PAIRING · Part II · The Campfire Contract
The numbers and sources
Chapter 8: The Father’s Falling Tide
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.
- 78 open
- 31 to check first
Numbers to carry
The figures the chapter is built on, each with its research document and its row in the log.
- 89 first-time parents: amygdala-to-cortex connectivity in fathers rises in a straight line with hours of childcare, and primary-caregiving gay fathers show both the "maternal" and the "paternal" network (Abraham, Feldman et al. 2014, PNAS, via FA-01). Rows C08-11, C08-12. Source: Abraham, Hendler, Shapira-Lichter, Kanat-Maymon, Zagoory-Sharon and Feldman 2014, PNAS 111:9792 to 9797.
- 25 mothers scanned before and after a first pregnancy: 100 per cent classifiable by grey-matter reduction, 91.67 per cent still at six years, oestradiol 3.42 to 12,400 pg/ml (Hoekzema et al. 2016, Nature Neuroscience, and Martínez-García et al. 2021, via FA-02). Rows C08-14, C08-16, C08-17, C08-19. Source: Hoekzema et al. 2016, Nature Neuroscience (online December 2016, print 2017), and Martínez-García et al. 2021, Brain Sciences.
- 38 first-time fathers: cortical volume reductions, largest in the precuneus, d above 0.5, tracking engagement and predicting poorer sleep and mood (Saxbe et al. 2024, Cerebral Cortex, via FA-02). Rows C08-01, C08-25, C08-26. Source: Saxbe, Martínez-García et al. 2024, Cerebral Cortex.
- Paternal perinatal depression 10.4 per cent across 43 studies and 28,004 participants, 25.6 per cent at three to six months, correlated 0.308 with maternal depression (Paulson and Bazemore 2010, JAMA, via FA-02 and FA-01). Rows C08-28, C08-29, C08-30. Source: Paulson and Bazemore 2010, JAMA 303:1961 to 1969.
- Ovarian reserve 12 per cent of peak at 30 and 3 per cent at 40 across 325 ovaries (Wallace and Kelsey 2010, PLoS ONE), and a 90 per cent chance of two children means starting by 27, or 31 with IVF (Habbema et al. 2015, Human Reproduction), both via FA-05. Rows C08-62, C08-64. Sources: Wallace and Kelsey 2010, PLoS ONE 5:e8772. Habbema, Eijkemans, Leridon and te Velde 2015, Human Reproduction 30:2215 to 2221.
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.
| Figure | What it measures | Study (authors, year, venue, N) | Document | Row | Layer | Status |
|---|---|---|---|---|---|---|
| 38 first-time fathers, scanned in the partner's pregnancy and at six months, own-baby versus unknown-infant pictures | Design of the opening scene | Saxbe, Martínez-García et al. 2024, Cerebral Cortex, N 38 | FA-02 | C08-01 | physiology | open (first names in this chapter supplied from general knowledge, see row C08-02) |
| Median testosterone fall 26 per cent waking, 34 per cent evening, or 20.5 per cent by one account, sample given as 394, 465, 624 or 683, sampled 2005 and 4.5 years later | Fatherhood and testosterone, Cebu | Gettler, McDade, Feranil and Kuzawa 2011, PNAS 108:16194 to 16199 | BI-02, BI-06, DH-08, chapter 3 rows C03-25 and C03-27 | C08-03 | physiology | open on sample and size, robust in direction |
| Men with higher baseline testosterone more likely to be partnered fathers by follow-up, the fall came after | Selection into fatherhood | Gettler et al. 2011 | DH-08, FA-02 | C08-04 | behaviour | robust |
| Fathers doing three hours or more of childcare a day had the lowest testosterone at follow-up | Dose by hours of care | Gettler et al. 2011 | DH-08, BI-06 | C08-05 | physiology | robust |
| Mating effort against parenting effort as the mechanism | Life-history framing | Gettler et al. 2011 discussion, no separate primary | DH-08 | C08-06 | physiology | robust as framing |
| BaYaka fathers rated more reliable providers and sharers had lower testosterone and higher oxytocin | Replication in a forager population, Congo Basin | Gettler, Boyette et al. 2020, American Journal of Human Biology | DH-08 | C08-07 | physiology | robust |
| Fathers' prolactin and oxytocin rise to infant cues | Paternal peptide response | Storey, Walsh, Quinton and Wynne-Edwards 2000, Evolution and Human Behavior | FA-02 | C08-08 | physiology | robust |
| 43 fathers over the first six months, oxytocin tracks affect synchrony, prolactin tracks coordinated exploratory play | What each peptide goes with | Gordon, Zagoory-Sharon, Leckman and Feldman 2010, journal not named in the document | DH-08 | C08-09 | physiology | robust |
| No baseline oxytocin difference between new mothers and new fathers, several cohorts | Parental oxytocin by sex | Feldman laboratory studies, no single primary named | FA-01 | C08-10 | physiology | robust, source open |
| 89 first-time parents, fMRI plus oxytocin, three groups | Design of the parental-brain study | Abraham et al. 2014, PNAS 111:9792 to 9797, place of study not given | FA-01 | C08-11 | physiology | open on place |
| Mothers' amygdala-centred network dominant and correlated with oxytocin and synchrony, secondary-caregiving fathers' cortical network dominant, primary-caregiving gay fathers both with amygdala-to-cortex connectivity, connectivity linear with childcare time across all fathers | Parental brain by role | Abraham et al. 2014 | FA-01 | C08-12 | physiology | robust |
| Oxytocin rises with affectionate touch and motherese in mothers, with stimulatory play in fathers | Context of the oxytocin rise | Feldman, Gordon, Schneiderman, Weisman and Zagoory-Sharon 2010, not named in the document beyond the laboratory | FA-01 | C08-13 | physiology | robust |
| 25 first-time mothers scanned pre-conception and postpartum, against 20 childless women, 19 first-time fathers, 17 childless men | Design of the pregnancy-brain study | Hoekzema et al. 2016, Nature Neuroscience, online December 2016, print 2017 | FA-02, BI-06 | C08-14 | physiology | robust, year contested between documents |
| Grey-matter reductions in hippocampus, amygdala, orbitofrontal and subgenual prefrontal cortex, right superior temporal gyrus, insula, fusiform gyrus, symmetrical | Where pregnancy changes the brain | Hoekzema et al. 2016 | FA-02 | C08-15 | physiology | robust |
| Classifier sorted every mother from every non-mother, 100 per cent, change six times cortical and three times medial temporal test-retest variability | Size and consistency of the change | Hoekzema et al. 2016 | FA-02 | C08-16 | physiology | robust |
| Oestradiol 3.42 to 12,400 pg/ml, progesterone 0.84 to 103 ng/ml | Pregnancy hormone ranges | Attributed by FA-02 to Hoekzema et al. 2016, check they are not from a source cited within it | FA-02 | C08-17 | physiology | open on attribution |
| Reduced regions overlap with regions activated by own-baby pictures, magnitude of change predicts postpartum attachment | Functional meaning of the change | Hoekzema et al. 2016 | FA-02 | C08-18 | physiology | robust |
| Same cohort at six years, 91.67 per cent classification, changes still track attachment | Persistence of the change | Martínez-García, Paternina-Die, Desco, Vilarroya, Hoekzema and Carmona 2021, Brain Sciences | FA-02 | C08-19 | physiology | robust |
| Second pregnancy: first pregnancy's reductions cover a 79 per cent larger area, median change 3.1 per cent, d above 1.0 | Second pregnancy versus first | Hoekzema et al. 2022, journal not named in the document | FA-02 | C08-20 | physiology | robust |
| Change described as pruning or specialisation, not loss of function | Interpretation | Hoekzema et al. 2016 discussion | FA-02, BI-06 | C08-21 | physiology | robust |
| 19 mothers, grey-matter increases in prefrontal, parietal and midbrain between two to four weeks and three to four months postpartum | Postpartum increase | Kim, Leckman, Mayes, Feldman, Wang and Swain 2010, Behavioral Neuroscience 124:695 to 700, exact windows to check | FA-02 | C08-22 | physiology | contested with Hoekzema on direction |
| Grey-matter increases from one to two days to four to six weeks postpartum | Early postpartum increase | Luders, Gingnell, Poromaa, Engman, Kurth and Gaser 2020, journal not named | FA-02 | C08-23 | physiology | contested with Hoekzema on direction |
| Biphasic model: gestational reduction by placental steroids then postpartum increase by infant stimuli, rodent transient swelling of emotion and reward regions | The two-tide reconciliation | Barrière et al. 2021, journal not named | FA-02 | C08-24 | physiology | contested (rests on rodent data and a review) |
| Cortical volume reduction and thinning, largest in precuneus, d above 0.5, larger reductions with stronger own-baby response, stronger prenatal bonding, more planned leave, more interaction | Fathers' brain change | Saxbe et al. 2024, Cerebral Cortex, N 38 | FA-02 | C08-25 | physiology | robust |
| Larger reductions with more sleep problems, anxiety, distress and depression at six months, controlling for prenatal levels | The cost of the fathers' change | Saxbe et al. 2024 | FA-02 | C08-26 | physiology | robust |
| UK Biobank, more than 30,000 participants, PNAS: more children parented, higher functional connectivity in mid and later life, both sexes, inverse of the ageing pattern | Parenthood and brain ageing | Author and year not named in FA-02, probably Orchard et al. 2025, PNAS | FA-02 | C08-27 | physiology | open |
| 43 studies, 28,004 participants, paternal perinatal depression 10.4 per cent, confidence interval 8.5 to 12.7 | Prevalence of paternal depression | Paulson and Bazemore 2010, JAMA 303:1961 to 1969, the 28,004 may include mothers | FA-02, FA-01 | C08-28 | behaviour | robust, N to check |
| 25.6 per cent at three to six months, interval 17.3 to 36.1, interview 4.9 per cent against self-report 11.0, US 14.1 against international 8.2 | Where and how the figure moves | Paulson and Bazemore 2010 | FA-02 | C08-29 | behaviour | robust, wide interval |
| Paternal and maternal depression correlation r 0.308, interval 0.228 to 0.384 | Co-occurrence in the couple | Paulson and Bazemore 2010 | FA-02, FA-01 (recovered from the PDF, p. 8) | C08-30 | behaviour | robust |
| Depressed fathers: long working hours odds ratio 5.34, isolation 5.28, father's risk up to 50 per cent higher if the mother is depressed | Presentation of paternal depression | Unnamed systematic review cited in FA-02 | FA-02 | C08-31 | behaviour | open |
| Edinburgh Postnatal Depression Scale validated on mothers, under-detects men | Screening bias | No primary named | FA-02 | C08-32 | behaviour | open |
| ALSPAC, more than 10,000 children, paternal depression at eight weeks, adjusted odds ratio 2.09 for emotional and behavioural problems at 3.5 years, 2.66 for conduct problems in boys, independent of maternal and later paternal depression | Effect of paternal depression on the child | Ramchandani, Stein, Evans, O'Connor and the ALSPAC team 2005, The Lancet 365:2201 to 2205 | FA-01 | C08-33 | behaviour | robust |
| Maternal postpartum depression 10 to 13 per cent in developed countries, up to 20 per cent in developing, or 10 to 15 per cent | Prevalence of maternal depression | No single primary named | FA-02, DH-08 | C08-34 | behaviour | robust as a range, documents differ |
| Postpartum depression as bargaining or costly signal of low support or low infant viability | Evolutionary reading one | Hagen 1999, Evolution and Human Behavior 20:325 to 359 | DH-08 | C08-35 | behaviour | contested (against Hahn-Holbrook) |
| Perceived lack of family support and low Apgar score the strongest predictors of onset, longitudinal | Support for the bargaining reading | Kardum et al., year and journal not given | DH-08 | C08-36 | behaviour | open |
| Postpartum depression as evolutionary mismatch, lost alloparental support | Evolutionary reading two | Hahn-Holbrook and Haselton 2018, year and journal to confirm | DH-08 | C08-37 | behaviour | contested (against Hagen) |
| Zuo yuezi, one month, China, and la cuarentena, forty days, Latin America | Postpartum confinement customs | No primary named | FA-02 | C08-38 | culture | open |
| German Wochenbett: statutory midwife home visits up to twelve weeks after birth | Public postpartum care in Germany | FA-02 sources an insurance blog and a thesis, check SGB V paragraph 24d and the Hebammenhilfe-Vertrag | FA-02 | C08-39 | culture | open |
| Parents against non-parents in marital satisfaction d minus 0.19, 38 per cent of mothers of infants against 62 per cent of childless women reporting high satisfaction | Birth and the couple | Twenge, Campbell and Foster 2003, Journal of Marriage and Family 65:574 to 583 | FA-02 | C08-41 | behaviour | robust |
| 218 couples over eight years, sudden drop at the birth | Shape of the decline | Doss, Rhoades, Stanley and Markman 2009, JPSP 96:601 to 619 | FA-02 | C08-42 | behaviour | contested (against Mitnick) |
| Non-parents also decline over the same period | Whether the child causes or advances the decline | Mitnick, Heyman and Smith Slep 2009, Journal of Family Psychology 23:848 to 852 | FA-02 | C08-43 | behaviour | contested (against Doss) |
| 95 studies, more than 3,000 children, infants slightly more secure with mothers, g minus 0.07, mother-father attachment r 0.32 | Attachment to the two parents | Lucassen, van IJzendoorn et al. 2011, Journal of Family Psychology 25:986 to 992 | FA-01 (recovered from the PDF, p. 3) | C08-44 | behaviour | robust |
| 174 studies, 22,914 dyads, maternal sensitivity r 0.26 (0.22 to 0.29), paternal r 0.21 (0.14 to 0.27) | Antecedents of attachment by parent | Document names van IJzendoorn, probable primary Madigan et al. 2024, Psychological Bulletin, or Deneault et al. | DH-08 (recovered from the PDF, p. 8) | C08-45 | behaviour | open on attribution |
| Integrative hypothesis: multiple secure attachments as insurance | Reading of the attachment evidence | van IJzendoorn, Sagi and Lambermon 1992, not named in the document | DH-08 | C08-46 | behaviour | robust as framing |
| Disorganised attachment d 0.77 higher in high-risk households | Risk and attachment | No primary named, probable Cyr et al. 2010 | DH-08 (recovered from the PDF, p. 9) | C08-47 | behaviour | open |
| Risky Situation procedure, 12 to 18 months, intrusive stranger, stairs, forbidden toy, under- and over-activated profiles with later anxiety and impulsivity | Activation parenting | Paquette and Bigras 2010, not named beyond Paquette | FA-01 | C08-48 | behaviour | robust |
| 40 studies, 3,607 fathers, 3,707 mothers, fathers higher on activation parenting g 0.27, p below .001 | Size of the activation difference | Sourced to a ResearchGate table, no author | FA-01 (recovered from the PDF, pp. 1 and 4) | C08-49 | behaviour | open on source |
| Overlap about 89 per cent for g 0.27 | Shared area of the two bells | Narrator's arithmetic from the chapter 4 formula, not in the documents | none | C08-50 | behaviour | robust as arithmetic |
| Rough-and-tumble play peaks around age four, more with sons, linked to emotion regulation and less aggression | Play by parent | No primary named, probable Fletcher, StGeorge and Freeman 2013 or Paquette 2003 | FA-01 | C08-51 | behaviour | open on source |
| Fathers' evolutionary role as the challenging parent | Theory of the activation difference | Bögels and Phares 2008, Clinical Psychology Review 28:539 to 558 | FA-01 | C08-52 | behaviour | contested (against Pleck) |
| 28 studies, 5,728 mother-child pairs, 12 studies, 1,019 father-child pairs, paternal not maternal challenging parenting linked to less child anxiety | Challenging parenting and anxiety | Author not named, probable Möller et al. 2016 | FA-01 (recovered from the PDF, p. 5) | C08-53 | behaviour | open on source |
| Rural dual-earner families, children at 24 months, language at 36 months, fathers fewer words, more word roots, more wh-questions, more clarification requests, paternal vocabulary diversity a unique predictor after controls | Fathers' language input | Pancsofar and Vernon-Feagans 2006, Child Development per the document, some versions in Journal of Applied Developmental Psychology | FA-01 | C08-54 | behaviour | robust, venue to check |
| 41 studies, 5,563 children, moderate correlation of father language with child language | Pooled father-language effect | Meta-analysis not named, probable Rowe et al. or Cabrera et al. | FA-01 (N recovered from the PDF, p. 6) | C08-55 | behaviour | open on source |
| Bridge Hypothesis, father as the less accommodating listener | Theory of the language difference | Gleason 1975, and Pancsofar and Vernon-Feagans | FA-01 | C08-56 | behaviour | contested (against Pleck) |
| Play and language differences as artefacts of less time | The objection | Pleck, year and work not named in the document | FA-01 | C08-57 | behaviour | contested, never tested in primary-carer fathers |
| Engagement, accessibility, responsibility, with responsibility barely shifted | Frame for fatherhood | Lamb, Pleck, Charnov and Levine 1985, American Zoologist 25:883 to 894 | FA-01 | C08-58 | culture | robust |
| Causal designs show smaller but consistent father-absence effects on high-school completion, social-emotional adjustment and adult mental health | Father absence | McLanahan, Tach and Schneider 2013, Annual Review of Sociology 39:399 to 427 | FA-01 | C08-59 | behaviour | robust |
| Cohort of 879, father loss and shorter telomeres | Biological trace of father loss | Mitchell, McLanahan, Schneper et al. 2017, Pediatrics 140:e20163245, document also says PNAS | FA-01 | C08-60 | physiology | open on venue |
| 54 effect sizes, 1,509 participants, children of gay fathers adjusted within or above the normal range | Control group for the father-absence argument | FA-01 text says Fedewa et al. 2015, reference list says Carone et al. 2024, Sexuality Research and Social Policy, mismatch confirmed in the PDF | FA-01 | C08-61 | behaviour | open on attribution, figure stands |
| Eight histological studies, 325 ovaries, age explains 81 per cent of variance, peak about 300,000 follicles per ovary at 18 to 22 weeks gestation, prediction interval 35,000 to 2,534,000, 12 per cent of reserve at 30, 3 per cent at 40 | Ovarian reserve | Wallace and Kelsey 2010, PLoS ONE 5:e8772 | FA-05 (81 per cent recovered from the PDF, p. 1) | C08-62 | physiology | robust |
| AMH and FSH poor predictors of fecundability in women without an infertility history | Reserve markers and the individual | Steiner et al. 2017, JAMA 318:1367 to 1376 | FA-05 | C08-63 | physiology | robust |
| 90 per cent chance of one, two or three children means starting by 32, 27 or 23 without IVF, 35, 31 or 28 with | Fertility timetable | Habbema, Eijkemans, Leridon and te Velde 2015, Human Reproduction 30:2215 to 2221 | FA-05 | C08-64 | physiology | robust |
| Students overestimate IVF's power to rescue fertility in the forties | Beliefs about IVF | Survey not named | FA-05 | C08-65 | culture | open |
| Male fertility declines from 35, markedly after 50 | Male age and conception | Dunson, Baird and Colombo 2004, Human Reproduction 19:1548 to 1553 | FA-05 | C08-66 | physiology | robust |
| Paternal age the main driver of de novo mutations | Male age and mutation | Kong et al. 2012, Nature 488:471 to 475, 78 trios. The document's "over 1.4 million individuals across four populations" is not this paper's sample | FA-05 | C08-67 | physiology | robust in finding, open on the document's sample description |
| 2,615,081 Swedish births, fathers 45 and over against 20 to 24: hazard ratio 3.45 autism (1.62 to 7.33), 13.13 ADHD, 24.70 bipolar | Paternal age and the child's psychiatric risk | FA-05 attributes all of it to Reichenberg et al. 2006, Archives of General Psychiatry 63:1026 to 1032, which was an Israeli cohort, autism only, odds ratio 5.75 for 40 and over. The Swedish 2.6 million cohort with the ADHD and bipolar hazards is D'Onofrio et al. 2014, JAMA Psychiatry. The document may have merged two papers | FA-05 | C08-68 | physiology | open on attribution. The chapter states the finding in a clause, gives no hazard ratio, and notes once that these are multiples of rare things |
| 185 studies, 42,935 men, 52.4 per cent decline in sperm concentration 1973 to 2011 | Sperm-count decline | Levine et al. 2017, Human Reproduction Update 23:646 to 659. Chapter 3 carries Levine 2022 (223 studies, 51.6 per cent, to 2018), a different paper | FA-05, BI-10 | C08-69 | physiology | contested (Boulicault and the GenderSci Lab glassware objection, chapter 3) |
| Long-run child penalty about 20 per cent Denmark, 60 per cent Germany, or 61 per cent West Germany | Earnings penalty of a first child | FA-05 cites Kleven, Landais and Søgaard 2019, AEJ Applied (the Denmark paper). HC-05 cites Kleven, Landais, Posch, Steinhauer and Zweimüller 2019, AEA Papers and Proceedings (the six-country comparison, where the German figure comes from) | FA-05, HC-05 | C08-70 | culture | robust in size, author list differs between documents |
| Germany total fertility rate about 1.4, Europe 1.4 to 1.5, ideal family size 2.0 to 2.2, fertility gap 0.35 | Fertility against stated ideals | Sobotka and Lutz 2010, Comparative Population Studies 35 | FA-05 | C08-71 | culture | robust |
| 68 per cent of the German fertility decline from fewer third and higher-order births | Decomposition of the decline | Bujard and Sulak 2016, KZfSS 68:487 to 514 | FA-05 | C08-72 | culture | robust |
| Fertility falls where public equity meets private inequity | Gender equity theory | McDonald 2000, Population and Development Review 26:427 to 439 | FA-05 | C08-73 | culture | robust as theory |
| South Korea total fertility rate 0.72 in 2023, about 280 trillion won of pronatal spending since 2006 | The demonstration case | FA-05 sources a news aggregator and an MDPI paper | FA-05 | C08-74 | culture | open on source |
| Family policies add 0.2 to 0.3 children per woman across the OECD | What policy buys | Luci-Greulich and Thévenon 2013, European Journal of Population 29:387 to 416 | FA-05 | C08-75 | culture | robust |
| Father presence improves child survival in about one third of 45 populations, Barí 822 pregnancies, 80 per cent against 64 per cent survival with secondary fathers | Ancestral value of fathers | Sear and Mace 2008, Evolution and Human Behavior 29:1 to 18. Beckerman et al. 1998, Current Anthropology 39:164 to 167 | DH-08, chapter 2 rows | C08-76 | behaviour | robust |
| Single-caregiver model a Western construct, money not fully separated from father presence | Critics of the father-absence literature | No specific primary for the money point | FA-01 | C08-77 | culture | open |
| 114 effects, 66 studies, direction of the fatherhood testosterone decline confirmed | Pooled fatherhood effect | Grebe et al. 2019, meta-analysis | chapter 3 row C03-28 | C08-78 | physiology | robust |
The case for and the case against
The case the chapter makes
- The parental brain follows the caregiving role, not the sex: Abraham, Feldman et al. 2014, 89 parents, connectivity linear with hours (FA-01, C08-11, C08-12).
- Fathers' testosterone falls with hours of care, not with the status: Gettler 2011 Cebu, Gettler and Boyette 2020 BaYaka (DH-08, BI-06, C08-03 to C08-07).
- Pregnancy reshapes the mother's social-cognition regions, lasting and predictive of attachment: Hoekzema 2016, Martínez-García 2021 (FA-02, C08-14 to C08-19).
- Fathers change in the same direction, scaled to engagement, with a mental-health cost: Saxbe 2024 (FA-02, C08-25, C08-26).
- Paternal depression at 10.4 per cent, correlated 0.308 with maternal, independently predictive for the child: Paulson and Bazemore 2010, Ramchandani 2005 (FA-02, FA-01, C08-28 to C08-33).
- Attachment nearly equal across parents: Lucassen 2011, the 174-study sensitivity meta-analysis (FA-01, DH-08, C08-44, C08-45).
- The two clocks: Wallace and Kelsey 2010, Habbema 2015, the paternal-age finding attributed to Reichenberg 2006 (FA-05, C08-62, C08-64, C08-68).
- The earnings penalty falls on the mother: Kleven et al. 2019, 60 per cent Germany against 20 per cent Denmark (FA-05, HC-05, C08-70).
The case against
- The chapter's role difference is not a sex difference, and the chapter's own best study shows it: Abraham 2014's gay fathers, Saxbe 2024's hours-scaled change, no baseline oxytocin difference (FA-01, FA-02, C08-10, C08-12, C08-25).
- Attachment to fathers differs from attachment to mothers by g minus 0.07, and paternal sensitivity predicts security at 0.21 against the mother's 0.26 with overlapping intervals (FA-01, DH-08, C08-44, C08-45).
- The activation difference, g 0.27, leaves about 89 per cent of the two bells shared, and Pleck's time-artefact reading has never been tested in primary-carer fathers (FA-01, C08-49, C08-50, C08-57).
- The Cebu testosterone finding is observational, with sample and drop reported four ways, and no one has assigned fathers to care at random (the BI digest, chapter 3, C08-03).
- The mother's brain change is contested in direction: Kim 2010 and Luders 2020 against Hoekzema 2016, with Barrière 2021's two-tide model resting on rodent data and a review (FA-02, C08-22 to C08-24).
- Paternal depression was counted by questionnaire more than interview, 11.0 against 4.9 per cent, and in American samples more than elsewhere, 14.1 against 8.2 (FA-02, C08-29).
- The ancestral record: Sear and Mace 2008 found father presence improving child survival in about a third of 45 populations, and the Barí children with a second father survived at 80 against 64 per cent. Fathers were useful, substitutable and optional (DH-08, C08-76).
- No fixed-effects design removes every confound from the father-absence literature, and the money a father brings has not been fully separated from the man (FA-01, C08-77).
Where the sources disagree
- Direction of maternal brain change: Hoekzema 2016 against Kim 2010 and Luders 2020, with Barrière 2021's two-phase synthesis (FA-02, C08-22 to C08-24).
- Whether the activation and language differences are sex or time: Paquette and Pancsofar against Pleck (FA-01, C08-48 to C08-57).
- Whether the birth causes the couple's decline or advances it: Doss 2009 against Mitnick 2009 (FA-02, C08-42, C08-43).
- Gettler 2011 sample: 394 and 465 (BI-02), 683 (BI-06), 624 (DH-08). Drop: 26 per cent waking and 34 per cent evening against 20.5 per cent. Carried from chapter 3 (C08-03). The narrated chapter rounds across the disagreement to "about a quarter to a third".
- Hoekzema's year: 2016 (FA-02, the CO and FA digest) against 2017 (BI-06, the BI digest). Online December 2016, print 2017 (C08-14).
- Kleven's German child penalty: 60 per cent (FA-05, citing Kleven, Landais and Søgaard 2019, AEJ Applied, which is the Denmark paper) against 61 per cent (HC-05, citing Kleven, Landais, Posch, Steinhauer and Zweimüller 2019). The narrated chapter gives "about 60 per cent" (C08-70).
- Maternal postpartum depression: 10 to 13 per cent developed, up to 20 developing (FA-02) against 10 to 15 (DH-08). The narrated chapter says "about as common" as the paternal figure (C08-34).
- Sperm decline: 52.4 per cent, 185 studies, 42,935 men, 1973 to 2011, Levine 2017 (FA-05) against 51.6 per cent, 223 studies, to 2018, Levine 2022 (BI-10, chapter 3). Two different papers. Contested by Boulicault and the GenderSci Lab (C08-69).
- Mitchell 2017 venue: Pediatrics (FA-01 source table) against "Pediatrics and PNAS" (FA-01 text) (C08-60).
- Fedewa 2015 (FA-01 text) against Carone 2024 (FA-01 reference 57) for the gay-father meta-analysis. Confirmed in the PDF (C08-61).
- Postpartum depression as bargaining (Hagen 1999) or mismatch (Hahn-Holbrook 2018) (DH-08, C08-35, C08-37).
- Paulson's 28,004 participants: may include mothers, the data edition's text said fathers (C08-28).
- Kim 2010 scan windows: the data edition said "second and fourteenth week", the document says two to four weeks and three to four months (C08-22).
- Dunson 2004: the data edition said conception odds fall "around 50", the document says "severe after 50" (C08-66).
Open questions
Every mark for checking from the data edition, with its row:
- First names supplied from general knowledge for every person the documents give by surname only: Saxbe, Abraham, Gordon, Boyette, Hoekzema, Martínez-García, Kim, Luders, Paulson, Bazemore, Ramchandani, Hahn-Holbrook, Twenge, Campbell, Foster, Doss, Mitnick, Heyman, Smith Slep, Lucassen, Bögels, Phares, Pancsofar, Vernon-Feagans, Lamb, Charnov, Levine, McLanahan, Tach, Schneider, Mitchell, Wallace, Kelsey, Steiner, Habbema, Dunson, Reichenberg, Levine, Kleven, Sobotka, Lutz, Bujard, Sulak, McDonald. Given in the documents: Gettler, Feldman, Paquette, Pleck, van IJzendoorn, Hagen, Sear, Mace, Bowlby. Storey, Barrière, Kardum and Kong left as surnames. Row C08-02. The narrated chapter names Darby Saxbe, Lee Gettler, Eyal Abraham, Ruth Feldman, Elseline Hoekzema, James Paulson, Sharnail Bazemore and Joseph Pleck.
- Abraham and Feldman 2014: place of the study not given. Row C08-11.
- UK Biobank parenthood-and-ageing PNAS paper: author not named in FA-02. Row C08-27.
- Paternal depression odds ratios 5.34 and 5.28: systematic review not named in FA-02. Row C08-31.
- Wochenbett details: FA-02's sources are an insurance blog and a thesis. Check SGB V paragraph 24d and the Hebammenhilfe-Vertrag. Row C08-39.
- Disorganised attachment d 0.77 in high-risk households: no primary named in DH-08, probable Cyr et al. 2010. Row C08-47.
- Activation parenting meta-analysis, 40 studies, g 0.27: sourced to a ResearchGate table, no author. Row C08-49.
- Anxiety meta-analysis, k 28 and k 12: author not named in FA-01, probable Möller et al. 2016. Row C08-53.
- Language meta-analysis, 41 studies, N 5,563: not named in FA-01. Row C08-55.
- Mitchell 2017 journal: FA-01 says Pediatrics and PNAS. Row C08-60.
- Fedewa 2015 against Carone 2024 for the 54-effect-size, 1,509-participant meta-analysis: confirmed in the PDF, text p. 8, reference 57 p. 16. Row C08-61.
- Kong 2012 sample, "over 1.4 million individuals across four populations": implausible for a de novo mutation study of 78 trios. Row C08-67.
- South Korea 280 trillion won: FA-05's source is a news aggregator and an MDPI paper. Row C08-74.
- Reichenberg 2006 attribution: FA-05 gives 2,615,081 Swedish births with hazard ratios for autism, ADHD and bipolar under one 2006 citation. Reichenberg 2006 was an Israeli conscript cohort reporting autism only. The Swedish 2.6 million cohort with the ADHD and bipolar hazards is D'Onofrio et al. 2014, JAMA Psychiatry. If the check confirms the split, the website should name both papers. Row C08-68.
- Whether the fertility section belongs in this chapter or in chapter 7 or 13. It is the one section the governing metaphor does not naturally reach. The narrated chapter keeps it, shortened, because it carries the child-penalty figure that the Preface and chapter 13 refer to.
- Whether to keep the couple-satisfaction paragraph at all, since chapter 6 owns the couple. The narrated chapter keeps it as three sentences without figures.
- Whether the German Wochenbett deserves a real scene with a primary source (a midwife, a date, the statutory basis), which FA-02 cannot supply.
- Whether to name the narrator's age at writing in the fertility section. Not done.
Points of the narrated chapter that are inference rather than figure, carried over from the data edition's assumptions:
- The opening scene is a composite of Saxbe 2024's design (pre-birth and six-month scans, own-baby against unknown-infant pictures). The mirror, the noise and the sequence are narrative dressing. The city is unnamed because FA-02 does not give one.
- "A multiple of a rare thing is still a rare thing" for the paternal-age hazards is an inference from the design, not a figure in FA-05. No absolute risk is given in the document.
- "Six years on, the same women could still be told from the controls" treats Martínez-García 2021 as a follow-up of the Hoekzema cohort, as FA-02's table says.
- "The care penalty falls on the father who is kept out" is the chapter's reading of Abraham 2014's hours line, Gettler 2011's hours effect and Saxbe 2024's leave effect together, not a figure from any one study.
- The In practice paragraph on where the baby sleeps is practice, not a finding. The data edition carried no study on co-sleeping. FA-02 lists breastsleeping in its contents.
The documents
FA-01: mothers and fathers as distinct contributions: Abraham and Feldman, Lucassen, Paquette, Bögels and Phares, Pancsofar and Vernon-Feagans, Lamb, McLanahan, Mitchell, Paulson, Ramchandani, Fedewa/Carone. FA-02: the postpartum period and the couple: Hoekzema, Martínez-García, Kim, Luders, Saxbe, Strathearn, Storey, Twenge, Doss, Mitnick, Paulson, the Wochenbett, zuo yuezi, breastsleeping. FA-05: fertility and the modern life course: Wallace and Kelsey, Steiner, Habbema, Dunson, Kong, Reichenberg, Levine, Kleven, Sobotka and Lutz, Bujard and Sulak, McDonald, Luci-Greulich and Thévenon, South Korea. BI-06: the endocrine life course: Hoekzema (dated 2017), Gettler with the three-hour figure and N 683, the oxytocin and prolactin rise in fathers. DH-08: ancestral child-rearing: Gettler 2011 with N 624, Gettler and Boyette 2020, Gordon and Feldman 2010, the 174-study sensitivity meta-analysis, the integrative hypothesis, Hagen, Hahn-Holbrook, Sear and Mace, the Barí.