Brain-imaging studies of first-time mothers find that pregnancy reduces grey-matter volume across most of the cortex, in changes that partly persist two years after birth and that tracked attachment to the infant in one cohort. Measured memory dips in late pregnancy, and prospective studies find no lasting decline. The evidence describes an adaptation of measurable size; it does not show a general cognitive improvement, and no study has yet identified the cellular mechanism.
On 6 October 2026 MIT Press published A Mother’s Brain: The New Science of the Neuro-Maternal Revolution, the English edition of a book by the Madrid neuroscientist Susana Carmona, credited with Lily Meyer.1 The book first appeared in Spanish as Neuromaternal, published by Ediciones B in 2024.2 Its publisher presents it as a challenge to “mommy brain” stereotypes. In an interview published on 4 November 2026, Carmona said that memory deficits are limited to the third trimester and the first postpartum years, and that cognitive function improves over the long term.3
Pregnancy produces a structural change in the adult brain larger than the variation measured in comparable non-pregnant adults: grey-matter volume falls across most of the cortex, part of the change persists for at least two years, and in one cohort it tracked the mother’s attachment to her infant. The measured cognitive cost is a small, temporary dip in memory in late pregnancy. The evidence supports adaptation; it does not support a general cognitive upgrade, and it does not establish synaptic pruning as the mechanism.
Three tests can falsify this reading on published data.
- Size and persistence. Longitudinal studies with a scan before conception must show grey-matter change beyond the variation seen in controls, lasting beyond the first postpartum months. If it does not, the thesis fails.
- Mechanism. Pruning must explain the data better than two rivals: fluid shifts during gestation, and the behavioural changes of parenthood that fathers share. If the rivals fit equally well, the mechanism claim fails.
- Cognition and scope. Objective tests against a pre-pregnancy baseline must show no lasting decline. An “upgrade” requires measured improvement. The claim holds only for the populations actually scanned.
The first test passes on small samples. The second fails for pruning in particular, because none of the imaging studies reviewed measured synapses and the authors of the most detailed study raise fluid as a candidate explanation themselves. The third passes for the absence of lasting cost and fails for the word upgrade, which is why the working title of this Dispatch was changed.
What the scanners measured
The modern evidence begins with a study by Elseline Hoekzema and colleagues, Carmona among them, published in the February 2017 issue of Nature Neuroscience.4 The team scanned 25 women before their first pregnancy and again after birth, at a mean of about 74 days postpartum. It compared them with 20 women who did not become pregnant over a similar interval, with 19 first-time fathers and with 17 childless men. The interval between scans averaged 463 days for the mothers and 413 days for the control women.
The mothers showed reductions in grey-matter volume concentrated in regions associated with social cognition. The pattern was distinct enough that a classification algorithm, trained only on the change in grey matter, separated the women who had been pregnant from those who had not with a balanced accuracy of 100% in leave-one-out testing. Analyses on predefined anatomical regions and on cortical surfaces gave accuracies between 69% and 96%, depending on the measure.
Three further results gave the study its interpretation. When 20 of the mothers were later shown pictures of their own infant and of other infants, nearly 30% of the voxels that responded more strongly to their own child lay within the grey matter that had lost volume, about seven times the overlap expected by chance. A multivariate model predicted two dimensions of a postpartum attachment questionnaire, quality of attachment and absence of hostility, from the grey-matter changes. And in the 11 mothers rescanned about two years after birth, the reductions persisted in every cluster except part of the left hippocampus, which partly recovered.
| Study | Design | Mothers | Scan before conception | Comparison group | Principal finding |
|---|---|---|---|---|---|
| Hoekzema et al., 2017 | Before and after pregnancy | 25 | Yes | 20 women, 19 fathers, 17 men | Grey-matter reductions in social-cognition regions; persisted at about two years in the 11 rescanned |
| Paternina-Die et al., 2024 | Late pregnancy and early postpartum | 110 | No | 34 nulliparous women | Lower cortical volume in all functional networks in late pregnancy, attenuated after birth; replicated in 29 mothers |
| Pritschet et al., 2024 | 26 scans of one woman | 1 | Yes | Control participants scanned over a similar span | Grey-matter decline in 80% of 400 cortical regions across gestation; partial rebound |
| Total, computed by Titanica | 136 | 26 | Excludes the replication sample of 29 |
The second large study came from the Neuromaternal laboratory that Carmona founded in 2017 at the Instituto de Investigación Biomédica of the Hospital Gregorio Marañón in Madrid,3 and was published in Nature Neuroscience in February 2024.5 María Paternina-Die and colleagues scanned 110 first-time mothers in late pregnancy and again in the early postpartum period, alongside 34 women who had never given birth. In late pregnancy the mothers had lower cortical volume than the controls across every functional network examined. The differences attenuated in the early postpartum session, more slowly in the default mode and frontoparietal networks, and mothers who had a scheduled caesarean section showed different trajectories. The main findings were replicated in an independent sample of 29 mothers and 24 controls.
The design trades one strength for another. It has more than four times as many mothers as the 2017 study, and it measures them near the point of greatest change, but it has no scan before conception, so differences between mothers and controls could in principle predate pregnancy. Read together, the two studies describe a reduction that is largest near birth, that partly reverses in the months that follow, and that leaves a measurable residue for at least two years.
| Measure | Reported value | Computed by Titanica |
|---|---|---|
| Classification of mothers versus controls from grey-matter change | 100% balanced accuracy, leave-one-out | |
| Responsive voxels (own baby over other baby) inside regions that lost volume | Nearly 30%, about 7 times chance | |
| Attachment: quality of attachment, residualised model | R = 0.38 (N = 24) | R² = 0.14, about 14% of variance |
| Attachment: absence of hostility, residualised model | R = 0.42 (N = 24) | R² = 0.18, about 18% of variance |
| Attachment: pleasure in interaction | Not significant (p = 0.985) | |
| Mothers rescanned about two years after birth | 11 of 25 | 44% of the cohort |
| Clusters with reductions persisting at two years | All except part of the left hippocampus | |
| Word-list learning, backward digit span, 2-back task | No significant change against controls |
The table also shows what the 2017 study did not find. The mothers’ scores on word-list learning, backward digit span and a working-memory task did not change significantly relative to controls, with only a trend toward fewer correct answers on word-list learning, and that trend did not track the change in grey matter. The structural change was clear and the cognitive change on these tasks was not detectable in a sample of this size.
One pregnancy, scanned twenty-six times
The third study traded sample size for density. A team led by Laura Pritschet and Emily Jacobs at the University of California, Santa Barbara, and Elizabeth Chrastil at the University of California, Irvine, scanned one healthy 38-year-old first-time mother, who had conceived by in vitro fertilisation, 26 times between three weeks before conception and two years after birth.6 The study appeared in Nature Neuroscience on 16 September 2024 and was funded in part by the National Institute on Aging and the National Institute of Mental Health.7
Total grey-matter volume declined almost linearly across gestation and rebounded only in part after birth. Eighty per cent of the 400 cortical regions analysed showed a negative relationship between volume and week of gestation. The authors report that the change in cortical volume was nearly three times larger than that seen in control participants scanned over a similar span, and about six times larger than the participant’s own variation before conception. Cortical thickness followed the same pattern, and its global reduction was still present two years after birth.
Two other measures moved the opposite way. Quantitative anisotropy, an index of white-matter microstructure, rose through the first and second trimesters and returned to baseline by the first postpartum scan. The volume of the lateral ventricles and of cerebrospinal fluid rose in the second and third trimesters and dropped sharply after birth. The hormone measurements show why gestation is a distinctive case for the brain: the participant’s oestradiol and progesterone rose by orders of magnitude, and the grey-matter reductions correlated significantly with both.
| Measure | Before conception | 3 weeks before birth | 3 months after birth | Change to late pregnancy, computed by Titanica |
|---|---|---|---|---|
| Oestradiol (pg/ml) | 3.42 | 12,400 | 11.50 | ×3,626 |
| Progesterone (ng/ml) | 0.84 | 103 | 0.04 | ×123 |
| Cortical regions with volume falling across gestation | 320 of 400 (80%) | |||
| Total grey-matter volume | Near-linear decline; partial rebound after birth | |||
| White-matter anisotropy | Rose in first and second trimesters; at baseline by first postpartum scan | |||
| Lateral ventricles and cerebrospinal fluid | Rose in second and third trimesters; fell sharply after birth | |||
A single participant cannot establish what is typical. The value of the study lies elsewhere: it shows the shape of the change, week by week, in a way that two-point designs cannot, and it shows that grey matter, white matter and fluid did not move together. That divergence is the key to the question of mechanism.
The pruning hypothesis and its rivals
The account popularised since 2017 holds that the loss of grey matter in pregnancy resembles the loss in adolescence, when the brain removes little-used synapses and becomes more efficient. The 2017 paper raised the comparison with adolescent pruning and described the result as a possible “adaptive refinement” or specialisation of social-cognition networks. It also stated that the reductions “cannot be pinpointed to a specific molecular mechanism”. No hormone series was available for that cohort: blood was collected from only two women during pregnancy.
Magnetic resonance imaging estimates the volume of tissue, and a fall in estimated grey matter can arise from several processes that a scanner cannot separate. The 2024 single-subject study set out the main candidates itself. Its authors suggested that the widespread pattern and the rise in cerebrospinal fluid “could reflect increased water retention and subsequent compression of cortical tissue”, noting that water retention can also alter MRI measurements. They proposed that the persistence of some changes after birth hints at cellular causes, “such as alterations in glia or neuron number, synaptic density and myelination”, and they described the decline as possible “fine-tuning” by hormones. Synaptic pruning is one item on that list, and none of the three studies measured synapses.
The time course allows a partial test. Fluid retention in pregnancy is a gestational state and should reverse after birth. In the single-subject study the ventricles and fluid did reverse sharply, white-matter anisotropy returned to baseline, and grey matter recovered only in part, with cortical thinning still measurable at two years. In the 2017 cohort, reductions persisted at two years in 11 mothers. Fluid can therefore account for part of the late-pregnancy trough and cannot account for the residue. What the residue consists of is unknown.
The second rival is parenthood itself, the change in sleep, attention, routine and social life that follows a birth whoever gave birth. Here the evidence comes from fathers. A Madrid study of 20 first-time fathers, scanned before their partner’s pregnancy and after the birth and compared with 17 childless men, found a significant reduction in cortical volume and thickness in the precuneus.8 Larger reductions were associated with stronger brain responses to pictures of the father’s own baby. The change in fathers was confined to one region. The change in gestational mothers covered most of the cortex. Carmona told Nautilus that her team finds more pronounced changes in gestational mothers than in non-gestational mothers or fathers, and that such changes may speed up caregiving without determining it.3
Population data push the comparison into middle age. In the UK Biobank, Ann-Marie de Lange and colleagues reported in 2019 that among 12,021 middle-aged women, those who had given birth showed less evidence of brain ageing than those who had not, an association they said could not be explained by common genetic variation or the confounders they measured.9 They called for prospective longitudinal studies before conclusions about cause. A year later Kaida Ning and colleagues at the University of Southern California analysed the same biobank and found the association in men as well.10
| Children | Women | Men | Men minus women, computed by Titanica |
|---|---|---|---|
| 1 | −0.21 (not significant) | −0.46 (not significant) | −0.25 |
| 2 | −0.52 | −0.62 | −0.10 |
| 3 | −0.72 | −0.68 | +0.04 |
| 4 or more | −0.69 | −0.41 (not significant) | +0.28 |
For two and three children, the associations in men and women differ by a tenth of a year or less. In the cognitive analyses of the same study, which covered 160,077 women and 143,119 men, having children was associated, in most categories, with faster reaction times and fewer visual-memory errors in both sexes, and the association was larger in men. The authors concluded that lifestyle and social factors, “rather than the physical process of pregnancy experienced only by females”, contribute to these associations. Since men do not gestate, the midlife brain-age association in mothers cannot be attributed to pregnancy on this evidence. Carmona made the same caveat in her interview: the biobank studies do not follow the same women over time, so differences that predate parenthood remain unknown.3
The mechanism test therefore divides. The short-term gestational change is specific to pregnancy in its extent, coincides with large hormonal shifts and is partly fluid. The persistent residue has no identified cellular basis. The long-term association with a younger-looking brain is shared by fathers and is cross-sectional. Pruning remains a hypothesis consistent with some of these facts and tested by none of them.
What the memory tests show
The belief that pregnancy impairs thinking is widespread, and the objective evidence on it is mixed in a specific way. The largest synthesis is a meta-analysis by Sasha Davies and colleagues, published online by the Medical Journal of Australia on 15 January 2018.11 It pooled 20 studies covering 709 pregnant and 521 non-pregnant women. General cognitive functioning and memory were significantly poorer in pregnant women, with standardised mean differences of 0.52 and 0.48. Cross-sectional differences in executive function and attention were not significant overall. In comparisons with non-pregnant women, the differences by trimester reached significance only in the third, where they were larger, though heterogeneity between studies exceeded 90%. Within the same women followed across pregnancy, general cognition and memory declined between the first and second trimesters, with standardised differences of about 0.3, and did not change significantly thereafter; the authors concluded that the differences primarily develop during the first trimester.
| Measure | Studies | Difference (SMD, 95% CI) | Share of controls outscoring the average pregnant woman, illustrative, computed by Titanica |
|---|---|---|---|
| General cognitive functioning, all | 20 | 0.52 (0.07 to 0.97) | 70% |
| Memory, all | 19 | 0.48 (0.04 to 0.92) | 68% |
| General cognition, first trimester | 4 | 0.84 (−0.47 to 2.15), not significant | |
| General cognition, third trimester | 6 | 1.28 (0.26 to 2.30) | 90% |
| Memory, third trimester | 5 | 1.47 (0.27 to 2.68) | 93% |
| Executive function, cross-sectional | 9 | −0.12 (−0.52 to 0.28), not significant | |
| Attention, cross-sectional | 8 | −0.22 (−0.70 to 0.26), not significant | |
| Prospective cohort, women aged 20–24 tested in 1999, 2003, 2007 | 1 | No significant difference in cognitive change by pregnancy or motherhood; decline in late pregnancy on one of four tests |
Two features of the meta-analysis matter for interpretation. The authors called the third-trimester decline in general cognition and memory clinically as well as statistically significant, and described the effects as small to moderate; they also reported that performance stayed within normal ranges and that the decrement would most likely show as minor lapses, such as forgetting appointments. And the trimester estimates rest on four to six studies each, with confidence intervals that run from modest to very large. The illustrative shares in the exhibit show what an effect of that size would mean if scores were normally distributed; the heterogeneity means the true average effect is poorly fixed.
The strongest evidence against lasting decline comes from a design that the imaging studies of cognition mostly lack. Helen Christensen and colleagues at the Australian National University followed 1,241 women aged 20 to 24 from 1999, testing them again in 2003 and 2007.12 Over those waves 76 were pregnant at a follow-up test, 188 became mothers and 542 remained without children. Because each woman had been tested before pregnancy, the study could compare change within the same person. It found no significant differences in cognitive change associated with pregnancy or motherhood, with a decline in late pregnancy on one of four tests. The authors suggested that earlier negative findings “may be a result of biased sampling”.
That phrase identifies the measurement problem the working thesis pointed to. Studies that recruit pregnant women and compare them with whoever is available as a control cannot separate the effect of pregnancy from differences between the groups, and studies that rely on self-report measure belief as well as performance. Reviewing the field in 2010, The Psychologist reported that about two-thirds of pregnant women say they have problems with memory or attention, and that a 2008 study found partners and people with no experience of pregnancy shared the expectation of mild decline, which its authors read as a cultural stereotype.13 The same review reported a 2007 meta-analysis of 14 studies that found a reliable but small deficit, concentrated on demanding tasks of free recall and working memory.
The record is therefore consistent on three points. A measurable decrement in memory is present by late pregnancy. It appears smaller in designs that follow the same women over time than in cross-sectional comparisons. And no prospective study reviewed here found a lasting decline. For Carmona’s further statement that cognition improves over the long term, the evidence she cites in the interview is the UK Biobank research, which is cross-sectional and finds the association in fathers too. On the evidence reviewed, “no lasting cost” is supported and “upgrade” is not.
Where the claim holds
The scope of the evidence is narrower than the language around it. The three imaging studies in Exhibit 01 enrolled first-time mothers in Spain and the United States, and the authors of the 2017 study listed the education level of their sample as a limitation. The total number of mothers with a scan before conception in those studies is 26, of whom 12 were followed for about two years. The single-subject study concerned an IVF pregnancy in a woman aged 38.
The cognitive evidence has its own limits. Laboratory tasks such as word lists and digit spans measure a narrow band of cognition. None of the studies reviewed tested the capacities that the imaging literature links to the affected regions, such as the social cognition that the 2017 study associated with them, and the one result that ties structure to behaviour, the attachment model in Exhibit 02, explains under a fifth of the variance in 24 women. Mothers who had a scheduled caesarean section followed different cortical trajectories in the 2024 Madrid cohort, a finding that indicates the circumstances of birth may also matter.
The field is also small relative to the question. Emily Jacobs, a co-author of the single-subject study, wrote in Nature in November 2023 that fewer than 0.5% of more than 50,000 human brain-imaging articles published since the 1990s consider health factors specific to women.14 The measurement bias the working thesis described therefore lies at least as much in what has been studied as in how. The tests used to look for a cognitive cost were tests of general memory, and the capacities that the changed regions are thought to serve were not measured in the studies reviewed here.
The strongest counter-argument
A specialist would object that a reduction in grey-matter volume is a measurement whose functional meaning is unknown, and that the interpretation of it as adaptation repeats the error it criticises in reverse. The same scans that are read as refinement would, in an older adult, be read as atrophy, and the only behavioural links offered are a modest attachment association in 24 women and neural responses to infant pictures, which do not by themselves show a benefit. On this view the honest conclusion is that pregnancy changes brain structure substantially and that what the change does is unknown.
The objection is largely correct and it narrows the thesis without reversing it. Three facts keep “adaptation” defensible as a description. The change is specific to gestation in its extent; unlike ageing, it partly reverses; and the regions that change overlap with those that respond preferentially to the mother’s own infant. None of these facts proves a benefit. They establish that the change is patterned and tied to the transition to caregiving, and that no lasting cognitive cost has been detected on the tests used. Adaptation in the biological sense, a change fitted to a new demand, is the most that the data allow.
A change fitted to a demand
Carmona’s book opens, according to the interview, with a line from Karl Popper: “Science must begin with myths, and with the criticism of the myths.”3 The myth in this case had two forms. The older one held that pregnancy degrades the mind. The newer one, carried in the phrase that gave this Dispatch its working title, holds that it upgrades it. The measurements reviewed here falsify the first in its strong form and leave the second untested.
What the scanners show is a brain that changes faster and more widely during gestation than adult brains are usually seen to change, that reverses part of that change after birth, and that keeps a residue whose cellular nature is unknown. What the memory tests show is a small dip near term and no lasting decline in the women followed from before conception. The thesis that survives is the narrower one: pregnancy remodels the brain, at measurable scale and in a pattern tied to the demands of caring for an infant, without a lasting cost on the measures so far applied. Whether the remodelling brings gains beyond that demand is a question the data do not yet reach.
- MIT Press, Susana Carmona and Lily Meyer, A Mother’s Brain: The New Science of the Neuro-Maternal Revolution, release date 6 October 2026, catalogue listing on Apple Books (publisher listing): 192 pages; description states that the book challenges “mommy brain” stereotypes and argues that pregnancy changes the brain more profoundly than adolescence. https://books.apple.com/us/book/a-mothers-brain/id6758185589 (accessed 10 November 2026) ↩︎
- Penguin Random House, Dra. Susana Carmona, Neuromaternal: ¿Qué le pasa a mi cerebro durante el embarazo y la maternidad?, Ediciones B, 16 July 2024 (publisher listing): Spanish edition, 304 pages. https://www.penguinrandomhouse.com/books/771135/neuromaternal-que-le-pasa-a-mi-cerebro-durante-el-embarazo-y-la-maternidad–neuromaternal-what-happens-to-my-brain-during-pregnancy-and-motherhood-by-dra-susana-carmona (accessed 10 November 2026) ↩︎
- Kristen French, “The Myth of Mommy Brain”, Nautilus, 4 November 2026 (press report; interview): Carmona on memory deficits in the third trimester and first postpartum years, long-term cognition, UK Biobank caveats, gestational and non-gestational parents, and the Popper epigraph of her book. https://nautil.us/the-myth-of-mommy-brain-1286337 (accessed 10 November 2026) ↩︎
- Hoekzema E., Barba-Müller E., Pozzobon C., Picado M., Lucco F., García-García D., Soliva J.C., Tobeña A., Desco M., Crone E.A., Ballesteros A., Carmona S., Vilarroya O., “Pregnancy leads to long-lasting changes in human brain structure”, Nature Neuroscience 20(2): 287–296, 2017, doi:10.1038/nn.4458 (peer-reviewed): 25 first-time mothers, 20 control women, 19 fathers, 17 control men; classification accuracy; attachment models; two-year follow-up of 11 mothers; cognitive tasks. Author manuscript at https://scholarlypublications.universiteitleiden.nl/handle/1887/57549 (accessed 10 November 2026) ↩︎
- Paternina-Die M., Martínez-García M., Martín de Blas D., Noguero I., Servin-Barthet C., Pretus C., Soler A., López-Montoya G., Desco M., Carmona S., “Women’s neuroplasticity during gestation, childbirth and postpartum”, Nature Neuroscience 27(2): 319–327, February 2024, doi:10.1038/s41593-023-01513-2 (peer-reviewed): 110 first-time mothers and 34 nulliparous women; replication sample of 29 mothers and 24 controls. Repository record at https://repisalud.isciii.es/entities/publication/dc6c88fd-629a-47e8-accc-2bc865cf285a/full (accessed 10 November 2026) ↩︎
- Pritschet L., Taylor C.M., Cossio D., Faskowitz J., Santander T., Handwerker D.A., Grotzinger H., Layher E., Chrastil E.R., Jacobs E.G., “Neuroanatomical changes observed over the course of a human pregnancy”, Nature Neuroscience, 16 September 2024, doi:10.1038/s41593-024-01741-0 (peer-reviewed): one 38-year-old first-time mother, 26 MRI sessions from three weeks before conception to two years after birth; hormone values; regional grey-matter, white-matter and cerebrospinal-fluid trajectories; discussion of mechanisms. https://escholarship.org/content/qt73w7c7pk/qt73w7c7pk.pdf (accessed 10 November 2026) ↩︎
- National Institutes of Health, “Brain changes observed during pregnancy”, NIH Research Matters, 1 October 2024 (official): summary of Pritschet et al. 2024, study leads and funding by the National Institute on Aging and the National Institute of Mental Health. https://www.nih.gov/news-events/nih-research-matters/brain-changes-observed-during-pregnancy (accessed 10 November 2026) ↩︎
- Paternina-Die M., Martínez-García M., Pretus C., Hoekzema E., Barba-Müller E., Martín de Blas D., Pozzobon C., Ballesteros A., Vilarroya O., Desco M., Carmona S., “The paternal transition entails neuroanatomic adaptations that are associated with the father’s brain response to his infant cues”, Cerebral Cortex Communications 1(1), 2020, doi:10.1093/texcom/tgaa082 (peer-reviewed): 20 first-time fathers and 17 childless men; precuneus volume and thickness. Repository record at https://e-archivo.uc3m.es/entities/publication/df6394c8-485b-4de9-a99d-8e9ef74064fa/full (accessed 10 November 2026) ↩︎
- de Lange A.-M.G., Kaufmann T., van der Meer D., Maglanoc L.A., Alnæs D., Moberget T. et al., Westlye L.T., “Population-based neuroimaging reveals traces of childbirth in the maternal brain”, Proceedings of the National Academy of Sciences 116(44): 22341–22346, 2019, doi:10.1073/pnas.1910666116 (peer-reviewed): 12,021 middle-aged women in the UK Biobank. University of Oslo record at https://www.sv.uio.no/psi/forskning/prosjekter/brainmint/publikasjoner/delange_pnas_2019.html (accessed 10 November 2026) ↩︎
- Ning K., Zhao L., Franklin M., Matlof W., Batta I., Arzouni N., Sun F., Toga A.W., “Parity is associated with cognitive function and brain age in both females and males”, Scientific Reports 10: 6100, 2020, doi:10.1038/s41598-020-63014-7 (peer-reviewed): UK Biobank; cognitive analyses of 160,077 women and 143,119 men; brain-age analyses of 6,822 women and 6,762 men; Table 4. https://www.nature.com/articles/s41598-020-63014-7.pdf (accessed 10 November 2026) ↩︎
- Davies S.J., Lum J.A.G., Skouteris H., Byrne L.K., Hayden M.J., “Cognitive impairment during pregnancy: a meta-analysis”, Medical Journal of Australia 208(1): 35–40, published online 15 January 2018, doi:10.5694/mja17.00131 (peer-reviewed): 20 studies, 709 pregnant and 521 non-pregnant women; pooled standardised mean differences by domain and trimester. https://www.mja.com.au/journal/2018/208/1/cognitive-impairment-during-pregnancy-meta-analysis (accessed 10 November 2026) ↩︎
- Christensen H., Leach L.S., Mackinnon A., “Cognition in pregnancy and motherhood: prospective cohort study”, British Journal of Psychiatry 196(2): 126–132, February 2010, doi:10.1192/bjp.bp.109.068635 (peer-reviewed): 1,241 women aged 20–24 assessed in 1999, 2003 and 2007; 76 pregnant at follow-up, 188 new mothers, 542 nulliparous. https://www.cambridge.org/core/journals/the-british-journal-of-psychiatry/article/cognition-in-pregnancy-and-motherhood-prospective-cohortstudy/812FCA4BC00BCCC65E27715C6FB4E89E (accessed 10 November 2026) ↩︎
- Christian Jarrett, “The maternal brain”, The Psychologist, British Psychological Society, 15 March 2010 (press report; review of studies): reports Crawley, Dennison and Carter (2003), Crawley, Grant and Hinshaw (2008) and Henry and Rendell (2007). https://bps.org.uk/psychologist/maternal-brain (accessed 10 November 2026) ↩︎
- Emily G. Jacobs, “Only 0.5% of neuroscience studies look at women’s health. Here’s how to change that”, Nature 623, 23 November 2023, doi:10.1038/d41586-023-03614-1 (commentary): fewer than 0.5% of more than 50,000 human brain-imaging articles consider health factors specific to women. Author copy at https://escholarship.org/content/qt4mm0n53s/qt4mm0n53s.pdf (accessed 10 November 2026) ↩︎
- Context
- The English edition of Susana Carmona's A Mother's Brain (MIT Press, 6 October 2026) and the claim that pregnancy upgrades the brain, set against the longitudinal neuroimaging studies of first-time mothers (Hoekzema et al. 2017; Paternina-Die et al. 2024; Pritschet et al. 2024), UK Biobank studies of parents in middle age, and meta-analytic and prospective evidence on cognition in pregnancy.
- Causes
- The hormonal changes of gestation, including oestradiol and progesterone rising by two to three orders of magnitude; fluid shifts during pregnancy; unidentified cellular changes that persist after birth; and the behavioural and social changes of parenthood that fathers share.
- Actors
- Susana Carmona and the Neuromaternal laboratory in Madrid; the research teams of Elseline Hoekzema, María Paternina-Die, Laura Pritschet, Emily Jacobs and Elizabeth Chrastil, Ann-Marie de Lange, Kaida Ning, Sasha Davies and Helen Christensen; the UK Biobank; the US National Institutes of Health; MIT Press and Ediciones B.
- Perspectives
- Separates structural measurement from function and from cognitive performance; distinguishes longitudinal designs with pre-pregnancy baselines from cross-sectional comparisons; tests the pruning hypothesis against fluid and parenthood explanations; labels illustrative calculations; records the strongest counter-argument; gives no health advice and makes no judgement of researchers or parents.
- Relationships
- Opens the Psyche Domain. Links to Oikos (Marriage Became Optional. Then It Became a Luxury, https://titani.ca/dispatch/marriage-became-optional/) and Soma (Pilates Promised Fitness at Nominal Cost. The Reformer Made It Bookable, https://titani.ca/dispatch/pilates-reformer-made-it-bookable/).
