Menopause averted a midlife energetic crisis with help from older children and parents: A simulation study
Abstract
The grandmother hypothesis, which proposes that ancestral women ceased reproduction midlife to instead provision their adult daughters and grandchildren, is the most influential account of the evolution of human menopause. Here I use simulations to investigate two theories that ground the evolution of menopause in provisioning by parents and grandparents of both sexes. Kaplan et al. (2010) proposed a “two-sex” account in which the high energetic burden of caring for multiple, slow-developing offspring was met with biparental investment. According to this model, menopause evolved because the physiological costs of pregnancy and childbirth increased with age, yet productivity also increased with age, and the benefits of both grandparents transferring resources to adult children and their offspring eventually outweighed the benefits of continued reproduction. Kuhle (2007) proposed the “father absent” hypothesis in which the higher mortality rate of husbands would often have left wives without the resources to raise young children, selecting for early reproductive cessation. Neither hypothesis considered juvenile productivity, which subsequent studies found to be surprisingly high. Simulations of hunter-gatherer energy consumption and production across the lifespan, taking account of age- and sex-specific survivorship, interbirth intervals, and varying rates of foraging skill acquisition typical of contemporary foragers, reveal a pronounced midlife energy deficit that could be averted with the increasing production of maturing juveniles, midlife cessation of reproduction, and energy transfers from from both older parents (i.e., grandparents), and also sometimes younger couples (e.g., brideservice). Menopause emerges as an integral and necessary component of the unique human pattern of relatively short interbirth intervals and a long period of juvenile dependency, supporting and extending both the ECM and the grandmother hypothesis. DOI: 10.5281/zenodo.18149332
1 Introduction
Prolonged post-reproductive lifespans are rare in wild animals (Chapman et al., 2024; Ellis, Franks, Nattrass, Cant, et al., 2018; Monaghan & Ivimey‐Cook, 2023; cf. Winkler & Goncalves, 2023), which is consistent with the expectation that somatic and reproductive functions should senesce at similar rates (Croft et al., 2015). The explanation for the oddly long post-reproductive lifespans of females in humans and a few other species, on the other hand, remains hotly debated (see Figure 1). In brief, byproduct or non-adaptive explanations include that these are artifacts of unusually benign environments (e.g., captivity, recent improvements in public health) that extend lifespan but not fertility; that these are epiphenomena of antagonistic pleiotropy; or that these are artifacts of selection for male longevity that extended female lifespan but not fertility.
Adaptive explanations include the influential grandmother hypothesis, in which the benefits of investing in grandchildren outweighed the fitness benefits of producing more children; the mother hypothesis, in which the benefits of maintaining investment in multiple slow-developing offspring outweighed the increasing risk of maternal morbidity and mortality from continued reproduction; the “aging eggs” hypothesis, in which increased risk of chromosomal abnormalities selected for a cessation of reproduction; and the avoidance of breeding competition with genetic kin. These and other hypotheses have been extensively reviewed elsewhere (Arnot, 2021; Croft et al., 2015; Ellis et al., 2024; Hawkes, 2020, 2025; Holmes, 2019; Monaghan & Ivimey‐Cook, 2023; Peccei, 2001; Sievert, 2024), so I will not revisit them here. Instead, I will investigate two related hypotheses for the evolution of human menopause that have received relatively little attention.
1.1 The “two-sex” embodied capital model of menopause
Kaplan et al. (2010) proposed a “two-sex” learning and skill-based model of the evolution of human menopause. In this account, humans evolved to exploit a highly knowledge- and skill-based “niche” providing energy-rich but difficult to extract resources that required the rearing of slow-developing, large-brained, and thus energetically expensive offspring (Kaplan et al., 2000). Offspring were produced every few years but required almost two decades to become self-sufficient, placing a potentially high energetic burden on mothers. This burden is met by a co-evolving sexual division of labor in long-term monogamous marriages in which husbands specialized in skill-based acquisition of high-risk, high-return resources, with their skills and productivity increasing for about two decades beyond the age at which they became self-sufficient. Mothers specialized in equally skill-based child rearing and acquisition of lower-risk, lower-return resources (Kelly, 2013). Menopause evolved because the physiological costs of pregnancy and childbirth increased with age, and oocyte quality decreased with age, yet productivity also increased with age. In this scenario, around midlife the fitness benefits of downward intergenerational transfers of resources (food sharing) by both parents, whose acquisition was based on four decades of skill development, termed embodied capital, outweighed the fitness benefits of continued offspring production. This account, which could be dubbed the grandparents hypothesis, is part of the highly influential Embodied Capital Model (ECM, Kaplan et al., 2000). See Figure 2A,B.
1.2 The absent father model of menopause
Kuhle (2007) proposed a two-part “absent father” hypothesis for the evolution of menopause in humans that assumes the evolutionary importance of a sexual division of labor in long-term marriages, similar to the scenario proposed by Kaplan and colleagues (Kaplan et al., 2010, 2000). The first part proposes that fathers who abandoned their wives for younger women would have left their ex-wives without “the necessary level of paternal investment to rear newborns to reproductive age” (p. 332). The second part proposes that because male mortality is greater than female mortality, and because husbands tend to be older than their wives, husbands would often have died before their wives, again leaving the wives without the resources necessary to rear newborns. Menopause would have limited the childrearing burden placed on older single mothers. Unlike the more influential mother and grandmother hypotheses, the Kaplan et al. (2010) and Kuhle (2007) hypotheses both propose that a sexual division of labor in long-term monogamous relationships played a central role in the evolution of human menopause.
Although serial monogamy is not uncommon in the ethnographic record, Kaplan et al. (2010) emphasize the cost of mate-switching, and provide evidence that in small scale societies most children are produced and raised in long-term monogamous relationships. I will therefore set the wife abandonment hypothesis aside and focus on the “two sex” aspect of the Kaplan et al. (2010) and Kuhle (2007) hypotheses.
1.3 New data on the ontogeny of foraging skills
The ECM, based on the data available at the time, emphasized that adult men produced a substantial surplus of energy from high return but difficult to acquire resources, with maximum production around age 35-40, reproductive-aged women produced approximately their own daily total energy expenditure (TEE), and post-reproductive women produced a modest surplus (Gurven et al., 2006; Kaplan et al., 2000; Walker et al., 2002). These surpluses supported a growing family of slowly developing offspring, each of whom would require energetic subsidies for up to 20 years. See Figure 2A,B.
Subsequently, two important sets of empirical results suggest a modification of the ECM might be necessary. First, in many populations, proficiency in hunting and other food production occurs earlier in adulthood, and is more constant throughout adulthood, than depicted for men in Figure 2A. A study of hunting based on ~23,000 records from more than 1800 individuals from 40 locations, for example — mostly men — found that skill and productivity increased more rapidly with age than envisioned in the ECM, peaking in the early 30’s, and remaining flatter across adulthood (Koster et al., 2020). See Figure 2E. A study of the ontogeny of a number of skills among Tsimane horticulturists found that most skills, including food production, had been acquired by age 20, although self- and peer-ratings of high proficiency in food production for both men and women increased into their 40’s, consistent with the ECM (Schniter et al., 2015). See Figure 2D.
Second, there is increasing evidence that children are able to acquire substantial calories from wild foods (Pretelli et al., 2024). Hadza children, for example, when followed during foraging trips (and not just measuring food returned to camp), often acquired more calories during the day than required by an adult (Crittenden et al., 2013; Froehle et al., 2019). See Figure 2C. A study of child foraging using published data from 28 societies found that their foraging returns, as a fraction of returns by young adults, increased approximately linearly with age, albeit more slowly for more skill-intensive, difficult-to-extract resources (tubers and game), but more rapidly for easier-to-extract resources (fruit and fish/shellfish) where adult levels of productivity were reached by adolescence (Pretelli et al., 2022).
1.4 Study Aims
This study used simulations to investigate the energetic consequences of ages of last birth ranging from much younger than typical of human hunter gatherers, to much closer to the end of life, resembling the pattern seen in other primates (Figure 1). It also investigated the energetic consequences of variation in skill acquisition rates, juvenile energy production, and interbirth intervals, with all life history, demographic, and energetic factors except ages of last birth falling within the ranges typical of contemporary hunter-gatherers.
Although this study was conceived, in part, as a test of the ECM “two sex” hypothesis, the grandmother hypothesis similarly depends on a midlife reproductive cessation and postreproductive energy transfers, differing from the ECM mainly in the claim that men transfer their surplus energy, not necessarily to provision their spouses, children, or grandchildren, but to increase their social status, thereby gaining additional mating opportunities with fertile women (mate guarding also plays an important role, Hawkes, 2020, 2025). The implications of the results for both hypotheses will therefore be addressed.
It should be acknowledged at the outset that if menopause is an adaptation it co-evolved with demographic and life history factors, so the counterfactual that a species with, e.g., modern human interbirth intervals, maturation times, body masses, total energy expenditures, and lifespans, but much later menopause, or no menopause, is unrealistic. Nevertheless, this exercise could illuminate why it is unrealistic.
2 Methods
Taking inspiration, in part, from Gurven & Walker (2005), who evaluated the energetic consequences of slow vs. fast childhood growth, energy production and consumption in hunter-gatherers was modeled in two ways.
2.1 The population model
The population model simulated the daily energy production and consumption of each age and sex class across the entire population, from birth to the maximum human lifespan (age 80), accounting for survivorship by age and sex. Energy consumption used age-, sex- and body-weight-adjusted total energy expenditure (TEE) values from the literature. Energy production was based on age- and sex-specific skills and physical strength (proxied by body weight), and varied across a range of parameter values for the relative importance of skill vs. strength, skill acquisition rates, and maximum adult energy production values (details below). Production, consumption, and their difference (energy balance) were computed for each age-sex class according to their proportions in a stationary population with a given life expectancy at birth, and then summed for all ages and sexes. Positive values of energy balance would indicate that the population as a whole produces more energy than it consumes on a daily basis, and negative values that it consumes more energy than it produces.
This model was used to assess how variation in skill ontogeny and juvenile energy production impacted energy balance at the population level. It had the advantages that the proportions of each age and sex in the population corresponded to standard life tables (details below), energy transfers were inherent, and it made no assumptions about social organization – it was entirely agnostic regarding which individuals producing energy surpluses transfer them to which individuals in energy deficit. It had the disadvantages that it did not include variation in interbirth intervals or ages of last birth and their consequences for energy supply and demand.
2.2 The family model
The family model simulated a hunter-gatherer nuclear family (mother, father, and their joint offspring) starting at age at first birth (AFB) and continuing with one-year time steps for a maximum human lifespan. The mother gave birth at a regular integer-valued interbirth interval (IBI) until her age at last birth (ALB), which was varied from 30 to 60 years old, in 10 year increments. Juveniles matured and left the family at their AFB. Age- and sex-specific mortality followed standard human life tables. Energy production and consumption were computed as in the population model. Family energy balance was total family production minus total family consumption computed for each year from the AFB to the maximum age (80).
All members of the population resided in identical families that, for a given combination of parameters, followed identical trajectories across the lifecourse. One family with energy surpluses at some life stages (e.g., parents or grandparents) was assumed to transfer them to one family with energy deicits at other life stages (e.g., to the family of an adult offspring). A negative sum of energy balance over the life of the family for a given combination of parameters would therefore indicate insufficient energy to support the families in the population, and thus a nonviable region of the parameter space.
This model was used to evaluate the impact of variation in ALB’s, IBI’s, and skill acquisition rates on energy balance. It had the advantage that it could vary ALB and IBI to determine their impact on family composition, and hence on energy balance. It had the disadvantage that varying ALB and IBI altered age-specific birth rates, violating population stationarity, as further discussed below.
This model conceptualized transfers of adult energy surpluses as biological parents feeding their spouses, joint offspring, and grandoffspring, per the ECM, but the underlying computations only assumed that the energy surpluses produced by some individuals were used to compenstate for the energy deficits of other individuals. In other words, the computations were agnostic regarding who transfered energy to whom. These transfers could therefore be equally conceptualized as, e.g., some men transferring their energy surpluses, not to a single wife and their joint offspring, but to any fertile woman and her offspring who are in energy deficit, in exchange for mating opportunities, per the grandmother hypothesis (Hawkes, 2020, 2025).
2.3 Age- and sex-specific survivorship (mortality)
The average life expectancy at birth for human hunter-gatherers of both sexes combined is \(e_0 = 30\) years (Davison & Gurven, 2021). For the Hadza, female \(e_0=35.5\) and male \(e_0=30.8\) (Blurton Jones, 2016). An analysis of !Kung mortality (Howell, 2017) found that it did not differ meaningfully from other human populations with similar life expectancy at birth (Coale & Demeny, 1966; Li & Gerland, 2012), a conclusion echoed by Gurven & Kaplan (2007), who concluded that “there is a characteristic life span for our species, in which mortality decreases sharply from infancy through childhood, followed by a period in which mortality rates remain essentially constant to about age 40 years, after which mortality rises steadily in Gompertz fashion” (p. 322).
The Kuhle (2007) hypothesis depends on a gap in the age of marriage (see Section 2.8) combined with a sex difference in mortality, which is common in many human populations as well as in many wild mammals. A study of 101 mammal species, for example, found that the median female lifespan was 18.6% longer than that of conspecfic males, and in humans the mean female advantage was 7.8% (Lemaître et al., 2020). For age- and sex-specific survivorship (mortality) I used the UN Model Life Tables (Li & Gerland, 2012; data and code from Gaddy et al., 2025) with \(e_0\) values similar to the Hadza and !Kung, i.e., a female life expectancy at birth, \(e_0 = 35\), a male life expectancy at birth, \(e_0 = 30\), and a sex ratio at birth (SRB) of 1.05 (i.e., a 5% male excess). See Figure S1.
2.4 Age at first birth (AFB), age at last birth (ALB), and interbirth interval (IBI)
Davison & Gurven (2021) compiled demographic and life history values for contemporary hunter-gather populations, small-scale subsistence populations, and chimpanzee populations. I primarily used their averaged values from the five hunter-gatherer populations: Ache, Agta, Hadza, Hiwi, and !Kung (Ju/’hoansi). Although Davison & Gurven (2021) did not provide a numerical value for mean AFB, their Figures 2 and S2 indicate hunter-gatherer values clustering from shortly before 20 years of age to shortly after. They found that the average AFB for all human subsistence populations was shortly before 20 years. Hunter-gatherer ALB ranged from 36 to 42, with a median of about 38, and mean total fertility was 6.2. Hunter-gatherer IBIs ranged from 2.8 years (Agta) to 3.3 years (Hiwi), with a mean value of 3.1 years. To match the foregoing values as closely as possible in the simulations, AFB = 20 and IBI = 3 years, although IBI = 4 years was also investigated.
To determine the energetic consequences of earlier vs. later ALBs, it was important to first determine the plausible range of ALBs. Females in most primates give birth until the end of life (Figure 1). A curvilinear regression model of the maximum ALB of mammals vs. maximum lifespan found that the predicted ALB of species with a maximum lifespan of 100 years, such as humans and narwals (Monodon monoceros), would be about 60 years (Huber & Fieder, 2018). The narwal ALB was 68 years, greater than predicted, and the human ALB was 50, less than predicted. Three other species had ALB ≥ 60 years (values that were greater than predicted): African bush elephants (Loxodonta africana), Asian elephants (Elephas maximus), and North Atlantic right whales (Eubalaena glacialis). Moreover, this study noted that “there is little indication for reproductive senescence in any of the baleen whales so far”, some of which have lifespans up to 200 years (Huber & Fieder, 2018, p. 2). Thus, because some mammals with maximum lifespans approaching or exceeding that of humans have ALB’s approaching or exceeding 60 years, simulations were run with ALB = 30, 40, 50, and 60 years. For simplicity, in all simulation runs fertility was constant until ALB, at which point it was set to 0.
Birth at age x was conditional on female survival to age x. For women who survived to the ALB, an IBI = 3 and an ALB = 40 resulted in a total fertility rate (TFR) of 7, slightly above the mean empirical value, and for IBI = 4, TFR = 6. Taking into account that not all women survive to the ALB, however, in the simulations with IBI = 3, TFR = 6.1, and with IBI = 4, TFR = 5.2.
2.5 Population stationarity and growth
Contemporary foraging populations typically exhibit high annual population growth rates in excess of 1%, which is unsustainable over the long term, leading Gurven & Davison (2019) to propose that humans evolved as a colonizing species, with outmigration during periods of growth interspersed with relatively frequent population crashes. Populations are stationary, on the other hand, when births equal deaths over time. Among contemporary foragers, the !Kung were an exception to high population growth, with a relatively low rate of r = 0.17% attributable to a life expectancy at birth e0 = 33.9, and a TFR = 4.3 (Gurven & Davison, 2019). Roughly, with IBI = 4, and taking maternal mortality into account, mothers have about four children, about half of whom survive to reproductive age, replacing their parents, yielding an approximately stationary population. The family model assumes a stationary population in which each family strives to support itself and transfers any surplus to one family in energy deficit, e.g., to the family of one of its two surviving (or locally settling) adult offspring.
The Hadza had a more typical high growth rate of r = 1.38%, attributable to a similar e0 = 34.7, but a higher TFR = 6.2 (Gurven & Davison, 2019). The simulations therefore used IBI values of 3 and 4 years, which combined with the e0 values specified in the previous section, approximate the TFR values (and hence the population growth rates) of the Hadza (high) and !Kung (low), respectively. In growing populations (IBI = 3), outmigration is assumed to keep the local population stationary.
2.6 Hunter-gatherer family energy consumption
Total energy expenditure (TEE) is a function of age, body mass, body composition (fat and fat-free mass), and activity levels. Age-related impacts on TEE include the energy required for development, age-related changes in activity levels, and tissue-specific metabolism. For example, the masses of energetically expensive organs, such as liver and brain, comprise a greater fraction of body mass in children than adults. A study of TEE measurements using doubly labeled water (the gold standard method) in a large, diverse sample found four distinct life stages: fat-free mass–adjusted expenditure accelerates rapidly from 0 to 1 year, declines slowly to adult levels by ~20 years, remains approximately stable from 20 to 60, even during pregnancy, and then declines in older adults (Bajunaid et al., 2025; Pontzer et al., 2021).
Somewhat counter-intuitively, after adjusting for age, sex, and body mass, TEE is relatively constant within these four phases, regardless of activity or immune activation. Instead, an individual’s fixed energy budget is allocated among, e.g., growth, immunity, and activity. Children in horticultural Shuar vs. industrial populations have very similar TEE, for example, but compared to children in industrial populations, Shuar children have higher immune activation and activity, and reduced growth (Urlacher et al., 2019).
Accordingly, hunter-gatherer TEE was modeled using the regression model in Bajunaid et al. (2025), which comprises an equation that predicts TEE using age (years), sex, body mass (kg), height (cm), ethnicity, and elevation (m) (see Table 1 in Bajunaid et al., 2025), except that TEE for infants (age 0-1) was from Pontzer et al. (2021). Body mass and height were modeled as the average of generalized additive models (GAMs) of body masses and heights by age and sex of the Ache, Hadza, and !Kung. Ethnicity and elevation were set as “African” and 1000 m, respectively, yielding a mean adult female TEE = 2019 kcals/day, and a mean adult male TEE = 2581 kcals/day. See Figure S2. Family energy consumption was the sum of TEE for the mother, father, and their joint children for each age of the mother starting at AFB.
2.7 Adult hunter-gatherer energy production
A compilation of adult hunter-gatherer energy production by sex found that males typically (but not always) produced a surplus of calories beyond the typical adult male TEE, women generally produced around an adult female TEE (but in some cases with a surplus and in others with a deficit), and the combined male and female production was generally a surplus that would support offspring (Kraft et al., 2021). See Figure 3.
Taking the average TEE of a hunter-gatherer adult to be 2300 kcals/day, and using the energy production values in Figure 3 from Kraft et al. (2021), adult hunter-gatherer women produced from 0.4 to 1.6 of daily adult TEE, men from 1 to 2.4 of daily adult TEE, and one woman plus one man produced from 1.7 to 3.3 of daily adult TEE. Accordingly, maximum female energy production was modeled as a range from 0.4 to 1.6 of adult TEE, and maximum male energy production as a range from 1 to 2.2 of adult TEE, with the constraint that 2.2 ≤ TEEprop,f + TEEprop,m ≤ 3.4 (i.e., the range needed to support at least two adults with some surplus for juvenile offspring, with a maximum just slightly above Kraft et al., 2021).
2.7.1 Age- and sex-specific energy production
Hunter-gatherer energy production (kcals/day) by age and sex is fairly uncertain because there have been very few such populations to study, and measurements might be restricted to food returned to camp (ignoring food consumed during foraging), perhaps obtained only during parts of the year, and/or data from adults only. Drawing inspiration from modeling approaches in Gurven & Walker (2005), Gurven & Kaplan (2006), Koster et al. (2020), and Schniter et al. (2015), I modeled daily foraging productivity across the lifespan as follows:
\[ productivity(age) = \mathrm{TEE}_{prop} strength(age)^\alpha skill(age)^{1 - \alpha} \tag{1}\]
where TEEprop is the proportion of adult TEE/day, and ranged from 0.4 to 1.6 for women and from 1 to 2.2 for men, and \(0 \le \alpha \le 1\) is the relative importance of strength vs. skill, and was set to 0.25 (mostly skill-based), 0.5 (equal parts skill and strength), and 0.75 (mostly strength-based), separately for males and females. Strength was proxied by age- and sex-specific body weight (kg) as a proportion of maximum adult male weight, multiplied by a function that declines rapidly as age approaches the maximum age, representing senescence (e.g., reductions in muscle mass and bone density in old age):
\[ strength(age, sex) = \frac{weight(age, sex) (1 - e^{b_0 (age_{\mathrm{max}}-age)})}{weight_{\mathrm{max}}} \tag{2}\]
where \(b_0\) is a parameter that determines the rate of strength decline in old age, and was fixed at -0.15 for both sexes in all simulation runs.
Skill is the product of a sigmoidal function that predominates at young ages, similar to that fit to the data in Schniter et al. (2015), with a function that declines rapidly as age approaches the maximum age:
\[ skill(age) = \frac{1}{1 - e^{-b_1(age - age_{50})}} (1 - e^{b_0(age_{\mathrm{max}}-age)}) \tag{3}\]
where \(b_1\) is a parameter that determines the rate of skill acquisition, and \(age_{50}\) is the age at which skill is 50% of maximum. These two parameters were combined to create Fast (\(b_1 = 0.4\), \(age_{50} = 10\)), Medium (\(b_1 = 0.25\), \(age_{50} = 15\)) and Slow (\(b_1 = 0.15\), \(age_{50} = 20\)) skill acquisition ontogenies, separately for each sex. \(b_0\) is a parameter that determines the rate of skill decline in old age, and was fixed at -0.15 for both sexes in all simulation runs. Productivity was set to 0 for children less than 3 years old. See Figure 2F and Table 1.
| Foraging skill ontogeny | \(b_1\) | 50% | 95% |
|---|---|---|---|
| Fast | 0.40 | 10 | 17 |
| Medium | 0.25 | 15 | 27 |
| Slow | 0.15 | 20 | 40 |
2.8 Age gaps in marriage
The absent father hypothesis assumes that, on average, husbands are older than wives (Kuhle, 2007). Binford (2001) compiled ages at first marriage from the ethnographic record for men and women in 177 hunter-gatherer societies (see Figure S3). The modal, median, and 75% quantile age differences were 2, 5, and 9.5 years, respectively. Marriage age gaps of 2, 5, and 10 years were therefore used to compute the probability that the husband and the wife would be alive, and the number of dependents, for each age of the wife (for simplicity, however, age gaps were only used for this computation, and not used in the family model).
2.9 Parameter space and simulated lifecourse
In the population model of the hunter-gatherer lifecourse, one simulation was run for every combination of the skill ontogeny and energy production parameters in Table 2 (i.e., omitting IBI and ALB). In the family model, one simulation was run for all combinations of parameters, including IBI and ALB. In both models, the sum of maximum adult male and female energy production was constrained to 2.2 ≤ TEEprop,f + TEEprop,m ≤ 3.4, as noted earlier.
In the family model, for each of the resulting 7776 combinations of parameter values the following vectors were computed in sequence:
- wife and husband ages were integer-valued vectors from AFB until the maximum age (80), at one-year intervals.
- wife’s probability of survival at each age relative to her AFB was computed with the female life table.
- husband’s probability of survival at each age relative to AFB was computed with the male life table (thus, there was slightly less than one husband per wife at AFB).
- pregnancies (coded as 0, 1) occurred every IBI years, starting at one year prior to AFB.
- births (coded as 0, 1) occurred every IBI years, starting at AFB and ending at ALB, and then scaled by mother survival to age x.
- child survival was age- and sex-specific survival multiplied by births (i.e., no new children in years with birth = 0).
- wife, husband, and child TEE’s at each age (as described earlier) were multiplied by their survival at each age.
- wife, husband, and child productivities at each age were computed using Equation 1 and multiplied by their respective survival at each age.
- The number, energy consumption, and energy production of all dependent children at each wife age were computed assuming offspring disperse (leave the family) at the AFB, and using a windowing function based on wife’s age and wife survival (such that the number of children was scaled by the number of women who survived to give birth; see code).
- family consumption and production at each wife age were the sums of wife’s, husband’s, and total dependent children’s consumption and production, respectively.
- family energy balance was family production - family consumption for each age.
- cumulative family energy balance was the cumulative sum of family energy balance at each age from AFB to the end of life (maximum age).
Neither family member survival, nor pregnancies, births, IBI’s, or any other factor were contingent on energy balance. Simulations were deterministic, i.e., there were no stochastic parameters. See code for details.
| Parameter | Description | Values |
|---|---|---|
| \(e_{0,f}\) | Female life expectancy at birth (years) | 35 |
| \(e_{0,m}\) | Male life expectancy at birth (years) | 30 |
| agemax | Maximum human age (years) | 80 |
| SRB | Sex ratio at birth | 1.05 |
| AFB | Age at first birth (years) | 20 |
| ALB | Age of last birth | 30, 40, 50, 60 |
| IBI | Interbirth interval (years) | 3, 4 |
| agegap | Marriage age gap (years) | 0, 2, 5, 10 |
| \(\alpha_f\), \(\alpha_m\) | Relative importance of skill vs. strength for females and males | 0.25, 0.5, 0.75 |
| age50,f, age50,m | Age of 50% skill acquisition for females and males | 10, 15, 20 |
| \(b_0\) | Rate of skill decline with increasing age | -0.15 |
| \(b_{1,f}\), \(b_{1,m}\) | Skill acquisition rate for females and males | 0.15, 0.25, 0.4 |
| TEEprop,f | Maximum productivity for females as a proportion of adult TEE | 0.4, 0.8, 1.2, 1.6 |
| TEEprop,m | Maximum productivity for males as a proportion of adult TEE | 1, 1.4, 1.8, 2.2 |
Simulations were conducted using R version 4.4.1 (2024-06-14). Code and data are available here: https://github.com/grasshoppermouse/menopause.
3 Results
The main results of the population model demonstrate that juvenile production was essential for maintaining population-level energy balance near or above zero across most of the parameter space, and was unnecessary only at the highest level of adult production (Figure 4). The mean effects of skill ontogeny parameter variation are depicted in Figure S6. Variation in maximum adult energy production (TEEprop) had by far the biggest effect, followed by the rate of skill acquisition (b1) and the relative importance of strength in energy production (\(\alpha\)).
The main results of the family model demonstrate that, averaging across the parameter space, as young couples produced offspring, family energy consumption outpaced production due to a short IBI, and families tended to fall into negative energy balance. Later in the lifecourse, reproductive cessation, the increasing production of older dependent offspring, and the departure of mature offspring enabled families to eventually produce a surplus that could be transferred to families in energy deficit (e.g., to the families of their adult offspring).
As ALB increased, however, families grew larger, and stayed large longer, thereby accumulating more energy “debt” earlier in the lifecourse, and were less able to “pay off” this larger debt because they had fewer years later in the lifecourse to generate an energy surplus that could be transferred to families in energy deficit. Averaging over the parameter space, ALBs ≤ 40 were in energy balance or produced a net energy surplus by the end of the lifecourse, whereas ALBs > 40 had a net energy deficit by the end of the lifecourse. See Figure 5.
More details on these factors are provided in the following sections.
3.1 Family size
Energy balance tracked the number and ages of dependents (Figure 5), which were determined by the IBI and ALB. For a given IBI, family trajectories were identical prior to the youngest ALB (30 years), at which point they diverged. The number of dependents reached its maximum of around 4.2 (IBI = 3) when parents were age 38 because they had their firstborns at age 20, who then dispersed 20 years later (at their AFB). If there were no mortality, then for ALBs > 40, the number of dependents would remain constant until the ALB because as the oldest dependent child dispersed a new child would be born. In the simulations, however, the number of family members, including dependent children, slowly decreased due to husband mortality and to maternal mortality that also reduced the birth of new children (Figure 5).
Critically, as ALB increased, the average number of dependent children and their daily energy needs increased because new children continued to be born as older children matured and dispersed, although child and maternal mortality also took their toll. For example, with ALB = 40, families were supporting an average of 1.3 dependent children each day from the parents’ AFB to the end of life, and dependents’ mean total energy need was 2077 kcals/day, whereas with ALB = 60, families were supporting an average of 2.2 dependent children each day, and their mean total energy need was 3487 kcals/day. See Figure S7.
3.2 Energy consumption, production, and balance
As family size grew and as children grew, daily family energy consumption grew, and as it shrunk, energy consumption shrunk, but critically, more rapidly for lower ALBs because dispersing offspring were not replaced by new offspring. Family energy production increased slowly for about the first decade after AFB, but more rapidly in the second decade as the productivity of maturing children increased. Subsequently, energy production declined due to wife and husband mortality, and the departure of older productive children as they reached their AFB. Averaging over the parameter space, consumption exceeded production early in the lifecourse, causing an increasing energy “debt” (a cumulative negative energy balance), which began to be “repaid” after reproductive cessation and the increasing production and eventual departure of older children. See Figure 6.
Energy balance trajectories varied due to variation in the parameters affecting productivity, primarily TEEprop and b1, each of which varied by sex. See Figure S8.
3.3 Brideservice and helpers-at-the-nest
Although in the simulations men and women both married at AFB, the age of marriage is often later for men than women. Whereas young women (age 20-25) produced a surplus in 52% of the parameter space, young men produced a surplus in 76% of the parameter space. See Figure S9. In hunter-gatherer societies in which young adults remain with their natal families prior to marriage, net energy producers are serving as “helpers-at-the-nest” (Hagen & Barrett, 2009; Hames & Draper, 2004). Brideservice, in which the bridegroom works for the bride’s family in exchange for marriage, is also common in many hunter-gatherer societies (Walker et al., 2011), so some surplus production of young men, including early in marriage when young couples would have few or no offspring, could be helping provision their wives’ younger siblings.
3.4 Parent mortality
Mortality of both wives and husbands is high. With no marriage age gap, by age 54 only one parent has survived (wife = 55%, husband = 46%). With IBI = 3 and ALB = 40, the single parent would be caring for 0.93 older dependents, on average, whereas with ALB = 60, it would be 3.13 dependents, some probably quite young (e.g., 2, 5, and 8). With a marriage age gap of 10 years and ALB = 40, the single parent is somewhat more likely to be a younger wife, age 49 (wife = 62%, husband = 37%), who would be caring for 1.72 older dependents, on average, whereas with ALB = 60, it would be 3.13 dependents. See Figure S10 and Figure S11.
4 Discussion
The key insight from the simulations is that with a counterfactual human ALB closer to the end of life, similar to other primates (Figure 1), and given the energy consumption and production rates of contemporary foragers, the modern human pattern of relatively short interbirth intervals, long periods of juvenile dependency, minimal female reproductive skew, and long lifespans, could not have evolved. Shortly after AFB in most simulations, parents with multiple highly dependent offspring would not be able to feed them, nor would surplus food be available from older families still feeding many of their own dependents (Figure 5 and Figure 6).
This energetic crisis is averted by ceasing reproduction midlife combined with increasing juvenile productivity and cooperative energy transfers (food sharing). Although the simulations did not explicitly include energy transfers between nuclear families, it was assumed that the energy surpluses that were present at some life stages in a family in most simulations, were transferred to a family in energy deficit at other life stages. Food sharing and need-based transfers have been documented in many small-scale societies (Ringen et al., 2019; Smith et al., 2019, and references therein). These results are broadly consistent with Kaplan et al. (2010) (see Figure S4), albeit with important distinctions noted below.
It is possible that, sans menopause, ancestral foragers could simply have extracted more energy from the environment, and learned to do so rapidly. Maximum family energy consumption in the simulations with IBI = 3 is 9,985 kcals/day, and with IBI = 4 is 8,288 kcals/day. In actual contemporary foraging populations, maximum parent energy production is 7,544 kcals/day by the Nukak, with Hadza and !Kung parents not far behind (Figure 3). In regions of the parameter space with high maximum productivity and rapid skill acquisition, dependent children could make up the difference. Under these conditions, however, families generally produce an energy surplus across the entire lifespan, implying that changes to human life history parameters would have evolved to take advantage of the surplus, such as changes to interbirth intervals, growth and maturation rates, and ages of first birth and dispersal.
The simulations therefore suggest the following role of menopause in the co-evolution of modern human life history traits. Relatively short interbirth intervals combined with long periods of juvenile dependence sow the seeds of a mid-life energetic crisis: families have produced many young offspring with high energy requirements but who, per the ECM, require many years to acquire the skills necessary to provision themselves. Parents averted this crisis with three evolved strategies. First, by ceasing the production of new offspring in midlife (menopause), around age 40, parents’ energetic investments would then go solely to existing juvenile offspring whose increasing skills enabled them to begin supporting themselves before their dispersal, at which point parents began climbing out of an energy deficit even without subsidies (Figure 6). This would have been especially effective if offspring skill acquisition was relatively rapid. This refocuses the mother hypothesis on the energy production capacity of the mother, father, and their older offspring.
Second, young adults with few or no children often produced a surplus (Figure S9) that could subsidize their immature siblings to whom they are as closely genetically related as to their own offspring, akin to the “helper at the nest” pattern seen in some other cooperatively breeding species (Hagen & Barrett, 2009; Hatchwell, 2009). This is also consistent with the brideservice practices of many hunter-gatherer societies in which the bridegroom works for the bride’s family in exchange for marriage; brideprice is a related practice in which the bridegroom’s family pays the bride’s family with some form of wealth (Goody & Tambiah, 1973; Walker et al., 2011).
Third, older parents, perhaps especially the wife’s parents (Daly & Perry, 2017), produced a surplus that could subsidize their younger adult children’s children (the grandparent hypothesis) because the older parents’ one or two dependent older children largely supported themselves or perhaps even produced a surplus, or their children have all dispersed.
4.1 Rethinking the father absent hypothesis to include mother absence
Kuhle (2007) proposed that menopause might have evolved to mitigate the negative consequences of higher paternal than maternal mortality. At age 54 in a species with ALB = 40, the surviving single parent would be caring for about 0.93 older children, whereas with ALB = 60 it would be 3.13 children, some quite young (Figure S10). As marriage age gaps increase, the age at which only one parent remains decreases, and it is more likely to be the wife, but not dramatically so (Figure S11). It is possible that limiting the childcare burden on a surviving parent to a few older dependent offspring, which probably would have been manageable, rather than the care of several younger ones, which probably would not have been manageable, was a selection pressure for menopause. However, because the death of the wife would also have been common, and because the fitnesses of monogamous parents are tightly coupled, it is parental absence, not necessarily father absence, that matters.1
4.2 Rethinking menopause and the ECM
There are three important distinctions between the results of the simulations reported here, and the Kaplan et al. (2010) “two sex” model. First, Kaplan et al. (2010) emphasize the importance of the increasing costs of pregnancy and childbirth with increasing age (e.g., decreasing oocyte quality): in their model, menopause evolves because at midlife these costs reduce the benefits of continued reproduction which then no longer outweigh the benefits of transferring resources to kin.
In the simulations, in contrast, menopause evolves simply because, given the amounts of energy that foragers can extract from environment, the relatively rapid production of offspring who require substantial calories, yet who cannot support themselves for a decade or more, limits the number of dependents to about three or four. With high infant mortality, this limit is reached about 20 years after age at first birth. Nevertheless, age-related increases in the costs of pregnancy and childbirth could be additional factors favoring menopause.
Second, juvenile productivity plays little role in the Kaplan et al. (2010) model but an essential role in the simulations, where, in combination with reproductive cessation, the increasing productivity of older juveniles (Pretelli et al., 2024, 2022) helps families escape an energy deficit. Moreover, without juvenile production, families would be in a pronounced negative energy balance for most of the wife’s lifespan – there would be virtually no energy surpluses to offset these deficits (Figure 4).
Third, although I have framed the simulations in terms of nuclear families comprising a wife and husband in a lifelong monogamous relationship raising their joint offspring — a key assumption of the Kaplan et al. (2010) “two sex” model — the results of the simulations do not rely on this assumption. The results are equally consistent with, e.g., a pooled energy model (Kramer & Ellison, 2010) in which all females reproduce (i.e., no non-reproductive females) and all individuals pool their resources to provision every member of the community according to their needs. The potential midlife energetic crisis is simply a consequence of producing offspring every three or four years who require substantial calories to reach adulthood yet cannot provide all those calories themselves for a decade or more. In this regard, results are also consistent with the venerable grandmother hypothesis (Hawkes, 2020, 2025), a point to which I will return. Note, however, that the simulations do not address the collective action problems or other evolutionary challenges of energy transfers.
Although ceasing reproduction midlife and transferring energy surpluses are key to averting an energetic crisis, the increasing productivity of dependent offspring is also key. Yet in mobile hunter-gatherers ranging over a large, spatially and temporally varying landscape replete with competitors, predators, and other sundry hazards (e.g., Blurton Jones et al., 1994), child productivity in ancestral foraging bands would have depended on foraging bands moving to the right place at the right time (Bettinger & Grote, 2016; Hamilton et al., 2016).
Embodied capital should therefore be conceptualized not only as the skills required to extract energy from a resource patch, but also as including the skills required to locate optimal resource patches. Older mothers and fathers would have had greater knowledge of the large ranges necessary to support the hunting and gathering lifestyle of a large-bodied hominin (Antón, 2002), and hence greater ability to lead their families to productive resource patches, a perspective that aligns with theories of menopause in toothed whales that emphasize the superior ecological knowledge of older matriarchs (Brent et al., 2015; Rendell et al., 2019), and given the lower mortality of females, perhaps also aligns with the predominance of female-biased leadership of collective movements in social mammals (Smith et al., 2022). It is also consistent with the anthropology of older women in traditional societies who increasingly take on important social roles (Brown, 1982; Jang et al., 2025), with research on the important ecological knowledge of older individuals in foraging societies (e.g., Biesele & Howell, 1981; Scalise Sugiyama, 2011; Wiessner, 2014; Wood et al., 2024), and with the computational services model, which emphasizes the cognitive challenges of raising multiple dependent offspring (Hagen et al., 2025).
4.3 The grandmother hypothesis vs. the ECM “two sex” hypothesis
Energy transfers from older adults to younger ones underlie both the grandmother hypothesis and the “two sex” hypothesis of menopause. It is therefore worth briefly comparing the two in light of the simulation results. In the simulations, parents become grandparents at age 40 when their first borns have their first child. At age 40, women with IBI = 3 have 3.3 dependent children, women with IBI = 4 have 2.9 dependent children, and both have 0.58 grandchildren (the proportion of firstborns that survive to AFB). On average across the parameter space when ALB = 40, families transition from an energy deficit to a surplus, and are thus able to invest in grandchildren, when wives are about 45 and there are either 2.6 dependent children (IBI = 3) or 2 dependent children (IBI = 4), and about 1.12 grandchildren (due to infant mortality). In other words, grandmothers are still mothering.
The primary difference between the ECM and the grandmother hypothesis is the degree of conflict over distribution of the energy surpluses that are generated by older juveniles and adults. According to the ECM, these individuals all prefer to transfer their surpluses to close biological kin, e.g., children, siblings, and the biological parents’ grandchildren, who are the nieces and nephews of the older juveniles, and whose families are likely now in energy deficit. According to the grandmother hypothesis, in contrast, there is a conflict of interest, with women preferring to transfer surpluses to their children and grandchildren, and men preferring to transfer surpluses to fertile women to outcompete other males for mating opportunities. High male intrasexual competition is a consequence of midlife reproductive cessation by females but not males, reducing the pool of fertile females (ages 18-40) to about 0.7 per adult male (ages 18-60) (Figure S5). The simulation models do not address these potential conflicts of interest and therefore cannot adjudicate between these two models.
More generally, because every individual has exactly one biological mother and one biological father, it was convenient to account for energy production and consumption in terms of nuclear families, i.e., biological parents and their offspring. Little hangs on this approach, however, because individuals of the same age and sex had identical energy production and consumption, mortality, IBI’s and ALB’s. Individuals whose production was less than their consumption required energy transfers (food sharing) from someone producing a surplus. In the real world, transfers required physical proximity. The substantial literature on various aspects of hunter-gatherer social organization, which can include patrilocality, matrilocality, bilocality, and neolocality, has ascribed these patterns to, e.g., inbreeding avoidance, male coalitions, paternity uncertainty, ecological variation, childcare, and warfare (e.g., Barnard, 1983; Ember, 1975; Hamilton et al., 2016, 2007; Kelly, 2013; Marlowe, 2004; Moravec et al., 2018). Simulation results suggest that age- and sex-based variation in fertility, mortality, energy consumption, energy production, and skill acquisition rates, which would determine feasible group compositions, and which undoubtedly depended on the local ecology, probably also played an important role.
4.4 Limitations
Parameters were varied independently (with the exception of male and female productivity, whose sum was constrained), yet in ancestral foragers, skill acquisition rates, productivity, the importance of strength, interbirth intervals, marriage age gaps, and so forth, likely covaried in complex ways, and were linked to ecological variation. The multidimensional parameter space used in the simulations did not incorporate the unknown covariation of these values across actual foraging societies, past or present, nor do averages across the parameter space represent an “average” foraging society.
Simulations also did not take into account that in real families and communities, parents and children adjust the time devoted to production in response to changes in demand (e.g., Kramer, 2005).
5 Concluding remarks
The ECM (Kaplan et al., 2010, 2000) is a leading explanation for the evolution of the distinctive human life history pattern involving relatively short interbirth intervals and long periods of dependency. Simulation results suggest that menopause was integral to the evolution of this pattern because it averted a midlife energetic crisis threatened by the rapid production of large bodied and large brained, and thus energetically expensive juveniles who would not become self-sufficient for a decade or more. Juveniles were subsidized, in part, by their older siblings and perhaps newly married sisters- or brothers-in-law, who in ethnographically documented hunter-gatherers, often lived with either the wife’s or husband’s parents (Hill et al., 2011). Young couples, initially with few or no children of their own, often produced an energetic surplus that could benefit, e.g., the wives’ younger siblings during brideservice (Walker et al., 2011). Couples’ later midlife reproductive cessation (menopause) optimized their family size at an energetically sustainable level as the energy production of their children ramped up (a variant of the mother hypothesis, albeit here also including fathers and older children), and energy transfers from couples’ parents, whose children have mostly dispersed, burgeoned (the grandmother hypothesis, albeit here also including grandfathers). Couples past the ALB with fewer but increasingly productive offspring (Pretelli et al., 2024, 2022) then became energetically self-sustaining, and finally energetically “profitable”, with the capacity to invest in their children’s families. Results are equally consistent, however, with a model in which men transfer surpluses to fertile women and their unrelated offspring in energy deficit to obtain mating benefits (Hawkes, 2020, 2025). High juvenile productivity probably depended on judicious resource patch choice by older individuals with substantial ecology knowledge, per the ECM.
In summary, the simulations revealed a human-specific midlife energetic constraint on continued reproduction that, perhaps in conjunction with other constraints such as mammalian reproductive senescence (Huber & Fieder, 2018) and kin competition in hominins (White et al., 2025), might help explain human’s distinctive midlife reproductive cessation and high levels of cooperative food sharing.
6 Acknowedgements
Thanks to Ray Hames, Nicole Hess, Ilaria Pretelli, Erik Ringen, Eric Schniter, and anonymous referees for helpful comments. AI was not used in any aspect of this study.
7 References
Antón, S. (2002). An ecomorphological model of the initial hominid dispersal from Africa. Journal of Human Evolution, 43(6), 773–785. https://doi.org/10.1006/jhev.2002.0602
Arnot, M. (2021). The evolutionary ecology of menopause. University College London.
Bajunaid, R., Niu, C., Hambly, C., Liu, Z., Yamada, Y., Aleman-Mateo, H., … Speakman, J. R. (2025). Predictive equation derived from 6,497 doubly labelled water measurements enables the detection of erroneous self-reported energy intake. Nature Food, 6(1), 58–71. https://doi.org/10.1038/s43016-024-01089-5
Barnard, A. (1983). Contemporary hunter-gatherers: Current theoretical issues in ecology and social organization. Annual Review of Anthropology, 193–214.
Bettinger, R. L., & Grote, M. N. (2016). Marginal value theorem, patch choice, and human foraging response in varying environments. Journal of Anthropological Archaeology, 42, 79–87. https://doi.org/10.1016/j.jaa.2016.03.002
Biesele, M., & Howell, N. (1981). ’The old people give you life’: Aging among !Kung hunter-gatherers. Other Ways of Growing Old: Anthropological Perspectives, 77–98.
Binford, L. R. (2001). Constructing frames of reference: An analytical method for archaeological theory building using ethnographic and environmental data sets. University of California Press.
Blurton Jones, N. G. (2016). Demography and evolutionary ecology of Hadza hunter-gatherers (Vol. 71). Cambridge University Press.
Blurton Jones, N. G., Hawkes, K., & Draper, P. (1994). Foraging returns of !Kung adults and children: Why didn’t !Kung children forage? Journal of Anthropological Research, 50(3), 217–248.
Brent, L. J. N., Franks, D. W., Foster, E. A., Balcomb, K. C., Cant, M. A., & Croft, D. P. (2015). Ecological knowledge, leadership, and the evolution of menopause in killer whales. Current Biology, 25(6), 746–750. https://doi.org/10.1016/j.cub.2015.01.037
Brown, J. K. (1982). Cross-cultural perspectives on middle-aged women. Current Anthropology, 23(2), 143–156.
Chapman, S. N., Ellis, S., Lahdenperä, M., Croft, D. P., & Lummaa, V. (2024). Menopause has not evolved as a general trait in mammals: A response to “Do mammals have menopause?” BioRxiv, 2024–2002.
Coale, A. J., & Demeny, P. (1966). Regional model life tables and stable populations. Princeton University Press.
Crittenden, A. N., Conklin-Brittain, N. L., Zes, D. A., Schoeninger, M. J., & Marlowe, F. W. (2013). Juvenile foraging among the hadza: Implications for human life history. Evolution and Human Behavior, 34(4), 299–304. https://doi.org/10.1016/j.evolhumbehav.2013.04.004
Croft, D. P., Brent, L. J. N., Franks, D. W., & Cant, M. A. (2015). The evolution of prolonged life after reproduction. Trends in Ecology & Evolution, 30(7), 407–416. https://doi.org/10.1016/j.tree.2015.04.011
Daly, M., & Perry, G. (2017). Matrilateral bias in human grandmothering. Frontiers in Sociology, 2. https://doi.org/10.3389/fsoc.2017.00011
Davison, R. J., & Gurven, M. D. (2021). Human uniqueness? Life history diversity among small-scale societies and chimpanzees. PLOS ONE, 16(2), e0239170. https://doi.org/10.1371/journal.pone.0239170
Ellis, S., Franks, D. W., Nattrass, S., Cant, M. A., Bradley, D. L., Giles, D., … Croft, D. P. (2018). Postreproductive lifespans are rare in mammals. Ecology and Evolution, 8(5), 2482–2494.
Ellis, S., Franks, D. W., Nattrass, S., Currie, T. E., Cant, M. A., Giles, D., … Croft, D. P. (2018). Analyses of ovarian activity reveal repeated evolution of post-reproductive lifespans in toothed whales. Scientific Reports, 8(1), 12833.
Ellis, S., Franks, D. W., Nielsen, M. L. K., Weiss, M. N., & Croft, D. P. (2024). The evolution of menopause in toothed whales. Nature, 627(8004), 579–585. https://doi.org/10.1038/s41586-024-07159-9
Ember, C. R. (1975). Residential variation among hunter-gatherers. Behavior Science Research, 10(3), 199–227.
Froehle, A. W., Wells, G. K., Pollom, T. R., Mabulla, A. Z., Lew-Levy, S., & Crittenden, A. N. (2019). Physical activity and time budgets of hadza forager children: Implications for self-provisioning and the ontogeny of the sexual division of labor. American Journal of Human Biology, 31(1), e23209.
Gaddy, H., Sear, R., & Fortunato, L. (2025). High rates of polygyny do not lock large proportions of men out of the marriage market. Proceedings of the National Academy of Sciences, 122(40). https://doi.org/10.1073/pnas.2508091122
Goody, J., & Tambiah, S. J. (1973). Bridewealth and dowry. CUP Archive.
Gurven, M. D., & Davison, R. J. (2019). Periodic catastrophes over human evolutionary history are necessary to explain the forager population paradox. Proceedings of the National Academy of Sciences, 116(26), 12758–12766. https://doi.org/10.1073/pnas.1902406116
Gurven, M., & Kaplan, H. (2006). Determinants of time allocation across the lifespan: A theoretical model and an application to the machiguenga and piro of peru. Human Nature, 17(1), 1–49.
Gurven, M., & Kaplan, H. (2007). Longevity among hunter-gatherers: A cross-cultural examination. Population and Development Review, 33(2), 321–365.
Gurven, M., Kaplan, H., & Gutierrez, M. (2006). How long does it take to become a proficient hunter? Implications for the evolution of extended development and long life span. Journal of Human Evolution, 51(5), 454–470. https://doi.org/10.1016/j.jhevol.2006.05.003
Gurven, M., & Walker, R. (2005). Energetic demand of multiple dependents and the evolution of slow human growth. Proceedings of the Royal Society B: Biological Sciences, 273(1588), 835–841. https://doi.org/10.1098/rspb.2005.3380
Hagen, E. H., & Barrett, H. C. (2009). Cooperative breeding and adolescent siblings: Evidence for the ecological constraints model? Current Anthropology, 50(5), 727–737.
Hagen, E. H., Garfield, Z. H., & Lightner, A. D. (2025). Headmen, shamans, and mothers: Natural and sexual selection for computational services. Evolution and Human Behavior, 46(1), 106651.
Hames, R., & Draper, P. (2004). Women’s work, child care, and helpers-at-the-nest in a hunter-gatherer society. Human Nature, 15(4), 319–341.
Hamilton, M. J., Lobo, J., Rupley, E., Youn, H., & West, G. B. (2016). The ecological and evolutionary energetics of hunter-gatherer residential mobility. Evolutionary Anthropology: Issues, News, and Reviews, 25(3), 124–132.
Hamilton, M. J., Milne, B. T., Walker, R. S., Burger, O., & Brown, J. H. (2007). The complex structure of hunter–gatherer social networks. Proceedings of the Royal Society B: Biological Sciences, 274(1622), 2195–2203. https://doi.org/10.1098/rspb.2007.0564
Hatchwell, B. J. (2009). The evolution of cooperative breeding in birds: Kinship, dispersal and life history. Philosophical Transactions of the Royal Society B: Biological Sciences, 364(1533), 3217–3227. https://doi.org/10.1098/rstb.2009.0109
Hawkes, K. (2020). The centrality of ancestral grandmothering in human evolution. Integrative and Comparative Biology, 60(3), 765–781. https://doi.org/10.1093/icb/icaa029
Hawkes, K. (2025). Revisiting “Grandmothers and the Evolution of Human Longevity”. American Journal of Human Biology, 37(4), e70045.
Hill, K. R., Walker, R. S., Božičević, M., Eder, J., Headland, T., Hewlett, B., … Wood, B. (2011). Co-residence patterns in hunter-gatherer societies show unique human social structure. Science, 331(6022), 1286–1289.
Hill, K., & Hurtado, A. M. (1996). Ache life history: The ecology and demography of a foraging people. Taylor & Francis.
Holmes, D. J. (2019). Do women age like whales? Integrating diverse perspectives on human menopause and postreproductive life spans. Current Opinion in Endocrine and Metabolic Research, 5, 58–67.
Howell, N. (2009). HTWT observations. Retrieved from http://hdl.handle.net/1807/17996
Howell, N. (2017). Demography of the Dobe! Kung. Routledge.
Huber, S., & Fieder, M. (2018). Evidence for a maximum "shelf-life"" of oocytes in mammals suggests that human menopause may be an implication of meiotic arrest. Scientific Reports, 8(1). https://doi.org/10.1038/s41598-018-32502-2
Jang, H., Kandza, V., Ntamboudila, F. K., & Boyette, A. (2025). Women’s self-reported autonomy in household decision-making varies between BaYaka foragers and Bandongo farmers in the Congo Basin. Hunter Gatherer Research, 1–24. https://doi.org/10.3828/hgr.2026.7
Kaplan, H., Gurven, M., Winking, J., Hooper, P. L., & Stieglitz, J. (2010). Learning, menopause, and the human adaptive complex. Annals of the New York Academy of Sciences, 1204(1), 30–42. https://doi.org/10.1111/j.1749-6632.2010.05528.x
Kaplan, H., Hill, K., Lancaster, J., & Hurtado, A. M. (2000). A theory of human life history evolution: Diet, intelligence, and longevity. Evolutionary Anthropology, 9(4), 156–185.
Kelly, R. L. (2013). The lifeways of hunter-gatherers: The foraging spectrum. Cambridge University Press.
Koster, J., McElreath, R., Hill, K., Yu, D., Shepard, G., Vliet, N. van, … Ross, C. (2020). The life history of human foraging: Cross-cultural and individual variation. Science Advances, 6(26). https://doi.org/10.1126/sciadv.aax9070
Kraft, T. S., Venkataraman, V. V., Wallace, I. J., Crittenden, A. N., Holowka, N. B., Stieglitz, J., … Pontzer, H. (2021). The energetics of uniquely human subsistence strategies. Science, 374(6575). https://doi.org/10.1126/science.abf0130
Kramer, K. L. (2005). Children’s help and the pace of reproduction: Cooperative breeding in humans. Evolutionary Anthropology, 14(6), 224–237.
Kramer, K. L., & Ellison, P. T. (2010). Pooled energy budgets: Resituating human energy-allocation trade-offs. Evolutionary Anthropology: Issues, News, and Reviews, 19(4), 136–147.
Kuhle, B. X. (2007). An evolutionary perspective on the origin and ontogeny of menopause. Maturitas, 57(4), 329–337.
Lemaître, J.-F., Ronget, V., Tidière, M., Allainé, D., Berger, V., Cohas, A., … Gaillard, J.-M. (2020). Sex differences in adult lifespan and aging rates of mortality across wild mammals. Proceedings of the National Academy of Sciences, 117(15), 8546–8553. https://doi.org/10.1073/pnas.1911999117
Li, N., & Gerland, P. (2012). Model life tables. Retrieved from https://www.un.org/development/desa/pd/data/model-life-tables
Marlowe, F. (2004). Marital residence among foragers. Current Anthropology, 45(2), 277–284.
Marwick, B., Johnson, A., White, D., & Eff, E. A. (2016). Binford: Binford’s hunter-gatherer data. Retrieved from https://CRAN.R-project.org/package=binford
Monaghan, P., & Ivimey‐Cook, E. R. (2023). No time to die: Evolution of a post‐reproductive life stage. Journal of Zoology, 321(1), 1–21. https://doi.org/10.1111/jzo.13096
Moravec, J. C., Atkinson, Q., Bowern, C., Greenhill, S. J., Jordan, F. M., Ross, R. M., … Cox, M. P. (2018). Post-marital residence patterns show lineage-specific evolution. Evolution and Human Behavior, 39(6), 594–601. https://doi.org/10.1016/j.evolhumbehav.2018.06.002
Peccei, J. S. (2001). A critique of the grandmother hypotheses: Old and new. American Journal of Human Biology, 13(4), 434–452.
Pontzer, H., Raichlen, D. A., Wood, B. M., Emery Thompson, M., Racette, S. B., Mabulla, A. Z. P., & Marlowe, F. W. (2015). Energy expenditure and activity among Hadza hunter-gatherers. American Journal of Human Biology, 27(5), 628–637. https://doi.org/10.1002/ajhb.22711
Pontzer, H., Yamada, Y., Sagayama, H., Ainslie, P. N., Andersen, L. F., Anderson, L. J., … Speakman, J. R. (2021). Daily energy expenditure through the human life course. Science, 373(6556), 808–812. https://doi.org/10.1126/science.abe5017
Pretelli, I., Crittenden, A. N., Dounias, E., Friant, S., Koster, J., Kramer, K. L., … Lew‐Levy, S. (2024). Child and adolescent foraging: New directions in evolutionary research. Evolutionary Anthropology: Issues, News, and Reviews, 33(2). https://doi.org/10.1002/evan.22020
Pretelli, I., Ringen, E., & Lew-Levy, S. (2022). Foraging complexity and the evolution of childhood. Science Advances, 8(41). https://doi.org/10.1126/sciadv.abn9889
Rendell, L., Cantor, M., Gero, S., Whitehead, H., & Mann, J. (2019). Causes and consequences of female centrality in cetacean societies. Philosophical Transactions of the Royal Society B: Biological Sciences, 374(1780), 20180066. https://doi.org/10.1098/rstb.2018.0066
Ringen, E. J., Duda, P., & Jaeggi, A. V. (2019). The evolution of daily food sharing: A Bayesian phylogenetic analysis. Evolution and Human Behavior, 40(4), 375–384.
Scalise Sugiyama, M. (2011). The forager oral tradition and the evolution of prolonged juvenility. Frontiers in Psychology, 2. https://doi.org/10.3389/fpsyg.2011.00133
Schniter, E., Gurven, M., Kaplan, H. S., Wilcox, N. T., & Hooper, P. L. (2015). Skill ontogeny among Tsimane forager-horticulturalists. American Journal of Physical Anthropology, 158(1), 3–18. https://doi.org/10.1002/ajpa.22757
Sievert, L. L. (2024). Evolutionary perspectives, comparative approaches, and the lived experience of menopause. American Journal of Biological Anthropology, 186(S78). https://doi.org/10.1002/ajpa.25067
Smit, N., & Robbins, M. M. (2025). Post-reproductive lifespan in wild mountain gorillas. Proceedings of the National Academy of Sciences, 122(42), e2510998122.
Smith, D., Dyble, M., Major, K., Page, A. E., Chaudhary, N., Salali, G. D., … Mace, R. (2019). A friend in need is a friend indeed: Need-based sharing, rather than cooperative assortment, predicts experimental resource transfers among Agta hunter-gatherers. Evolution and Human Behavior, 40(1), 82–89. https://doi.org/10.1016/j.evolhumbehav.2018.08.004
Smith, J. E., Fichtel, C., Holmes, R. K., Kappeler, P. M., Vugt, M. van, & Jaeggi, A. V. (2022). Sex bias in intergroup conflict and collective movements among social mammals: Male warriors and female guides. Philosophical Transactions of the Royal Society B, 377(1851). https://doi.org/10.1098/rstb.2021.0142
Urlacher, S. S., Snodgrass, J. J., Dugas, L. R., Sugiyama, L. S., Liebert, M. A., Joyce, C. J., & Pontzer, H. (2019). Constraint and trade-offs regulate energy expenditure during childhood. Science Advances, 5(12), eaax1065.
Venkataraman, V. V., Hagen, E. H., Stibbard-Hawkes, D., Hames, R., Lew-Levy, S., Jang, H., et al.others. (2026). The meanings and dividends of man the hunter. Evolution and Human Behavior, 47(3), 106840.
Walker, R. S., Hill, K. R., Flinn, M. V., & Ellsworth, R. M. (2011). Evolutionary history of hunter-gatherer marriage practices. PLoS ONE, 6(4), e19066. https://doi.org/10.1371/journal.pone.0019066
Walker, R., Hill, K., Burger, O., & Hurtado, A. M. (2005). Life in the slow lane revisited: Ontogenetic separation between chimpanzees and humans. American Journal of Physical Anthropology, 129(4), 577–583. https://doi.org/10.1002/ajpa.20306
Walker, R., Hill, K., Kaplan, H., & McMillan, G. (2002). Age-dependency in hunting ability among the ache of eastern paraguay. Journal of Human Evolution, 42(6), 639–657. https://doi.org/10.1006/jhev.2001.0541
White, L. C., Lukas, D., Langergraber, K. E., & Vigilant, L. (2025). Avoidance of reproductive conflict and the evolution of menopause in chimpanzees. Royal Society Open Science, 12(6). https://doi.org/10.1098/rsos.250385
Wiessner, P. W. (2014). Embers of society: Firelight talk among the Ju/’hoansi Bushmen. Proceedings of the National Academy of Sciences, 111(39), 14027–14035.
Winkler, I., & Goncalves, A. (2023). Do mammals have menopause? Cell, 186(22), 4729–4733.
Wood, B. M., Negrey, J. D., Brown, J. L., Deschner, T., Thompson, M. E., Gunter, S., … Langergraber, K. E. (2023). Demographic and hormonal evidence for menopause in wild chimpanzees. Science, 382(6669), eadd5473.
Wood, B. M., Raichlen, D. A., Pontzer, H., Harris, J. A., Sayre, M. K., Paolo, B., … Mabulla, A. Z. P. (2024). Beyond the here and now: Hunter-gatherer socio-spatial complexity and the evolution of language. Philosophical Transactions of the Royal Society B: Biological Sciences, 379(1912). https://doi.org/10.1098/rstb.2022.0521
8 Supplementary information
8.1 Hunter-gatherer demography and life history
8.2 Effects of variation in skill ontogeny parameters on energy balance
8.3 Effects of variation in ALB and IBI on family composition and energy consumption
8.4 Energy balance distributions
8.5 Dependents and surviving parents
Footnotes
As an aside, the not infrequent death of mothers with multiple dependent children could have helped select for female fidelity to ensure surviving fathers would continue to invest in their joint offspring.↩︎