et al. 2011 (BI-02, BI-06, the BI digest). Row C03-25, cross-reference Preface row P-24. Source: Gettler et al. 2011, PNAS 108:16194 to 16199.
  4. Hormonal contraception and antidepressant use: 1.7 to 2.2 per 100 woman-years, relative risk 1.23 to 1.60 by method, 1,061,997 women, Skovlund et al. 2016 (BI-08, BI-10). Row C03-49. Source: Skovlund et al. 2016, JAMA Psychiatry 73:1154 to 1162.
  5. Cycle length: 13 per cent of more than 600,000 logged cycles exactly 28 days, mean 29.3, Bull et al. 2019 (BI-04). Row C03-20. Source: Bull et al. 2019, npj Digital Medicine 2:83.

Every figure

One row for each figure or finding in the data edition's chapter, whether or not the narrated edition kept it. Layer says whether it belongs to physiology, behaviour or culture. Status is how the finding stands in the research: robust, contested, open or fallen. It is not yet the log's verdict.

FigureWhat it measuresStudy (authors, year, venue, N)DocumentRowLayerStatus
Testosterone peak 100 to 400 ng/dL at 1 to 3 months (or a window of 1 to 6 months)Mini-puberty in male infantsKuiri-Hänninen et al. 2014, Hormone Research in Paediatrics 82:73 to 80BI-06, BI-01C03-01physiologycontested (window differs between documents)
Near zero by first birthday, low until pubertyMale infant testosterone after mini-pubertyAs BI-01 gives it, no primary namedBI-01C03-02physiologyrobust
FSH elevated until age 3 to 4Female infant gonadotropinKuiri-Hänninen et al. 2014BI-06C03-03physiologyrobust
SRY acts at weeks 6 to 7 via SOX9Testis determinationTextbook developmental biology as BI-01 gives itBI-01C03-04physiologyrobust
Ovaries require WNT4 and RSPO1Ovary determination, the female default is wrongAs BI-01 gives itBI-01C03-05physiologyrobust
Fetal testosterone surge weeks 8 to 24, DHT shapes external genitaliaPrenatal androgen exposureAs BI-01 gives itBI-01C03-06physiologyrobust
Puberty as a second organisational windowAdolescent brain organisationMcCarthy 2015, Nature Neuroscience (first name Margaret to confirm, possibly McCarthy and Nugent, Nat Neurosci 18:1 to 8)BI-01C03-07physiologyopen
Prenatal testosterone organises adult sexual behaviour in guinea pigsOrganising versus activating effectsPhoenix, Goy, Gerall and Young 1959, Endocrinology 65:369 to 382, guinea pigsAN-08, BI-01C03-08behaviouropen (laboratory given as University of Kansas, marked (to be checked) in the data edition)
Toy play shifted about 60 per cent of the way toward the male mean, cohorts of 40 to 125CAH girls' play behaviourHines, Berenbaum and successors as BI-01 summarises themBI-01C03-09behaviourrobust
81 CAH children, 72 siblings, parents encouraged female-typical play more in CAH girlsParental socialisation against the shiftPasterski et al. 2005, Child Development 76:264 to 278the AN digestC03-10behaviourrobust
Cohorts of 10 to 30, female-typical in every measureComplete androgen insensitivityAs BI-01 gives itBI-01C03-11behaviourrobust
18 raised as girls, 17 adopted male identity at puberty5-alpha-reductase deficiency, Dominican RepublicImperato-McGinley et al. 1974, Science 186:1213 to 1215 (the 18 and 17 may come from the 1979 NEJM follow-up)BI-01C03-12behaviouropen
Castrated guinea pigs return to their own rank on an identical doseTissue sensitivity, not blood levelGrunt and Young 1953, Journal of Comparative and Physiological Psychology 46:138 to 144AN-08C03-13behaviourrobust
10 per cent of normal testosterone restores copulation in castrated ratsThreshold effectDamassa et al. 1977, EndocrinologyAN-08C03-14physiologyrobust
Men 10.4 to 34.7 nmol/L, women 0.5 to 2.4, no overlapAdult testosterone reference rangesLaboratory reference intervals, no laboratory namedBI-02C03-15physiologyrobust as a range, source open
138 to 690 nmol/L in both sexesCortisol reference rangeAs BI-02 gives itBI-02C03-16physiologyrobust
Men 36.7 to 146.8 pmol/L, women 55 to 1,468 across the cycleOestradiol reference rangesAs BI-02 gives itBI-02C03-17physiologyrobust
Men below 3.18 nmol/L, women 0.3 to 63.6Progesterone reference rangesAs BI-02 gives itBI-02C03-18physiologyrobust
Peak 06:00 to 08:00, tied to REM sleepDaily testosterone rhythmAs BI-02 gives itBI-02C03-19physiologyrobust
124,648 users, more than 600,000 cycles, 13 per cent exactly 28 days, mean 29.3, luteal 12.4 daysMenstrual cycle lengthBull et al. 2019, npj Digital Medicine 2:83 (first name Jonathan)BI-04C03-20physiologyrobust
34 studies, 640 women, 710 men, men higher cortisol peakStress response in the Trier Social Stress TestLiu et al. 2017, Psychoneuroendocrinology 82:26 to 37BI-02C03-21physiologyrobust
Women report more distress, men higher cortisolSex difference in stress displayKudielka and Kirschbaum 2005, Biological Psychology 69:113 to 132BI-06C03-22behaviourrobust
10 men, five-hour nights for a week, testosterone down 10 to 15 per centSleep restriction and testosteroneLeproult and Van Cauter 2011, JAMA 305:2173 to 2174BI-02, BI-05C03-23physiologycontested (BI-02 calls it robust, BI-05 fallen)
18 controlled studies, partial restriction SMD minus 0.22 not significant, total deprivation about 25 per cent dropSleep loss and testosterone, pooledSu et al. 2021, Sleep Medicine, N strippedBI-05C03-24physiologycontested
Median testosterone down 26 per cent waking, 34 per cent evening (or 20.5 per cent), N 394, 465, 624 or 683Fatherhood and testosteroneGettler et al. 2011, PNAS 108:16194 to 16199, Cebu, saliva 2005 and 4.5 years laterBI-02, BI-06, DH-01, FA-03, CO-02, SY-04, DH-08C03-25physiologyopen on N and drop size, robust in direction
Higher baseline testosterone predicts partnering and fatherhoodSelection into fatherhoodGettler et al. 2011BI-02C03-26behaviourrobust
Fathers doing three or more hours of childcare a day have the lowest testosteroneDose by childcareGettler et al. 2011BI-06C03-27physiologyrobust
114 effects, 66 studies, partnered and fathers lowerPooled pair-bonding and testosteroneGrebe et al. 2019, Neuroscience and Biobehavioral Reviews 98:221 to 233 (first name Nicholas), effect sizes lostBI-05C03-28physiologyopen (effect sizes)
60 effects, more than 2,500 participants, winners rise more, larger in real venuesWinner and loser effectGeniole et al. 2017, Hormones and Behavior 92:37 to 50, d lostBI-05C03-29physiologyopen (effect size)
Challenge hypothesisTestosterone and seasonal breeding in birdsWingfield et al. 1990, American Naturalist 136:829 to 846AN-08C03-30behaviourrobust
Male to female interactions provoke larger testosterone pulses than male to maleChallenge hypothesis revisedGoymann, Moore and Oliveira 2019, BioScience (first name Wolfgang)AN-08C03-31behaviourrobust
45 studies, N 9,760, baseline r = 0.054 (0.028 to 0.080), men 0.071, women 0.002, manipulated 0.046 not significant, or r = 0.14Testosterone and aggressionGeniole et al. 2020, Hormones and Behavior 123:104644BI-03, AN-08, BI-02C03-32behaviouropen (0.054 versus 0.14), robust in being small
1,000 men, 11 mg intranasal testosterone, no effect on trust, risk, competitiveness or fairnessTestosterone and economic preferencesNadler, Zava, Camerer et al. 2024, PNAS, or Nave et al. 2018BI-03, BI-05C03-33behaviouropen (authors and year)
N 121, aggression up only in high-dominance, low-self-control menModerated testosterone effectCarré et al. 2017, Biological Psychiatry 82:249 to 256 (first name Justin)BI-03C03-34behaviourrobust
40 men, both punishment and generosity upTestosterone and social decisionsDreher et al. 2016, PNAS 113:11633 to 11638 (first name Jean-Claude)BI-12C03-35behaviourrobust, small sample
About 33 per cent lower than age-matched Americans, no age decline, 63 men, 30 per cent surge after football, 15 per cent at one hourTsimane testosteroneTrumble et al. 2012, Evolution and Human Behavior, and Proceedings of the Royal Society B (Trumble and Gurven)BI-10C03-36physiologyrobust
736-night census, about 110 lifetime menses against 400 to 450 modern, or 100 to 160 ancestralLifetime menstruation, Dogon, MaliStrassmann 1997, Current Anthropology 38:123 to 129 (first name Beverly), Eaton et al. 1994 for 100 to 160BI-04, BI-10C03-37physiologycontested between documents
Dogon men watched the hut as a paternity signalMenstrual hut as informationStrassmann 1992 and 1997BI-04C03-38culturerobust
134 effects, 50 studies found a shift, re-analysis of 58 reports found noneOvulatory shift in mate preferenceGildersleeve et al. 2014, Psychological Bulletin 140:1205, and Wood et al. 2014, Emotion Review 6:229 (first names Kelly, Wendy)BI-04C03-39behaviourfallen
N 584, more than 5,000 sessions, no hormone to face-preference linkOvulatory shift, large studyJones et al. 2018, Psychological Science (first name Benedict)BI-04C03-40behaviourrobust null
N 257, 1,028 sessions, no ovulatory shiftOvulatory shift, replicationStern et al. 2021 (first name Julia)BI-04C03-41behaviourrobust null
Oestradiol predicts desire positively, progesterone negativelyCycle and desireRoney and Simmons 2013, Hormones and Behavior 63:636 to 645 (first names James, Zachary)BI-04C03-42behaviourrobust
About 26,000 diary entries, mid-cycle desire peakCycle and desire, diariesArslan et al. 2021 (first name Ruben)BI-04C03-43behaviourrobust
Partners do not detect fertilityPartner perception of the cycleSchleifenbaum et al. 2022 (first name Lara)BI-04C03-44behaviourrobust
47.8 per cent pooled prevalence across 17 studiesPremenstrual symptomsDirekvand-Moghadam et al. 2014BI-04C03-45behaviourrobust
3 to 8 per cent, or about 5 per centPremenstrual dysphoric disorder prevalenceAs BI-04 and BI-03 give itBI-04, BI-03C03-46behaviourcontested between documents
22 women, leuprolide, add-back reproduces symptoms only in PMDD, symptoms return in month one and are gone by month threePMDD as a response to hormonal changeSchmidt and Rubinow 1998, NEJM 338:209 to 216, NIMH Bethesda, and the 2017 American Journal of Psychiatry follow-upBI-03, BI-04C03-47physiologyrobust
32,748 Dutch women, 8.9 days lost productivity a year, 80.7 per cent presenteeism, 20 per cent disclosed to an employerCost of menstrual symptoms at workSchoep et al. 2019, BMJ OpenBI-04C03-48culturerobust
1,061,997 Danish women 15 to 34, RR 1.23 combined pill, 1.34 progestogen-only, 1.40 hormonal coil, 1.60 ring, adolescents higher, absolute 1.7 to 2.2 per 100 woman-years, risk peaks about six months after startingHormonal contraception and first antidepressantSkovlund et al. 2016, JAMA Psychiatry 73:1154 to 1162, Copenhagen (first name Charlotte Wessel)BI-08, BI-10C03-49behaviourrobust in direction, causation contested
475,802 women, 6,999 suicide attempts, 71 suicides, RR 1.97 attempts, 3.08 suicide, peak at two monthsHormonal contraception and suicideSkovlund et al. 2018, American Journal of PsychiatryBI-08C03-50behaviourrobust in direction, causation contested
264,557 women, 7,354 sibling pairs, HR 1.71 in the first two yearsHormonal contraception and depression, sibling designJohansson et al. 2023, UK BiobankBI-08C03-51behaviourrobust in direction
RCT, N 340, lower general well-being, no difference on a depression scaleHormonal contraception, randomisedZethraeus et al. 2017, Fertility and Sterility 107:1238 (first name Niklas)BI-08C03-52behaviourrobust
Allopregnanolone from progesterone acts on GABA-A, synthetic progestins are not metabolised to itMechanismAs BI-10 gives itBI-10C03-53physiologyrobust as mechanism, causal weight open
SHBG up 80 to 300 per cent, about 15 per cent lower desire, more than 80 per cent no changePill and desirePastor et al. 2013BI-08C03-54physiologyrobust
365 couples, women who met partners on the pill lose attraction off itCongruency hypothesisRoberts et al. 2014, Psychological Science (first name S. Craig), Stern 2019 and 2021 as failed replicationsBI-10, BI-08C03-55behaviourcontested
N 157 (52 women, 105 men), intrinsic period 24.09 versus 24.19 hours, 35 versus 14 per cent under 24 hoursCircadian period by sexDuffy et al. 2011, PNAS 108 Suppl 3:15602 to 15608, Boston (first name Jeanne)BI-09C03-56physiologyrobust
169 studies, women more slow-wave sleep, insomnia RR 1.41, men more apnoeaSleep by sexBoulos et al. 2019, Sleep, and Zhang and Wing 2006BI-09C03-57physiologyrobust
ASHRAE 55 calibrated to a 70 kg male, overestimates female heat production by up to 35 per cent, 16 womenOffice thermal comfortKingma and van Marken Lichtenbelt 2015, Nature Climate Change 5:1054 to 1056, Maastricht, critic Bjarne Olesen on clothing insulationBI-09C03-58physiologycontested, critic's first name open
Preferred temperature women 24 to 25 degrees C, men 21 to 22, dissatisfaction ratio 1.74, luteal set-point plus 0.3 to 0.5 degreesThermal preferenceKarjalainen 2011BI-09C03-59physiologyrobust
Pain threshold d = 0.66, tolerance d = 0.79, about three-quarters overlap, men's threshold lower with a female examinerPain by sexRiley 1998, Racine 2012 (the overlap is the narrator's computation from d)BI-09C03-60physiologyrobust
Men about 60 per cent more muscle, mostly upper body, women about 1.6 times the body fatBody compositionLassek and Gaulin 2009Preface rows P-17, P-20, SY-01C03-61physiologyrobust
Grip strength 541 N men, 329 N women, N 2,187, 90 per cent of women below 95 per cent of men, gap above two standard deviationsGrip strengthLeyk et al. 2007, European Journal of Applied Physiology 99:415 to 421HC-04, SY-04, SY-02Preface rows P-15, P-16physiologyrobust
About 100 women (60 in one document, 120 in another), 0.5 mg sublingual testosterone, fairer offers, belief lowered offersTestosterone, belief and fairnessEisenegger et al. 2010, Nature 463:356 to 359, ZurichSO-07, BI-05Preface row P-26behaviourrobust in finding, N differs between documents
NHANES 4,207 men 15 to 39, 605.39 to 451.22 ng/dL 1999 to 2016, normal weight 664.79 to 529.24Secular decline in testosteroneLokeshwar et al. 2020 or 2021, European Urology Focus (first name Soum)BI-05, BI-10C03-62physiologyopen (year)
Minus 51.6 per cent 1973 to 2018, 223 studies, 53 countries, minus 2.64 per cent a year after 2000Sperm concentration declineLevine et al. 2022, Human Reproduction Update, critics Boulicault et al. 2021 and Allan PaceyBI-10C03-63physiologycontested
Thelarche 0.24 years earlier per decade, 30 studiesEarlier puberty in girlsEckert-Lind et al. 2020, JAMA Pediatrics (first name Camilla)BI-10C03-64physiologyrobust
Prenatal phthalates shorten male anogenital distanceEndocrine disruptionvan den Driesche reviews as BI-10 gives themBI-10C03-65physiologyrobust
898 people (440 trans men, 468 trans women), 36 months, no correlation serum testosterone to state anger at 3, 12 or 36 months, transient rise at three months, anger tied to baseline psychopathologyHormone therapy and angerDefreyne et al. 2018, ENIGI, Endocrine Abstracts 57 and PubMed 30822410, Ghent (city not in BI-12)BI-12C03-66behaviourrobust
28 men, 1,000 mg testosterone undecanoate, POMS anger 7.48 to 10.71, no change in behavioural aggressionSupraphysiological testosterone in cisgender menO'Connor, Archer and Wu 2004, JCEM 89:2837 to 2845BI-12C03-67behaviourrobust
26 interviews (12 feminising, 14 masculinising), crying threshold lower on oestrogen and higher on testosteroneFirst-person accounts of hormone therapyvan de Grift et al., Journal of Sexual Medicine, dated 2017 and 2026 in the same document (first name Tim)BI-12C03-68behaviouropen (year)
Emotional lability and weepiness on androgen deprivationProstate cancer patientsProQuest thesis as BI-12 cites it, primary to findBI-12C03-69behaviouropen (source)
73.9 per cent of trans men report increased sexual desire, solitary not dyadicTestosterone and desireENIGI, Wierckx et al.BI-12C03-70behaviourrobust
Total brain volume up 31 mL in four monthsTestosterone and brain volumeHulshoff Pol et al. 2006, European Journal of Endocrinology 155:S107 to S114, Utrecht (first name Hilleke)BI-12C03-71physiologyrobust
Spatial up, verbal fluency down, spatial half replicated onlyTestosterone and cognitionvan Goozen et al. 1995 and Slabbekoorn et al. 1999, Psychoneuroendocrinology 20:343 and 24:423BI-12C03-72behaviourcontested
Treatment-naive trans people perform like lived gender on mental rotationCognition before hormonesBurke et al. 2016 (first name Sarah)BI-12C03-73behaviourrobust
27 RCTs, 1,890 men, depressive symptoms reduced, against 5,601 men below 12 nmol/L with no BDI improvementTestosterone as antidepressantWalther et al. 2019, JAMA Psychiatry, and Hudson et al. 2023 individual-patient meta-analysisBI-12C03-74behaviourcontested
Humans emancipated from hormonal control of matingThe critics' frameFine, Delusions of Gender, Jordan-Young, Brain Storm, Wallen 1999, Hormones and Behavior 36:236 to 250AN-08, BI-12C03-75culturerobust as argument
Plasma oxytocin 1.53 versus 4.53 pg/mL by assayMeasurement disagreementAs BI-02 gives itBI-02C03-76physiologyrobust
About 1 per cent a year after 30Male testosterone decline with ageHarman 2001, Baltimore, as BI-06 cites itBI-06C03-77physiologyrobust
99.7 per cent of the variance, a third of one per cent, 2 per centArithmetic from r = 0.054 and r = 0.14Narrator's computation, not a sourcenoneC03-78behaviourrobust as arithmetic
Narrator 51 at writingBiography, the data edition working notes onlyBorn 1975concept noteC03-79nonerobust

The case for and the case against

The case the chapter makes

  • The sexes differ absolutely in one measurable ingredient, testosterone, whose adult reference ranges do not overlap (BI-02, C03-15), and share the rest of their chemistry on different timetables (BI-02, C03-16 to C03-19).
  • Hormones organise tissue before birth and activate it later, the frame from Phoenix et al. 1959 (AN-08, BI-01, C03-08), confirmed in humans by CAH, CAIS and the Dominican families (BI-01, the AN digest, C03-09 to C03-12).
  • Above a threshold the level of a hormone tells you little about the person, from Grunt and Young 1953 and Damassa 1977 (AN-08, C03-13, C03-14) and from the PMDD add-back experiment (BI-03, BI-04, C03-47).
  • Testosterone responds to life, falling with hands-on fatherhood in Cebu (BI-02, BI-06, C03-25 to C03-27) and across pooled studies (BI-05, C03-28), rising with winning and with courtship (BI-05, AN-08, C03-29 to C03-31).
  • Testosterone does not drive aggression in any simple way, r about 0.05 at baseline and no effect when given (BI-03, AN-08, C03-32, C03-33), and belief about the hormone moves behaviour where the hormone does not (Eisenegger, P-26).
  • The cycle's effects on the mind are smaller than the culture suggests, the ovulatory-shift findings fell (BI-04, C03-39 to C03-41) and what survived is desire tracking oestradiol (BI-04, C03-42 to C03-44).
  • The pill's effect on mood is real in direction and small in absolute size, one extra woman in two hundred (BI-08, BI-10, C03-49 to C03-52).
  • People who changed one hormone describe moved thresholds for tears and desire and an unchanged self, consistently across ENIGI, O'Connor and van de Grift (BI-12, C03-66 to C03-70).

The case against

  • The correlations are so weak that the chapter describes a rumour: r = 0.054 (BI-03, AN-08, C03-32), a thousand men dosed with no effect (BI-03, C03-33), 898 people in transition with no hormone to anger link (BI-12, C03-66).
  • What moves, moves with culture: the German border in the Preface, the pill's one in two hundred (BI-10, C03-49), the cycle effects that fell with sample size (BI-04, C03-39), Eisenegger's belief effect (P-26).
  • Fine, Jordan-Young and Wallen: the organising-activating model imports rodent determinism into a species emancipated from hormonal control (AN-08, BI-12, C03-75).
  • Measurement: oxytocin assays disagree threefold (BI-02, C03-76), the documents disagree on Gettler's N by nearly a factor of two (C03-25) and on Geniole's r by a factor of three (C03-32).
  • From the data edition's Contested list: sleep restriction, Leproult against the 2021 meta-analysis (C03-23, C03-24).
  • The sperm decline, Levine against Boulicault and Pacey (C03-63).
  • The congruency hypothesis, Roberts against Stern (C03-55).
  • The thermostat, Kingma against Olesen (C03-58).
  • Causation in the pill data, registers against trials (C03-49 to C03-52).
  • Testosterone for depression, 2019 against 2023 (C03-74).
  • Fallen and named as such in the data edition: the ovulatory face preference (C03-39 to C03-41), and testosterone and mind-reading, Nadler 2019 (BI-12, listed in the data edition's Contested list, not in its prose).

Where the sources disagree

  • Gettler 2011 N: 394 (SY-04), 465 (BI-02), 624 (DH-01, DH-08), 683 (BI-06). Drop: 26 per cent waking and 34 per cent evening (BI-02, DH-01, FA-03) against 20.5 per cent (CO-02). The narrated edition says roughly a quarter to a third. Row C03-25.
  • Eisenegger 2010 N: 60 (BI-05), 120 (SO-07). The data edition used no number. The narrated edition says about a hundred women. Preface row P-26.
  • Geniole 2020 r: 0.054 (BI-03, AN-08) against 0.14 (BI-02). The narrated edition says about 0.05. Row C03-32.
  • The 1,000-man intranasal trial: Nadler, Zava, Camerer et al. 2024, PNAS (BI-03) against Nave et al. 2018 (BI-05). The narrated edition names nobody. Row C03-33.
  • Wu 2010 EMAS journal: JCEM (BI-02) against NEJM (BI-06). Not used in either edition.
  • De Dreu 2010 N: 480 (BI-06) against 870 (BI-03). Not used in either edition.
  • Dogon lifetime menses: 110 (BI-04) against 100 to 160 ancestral (BI-10, citing Eaton et al. 1994). The narrated edition gives no number. Row C03-37.
  • Mini-puberty window: 1 to 3 months (BI-06) against 1 to 6 months (BI-01). The narrated edition says the first months. Row C03-01.
  • Leproult 2011 sleep finding: robust (BI-02) against fallen (BI-05). The narrated edition calls it contested and small. Rows C03-23, C03-24.
  • Congruency hypothesis: replicated (BI-10) against failed to replicate, Stern 2019 and 2021 (BI-08). The narrated edition calls it contested. Row C03-55.
  • PMDD prevalence: 3 to 8 per cent (BI-04) against about 5 per cent (BI-03). The narrated edition says a small minority. Row C03-46.
  • Lokeshwar year: 2020 (BI-10) against 2021 (BI-05). The narrated edition names neither. Row C03-62.
  • van de Grift year: 2017 and 2026 in the same document (BI-12). Row C03-68.
  • Sperm decline: Levine 2022 against Boulicault 2021 and Pacey. The narrated edition says fought over. Row C03-63.
  • Thermostat: Kingma 2015 against Olesen on clothing insulation. The narrated edition keeps both sides without numbers. Row C03-58.
  • Testosterone for depression: Walther 2019 against Hudson 2023. The narrated edition says works in one pooled analysis and fails in a larger one. Row C03-74.
  • Pill and mood causation: registers (Skovlund, Johansson) against the Zethraeus trial. The narrated edition says registers say risk, trials say mood. Rows C03-49 to C03-52.

Open questions

marks for checking from the data edition, with rows:

  • University of Kansas as the 1959 guinea-pig laboratory. Row C03-08.
  • Gettler sample size, 394, 465, 624 or 683. Row C03-25.
  • Grebe 2019 effect sizes, lost in extraction and absent from the PDF. Row C03-28.
  • Geniole 2017 winner-effect d, lost in extraction and absent from the PDF. Row C03-29.
  • Geniole 2020 r = 0.054 against 0.14. Row C03-32.
  • The 1,000-man trial as Nadler 2024 PNAS against Nave 2018. Row C03-33.
  • Bjarne Olesen as the ASHRAE critic, BI-09 gives the surname only. Row C03-58.

Open questions from the data edition's working notes:

  • Whether the Dominican families belong here or in a later chapter on identity.
  • Whether the modern-environment material should grow (BI-10 has PFAS, blue light and allostatic load, all cut) or move to chapter 9. The narrated edition keeps one paragraph.
  • Whether to name the narrator's age in the kitchen-table close (born 1975, so 51 at writing). The narrated edition does not.
  • Whether the trans and prostate-cancer memoirists should get a full scene rather than a survey. The narrated edition gives them a section and lets them carry the close.

Figures not found in the PDFs either, per the data edition: Geniole 2017 winner-effect d, Geniole 2020 point estimates in the AN-08 and BI-05 tables (present as text in BI-02 and BI-03), Grebe 2019 effect sizes, Su 2021 sample size, ENIGI STAXI-2 point changes, ENIGI solitary-desire correlation, toy-preference meta-analytic d in AN-08, vervet colony N in AN-08.

Every row in the log (C03-01 to C03-79) was OPEN on 2026-09-30, meaning not yet checked against the primary paper. Rows marked with a warning there: C03-08, C03-12, C03-15, C03-20, C03-25, C03-32, C03-33, C03-37, C03-47, C03-49, C03-50, C03-56, C03-58, C03-62, C03-63, C03-66.

The documents

BI-01: how a body becomes male or female, the natural experiments, and the organising-activating frame. BI-02: the two endocrine systems compared, reference ranges, daily and monthly rhythms, stress. BI-03: hormones and behaviour, Geniole 2020, Carré, Nadler, oxytocin. BI-04: the menstrual cycle and mind, Strassmann, PMDD, what fell in replication. BI-05: male hormonal dynamics, winner effect, sleep, Grebe 2019. BI-06: the life course from mini-puberty to old age, TSST, Gettler childcare hours. BI-08: hormonal contraception, Skovlund, Zethraeus, congruency. BI-09: bodily differences in a shared flat, clock, thermostat, pain, immunity. BI-10: the modern endocrine environment, Levine, Travison, Tsimane, thelarche. BI-12: first-person evidence from hormone therapy, ENIGI, O'Connor, van de Grift, Hulshoff Pol. AN-08: hormones and behaviour in animals, Phoenix 1959, Grunt and Young, challenge hypothesis, Wallen.

Its documents in the Library

159 160 161 162 163 164 166 167 168 170 178 181 188 196 209 217 233 234 235 237