Psychology by BogdanovaThe Encyclopedia of Psychology for the Artificial Era

RESEARCH DOSSIER

Gene–Culture Coevolution and Human Niche Construction

Abstract. Gene–culture coevolution studies reciprocal relations between genetic inheritance and socially transmitted practices, while human niche construction investigates how organized activities modify ecological conditions encountered by later generations. This fundamental critical narrative research monograph examines mathematical dual-inheritance theory, culturally mediated selection, genetic drift and migration, developmental and ecological inheritance, social-learning experiments, cumulative culture, archaeology, population genomics, and ancient DNA. Extended case analyses evaluate lactase persistence and dairying, convergent African adaptations, the revised famine–pathogen hypothesis, structural variation at the amylase locus, and multispecies domestication. The review compares explanatory frameworks, reconstructs the history of major theories, separates causal identification from association, and assesses evidentiary limits across laboratory, historical and computational methods. The scientific synthesis places cultural transmission, niche modification and population-genetic processes within a testable explanatory sequence and identifies opportunities for future research on constructed cognitive environments in the Artificial Era. This manuscript is a critical narrative synthesis of existing published literature and reports no original experiment or formal systematic-review protocol.

Keywords: gene–culture coevolution; human niche construction; dual inheritance theory; cultural evolution; ecological inheritance; social learning; cumulative culture; population genetics; lactase persistence; dairying; amylase copy number; ancient DNA; cultural transmission bias; archaeology; gene–environment interplay; cultural demography; Artificial Era

The Scientific Problem and Its Units of Analysis

Gene–culture coevolution concerns reciprocal evolutionary relations between genetically inherited variation and socially transmitted practices. Human niche construction concerns the ways organismal activities, particularly culturally organized activities, alter developmental conditions and ecological selection pressures. The central empirical problem is to establish which cultural practices change environments, which biological traits affect the acquisition or consequences of those practices, and whether their interaction changes genetic frequencies or phenotypic distributions over generations. This definition makes a causal sequence testable: cultural inheritance can modify exposure, exposure can alter survival and reproduction, and genetically influenced dispositions can change the likelihood of adopting, maintaining, or transmitting the culture Feldman & Cavalli-Sforza (1976) Laland et al. (2000) Laland et al. (2010).

The appropriate observational unit depends on the claim. An individual provides measurements of learning, choice, phenotype, genotype, and developmental experience; a community provides transmission networks and institutions; a population provides frequencies and demographic histories; archaeological deposits register ancient practices; genomes retain traces of ancestry, selection, migration, and drift. Evidence established at one level cannot automatically supply a mechanism at another. An association between present-day diet and an allele, for example, suggests a historical question but cannot alone identify the date, selective agent, or cultural transmission pathway that generated the association Creanza et al. (2017) Deffner et al. (2024).

This critical narrative synthesis evaluates mathematical models, experiments in social learning, archaeological material, historical and ancient DNA, contemporary population genetics, and comparative evidence. Its questions concern the origin of the framework, the psychological mechanisms that transmit culture, the conditions that produce evolutionary feedback, the relative roles of selection and neutral processes, and the limits of causal inference. The literature is arranged by explanatory problem rather than by a claim that a single general theory accounts for every human cultural difference. The resulting account distinguishes direct measurement, inference from combined data, and theoretical possibility Fogarty et al. (2025) Laland et al. (2010).

Definitions of Culture, Genetic Inheritance, and Constructed Niches

Culture here denotes information and practices acquired or modified through social learning and capable of transmission between individuals or generations. Its operational indicators include observed copying, teaching, communication, conventional tool use, patterned food preparation, institutionally reinforced norms, and the spatial distribution of practices. Cultural persistence does not require an invariant symbolic representation: learners routinely reconstruct, combine, and modify behaviors. Models therefore distinguish a cultural trait's prevalence from the fidelity and pathways of its transmission, and they distinguish the functions a practice performs from the particular variant used to perform them Henrich & McElreath (2003) Mesoudi et al. (2006) Creanza et al. (2017).

Genetic inheritance involves DNA variation passed through reproductive lineages. An allele's frequency can change through selection, genetic drift, mutation, and migration; structured mating and demographic history can produce strong geographical associations without a cultural practice directly selecting that allele. Gene–culture association describes a statistical relationship between genetic and cultural states; gene–culture coevolution is a dynamic explanation of how their transmission and evolutionary changes interact. A strong causal claim identifies model assumptions and data that support reciprocal effects, rather than equating simultaneous genetic and cultural differentiation with demonstrated coevolution Feldman & Cavalli-Sforza (1976) Fogarty et al. (2025).

Niche construction comprises activities that modify environmental features consequential for organisms and their descendants. Humans construct agricultural landscapes, dwellings, roads, storage systems, irrigation works, disease environments, and institutional routines, often in conjunction with other species. Such modifications may persist after their original builders have died and may therefore constitute ecological inheritance. A modified niche is an external arrangement of opportunities, hazards, and constraints; inheriting it is analytically distinct from inheriting the genetic or cultural traits that produced it. Culturally built environments can nevertheless redirect the fitness consequences of both forms of inheritance Laland et al. (2000) Laland et al. (2001).

Intellectual History of Dual Inheritance and Cultural Evolution

Population-genetic approaches to cultural transmission developed from dissatisfaction with treating socially learned phenotypes as if their inheritance were entirely Mendelian. The mathematical research program of Cavalli-Sforza and Feldman formalized evolutionary consequences of complex transmission; their 1976 treatment of genetic and learned phenotypes demonstrated that acquisition and selection could be considered jointly. Boyd and Richerson subsequently developed an extensive theory of cultural transmission, biased learning, and population-level cultural evolution. Their 1985 volume Culture and the Evolutionary Process supplied a foundation for later models of conformity, prestige, and interactions between learning and selection Feldman & Cavalli-Sforza (1976) Boyd & Richerson (1985).

During the following decades the framework became connected to evolutionary psychology, behavioral ecology, archaeology, and comparative cognition. Henrich and McElreath articulated how evolved capacities for social learning and population processes could jointly account for the diversity of human behavioral adaptations. Richerson and Boyd described cultural inheritance as a major contributor to human evolutionary history, while Mesoudi, Whiten, and Laland argued for disciplinary integration across experiments, ethnography, historical observations, and evolutionary modeling. These contributions established a methodology in which cultural variants have transmission histories and population dynamics that can be empirically reconstructed Henrich & McElreath (2003) Richerson & Boyd (2005) Mesoudi et al. (2006).

Niche-construction research introduced a complementary emphasis on environmental modification. The framework developed by Laland, Odling-Smee, and Feldman examined how environmental changes made by organisms influence the selective conditions encountered by later generations. Their 2000 conceptual synthesis and 2001 mathematical analysis placed culturally generated environmental feedback within evolutionary models. The subsequent interdisciplinary literature connected this claim to genetic variation, domestication, and dietary change. The resulting history is a convergence of distinct research programs—cultural transmission theory, environmental modification theory, and population genomics—rather than one homogeneous school Laland et al. (2000) Laland et al. (2001) Laland et al. (2010).

The Psychological Machinery of Social Learning

Social learning spans observation, attention to others, emulation of outcomes, imitation of actions, verbal instruction, testimony, and apprenticeship. Whether a transmitted solution persists depends on how learners encode information, infer intentions, evaluate demonstrators, and integrate experience with locally available rewards. Models of cultural inheritance often idealize these mechanisms as transmission probabilities. Psychologically realistic testing requires task designs that separate copying from independent discovery and distinguish accurate reproduction from convergence on the same functional solution. A cultural practice can spread because it is repeatedly reinvented, widely demonstrated, actively taught, or institutionally compelled; these possibilities leave different empirical signatures Henrich & McElreath (2003) Dean et al. (2012).

Social information is not necessarily copied at random. Learners may attend to skilled or prestigious demonstrators, frequent local behavior, kin, or members of trusted groups. Payoff bias and frequency-dependent bias can generate different population trajectories, yet an observed majority choice is insufficient evidence that a conformist bias caused it. Similar outcomes can follow from common incentives, shared information, experimenter cues, or individual adaptation. The theoretical value of transmission biases lies in articulating discriminating predictions about changes in the distribution of variants under specified learning environments Henrich & Boyd (2001) Deffner et al. (2024).

Rendell and colleagues' 2010 computational tournament is informative because it compared explicit strategies in a simulated variable environment. Social-learning-heavy strategies performed unexpectedly well in that setting partly because individuals tended to demonstrate rewarding behavior, thereby filtering information for observers. The result describes performance under the tournament's rules, not an invariant human preference for imitation. Dean and colleagues' 2012 puzzle-box comparison identified associations between children's success and teaching, imitation, and prosocial behavior absent from the same task pattern in the tested chimpanzee and capuchin groups. These studies provide complementary computational and comparative evidence for mechanisms that can sustain cumulative transmission Rendell et al. (2010) Dean et al. (2012).

Cultural Accumulation, Innovation, and Demographic Conditions

Cumulative cultural evolution occurs when socially transmitted modifications are preserved and elaborated across learning episodes, allowing later learners to employ solutions they could not readily invent individually. The minimal ingredients include a supply of variation, means of retaining improvements, opportunities for recombination, and sufficiently reliable transmission. The relevant cognitive capacities vary across tasks: language can communicate design reasons, imitation can preserve motor sequences, teaching can reduce exploration costs, and cooperative activity can distribute complementary skills. Claims about a universal threshold of cumulative culture require explicit performance metrics and comparable testing environments Dean et al. (2012) Richerson & Boyd (2005).

Henrich's 2004 model associated the maintenance of complex technologies with the effective pool of available social learners and the distributions of skill and copying error. Its application to Tasmanian archaeological history is an influential explanatory hypothesis rather than an experimental reconstruction of the actual loss of every proposed technology. The model illustrates a general principle: cultural complexity can change when population connectivity and opportunities to learn change, even if individual cognitive capacities remain constant. Population size, network reach, specialist availability, teaching institutions, and ecological necessity may have different effects and should be operationalized separately Henrich (2004).

The ability of culture to accumulate introduces a historical dependence that can amplify niche construction. A community can inherit reservoirs, field systems, husbandry knowledge, manufacturing routines, and food-processing methods that were constructed by earlier generations. These structures change which innovations are feasible and what counts as competence for a newcomer. Whether such accumulation promotes adaptation depends on ecological conditions and can be maladaptive when inherited practices become costly. Historical durability therefore establishes a candidate cultural mechanism; it does not by itself demonstrate a genetic response. The latter requires additional evidence concerning the reproductive consequences of the constructed environment Laland et al. (2001) Creanza et al. (2017).

Population-Genetic Models of Coupled Cultural and Biological Inheritance

A useful model of gene–culture coevolution specifies genetic states, cultural states, their joint frequencies, and the processes that map one generation into the next. Genetic transmission follows reproductive lineages; cultural transmission can be vertical, oblique, or horizontal. The two pathways can produce associations even when a cultural trait is acquired late in life or crosses family boundaries. Researchers track the conditional probability that an individual acquires a cultural behavior given parental, peer, demonstrator, or institutional information, and the expected reproductive contribution of individuals with different genotype–culture combinations Feldman & Cavalli-Sforza (1976) Boyd & Richerson (1985).

A simplified conceptual recurrence can be expressed as p(t+1) = T[p(t), q(t), E(t), D(t)] for the joint distribution p of genetic and cultural states, where q represents cultural transmission and E the niche state, while D represents demographic exchange. The equation is schematic, not an estimated evolutionary law. Its purpose is to show where a hypothesis must declare its assumptions. In particular, a cultural frequency change may precede an allele-frequency change, follow one, be driven by the same ecology, or occur independently; these sequences are not observationally interchangeable Creanza et al. (2017) Fogarty et al. (2025).

The mathematical literature distinguishes selection on genetic types, selection on cultural types, and associations generated by transmission itself. Fogarty, Zhang, and Feldman (2025) demonstrated model conditions under which gene–culture association persists, polymorphic equilibria arise, and coevolution occurs without direct selection on the cultural trait. These are properties of specified theoretical systems. Their value for empirical research lies in preventing a verbal model from assuming that any cultural adaptation must inevitably favor a particular allele. An estimated interaction should be interpreted only after alternative demographic and transmission pathways have been represented Fogarty et al. (2025).

Models are evaluated by recoverable parameter predictions: for example, whether observed ancestry-adjusted allele trajectories, reconstructed dairy practices, and archaeological chronology are jointly likely under competing selection regimes. A model capable of generating a historical pattern is not thereby uniquely confirmed, since distinct histories may yield similar contemporary observations. Model identification requires informative data across time and space, diagnostics for sensitivity to assumptions, and, where possible, independent tests that were not used to choose the original model Itan et al. (2009) Deffner et al. (2024).

Cultural Selection Biases and Evolutionary Dynamics

Cultural transmission can favor behaviors because learners expect a high payoff, see them used by many others, or regard demonstrators as reliable. Each mechanism may accelerate or delay the spread of a trait. Frequency-dependent transmission may create multiple stable traditions under similar material conditions, while payoff-sensitive transmission may produce rapid change after an environmental shock. As a result, cultural frequencies need not track genetic frequencies or immediate fitness consequences. An explanatory analysis should reconstruct what individuals actually observed, which choices were available, and how the cost of a mistaken imitation varied across social positions Boyd & Richerson (1985) Richerson & Boyd (2005).

Henrich and Boyd's 2001 model showed that limited conformist transmission can stabilize norms of costly punishment in cooperative dilemmas under particular assumptions about multistage enforcement. The theoretical implication is that institutions may generate persistent selective environments for cooperative behavior. It would be incorrect to infer from the model alone that observed punitive norms are genetically caused or that real societies occupy the modeled equilibrium. Institutions themselves have changing political, economic, and interpersonal histories; these can be measured through longitudinal comparative studies and experimental analysis of social learning Henrich & Boyd (2001).

A further distinction separates cultural success from biological fitness. A practice may spread because it is entertaining, memorable, prestigious, mandated, or immediately profitable while imposing long-term costs on some participants. Cultural innovations can also modify fertility, partnership systems, disease exposure, and age-specific mortality with effects that differ across genetic backgrounds. Linking these effects to evolutionary change requires time-indexed demographic information, not only contemporary endorsement of a norm. Mathematical treatments can represent complex selection and transmission without presuming that every institution benefits its practitioners Feldman & Cavalli-Sforza (1976) Fogarty et al. (2025).

Cultural selection is particularly susceptible to measurement error because a reported belief, a visible act, and sustained practical competence are different phenotypes. Laboratory copying tasks measure behavioral choices in highly constrained conditions; ethnographic observation can reveal apprenticeship, sanction, and symbolic significance; digital traces offer frequency information while introducing platform selection. When these measures disagree, the discrepancy is informative about the distinction between cultural representation and cultural performance. A sound coevolutionary model specifies which of these dimensions is transmitted and which is expected to change selective exposure Deffner et al. (2024) Creanza et al. (2017).

Ecological Inheritance and Feedback Through Built Environments

A constructed niche can persist as material infrastructure, a managed species assemblage, a modified soil or water regime, or a patterned social environment. When later generations encounter such changes, their fitness landscape partly reflects prior activity. This creates a feedback sequence in which organismal behavior changes the environment, the changed environment alters selection, and later behavior or genotype frequencies influence subsequent modifications. In humans, shared tools, cooperative labor, and records extend that feedback across individuals and timescales. The ecological inheritance is the inherited state of the world, rather than a third mechanism of DNA transmission Laland et al. (2000) Laland et al. (2001).

Niche construction is an explanatory hypothesis only when the relevant environmental modification can be specified. A field changes plant communities and pathogen exposure; a granary changes seasonal risk and storage incentives; an animal enclosure changes contact with domesticated species; settlement crowding may alter infection hazards. Each pathway has different measurable mediators and time lags. A generic statement that humans transform environments is insufficient to establish how a particular genetic variant spread. The decisive step is to identify differential survival, fertility, or population movement plausibly caused by the transformed exposure Laland et al. (2010) Evershed et al. (2022).

Cultural niche construction can also influence other species, making the evolutionary system multispecies. Herding and domestication modify selective conditions on animals while changing human diets, work allocation, cooperation, and ecological contact. Beja-Pereira and colleagues identified correlations among geographic patterns of cattle milk-protein gene diversity, early livestock contexts, and modern human lactose tolerance. Those associations motivated a historical coevolutionary interpretation, but independent reconstructions are required to separate shared population history, deliberate selection, and environmental effects. The animals and the people form connected, measurable evolutionary populations rather than one undifferentiated evolutionary subject Beja-Pereira et al. (2003) Van Amburg et al. (2026).

A new mathematical treatment by Van Amburg, Babel, and Feldman (2026) explicitly connected culturally transmitted human preferences with selection on the genetics of domesticates. The model can maintain genetic and cultural polymorphisms and can yield cyclic dynamics under specified conditions. It establishes the possibility of feedback between human cultural traits and another species' genetic composition. It is a theoretical contribution, not a direct demonstration that any particular historical livestock population followed a predicted cycle. Empirical tests would need time-resolved animal genotypes, changes in human preference, and controls for trade, ecology, and husbandry Van Amburg et al. (2026).

Evolutionary Feedback Versus Common Ecological Causes

Cultural practice and genetic variation may covary because both respond to the same ecological gradient. Temperature, altitude, pathogen load, food availability, and population ancestry can structure behavior and allele frequencies simultaneously. In that case, a correlation is consistent with coevolution yet also consistent with independent responses to environment. Distinguishing the hypotheses requires triangulation among chronology, natural history, ancestry modeling, and mechanistic measurements. Cross-sectional population differences are especially vulnerable to confounding by migration, intermarriage, language family, and sampling schemes Laland et al. (2010) Henrich et al. (2010).

The temporal order of events matters. If a culturally constructed resource preceded a genetically associated response, that ordering supports a necessary component of a feedback hypothesis. It does not prove that the resource was the selective agent: the same settlement transition may change infection, crowding, trade, and nutrition at once. Ancient DNA offers direct observations of past genetic composition, while lipid residues and artifacts provide partially independent evidence of practice. Their combination can reject simple narratives that could not be evaluated from present-day populations alone Evershed et al. (2022) Gerbault et al. (2011).

Causal graphs are useful because they force researchers to distinguish mediators from confounders. For example, an agricultural transition may affect starch consumption, population density, disease burden, migration, and fertility, while ancestry influences the distribution of genetic variants and subsistence behavior. Conditioning on a variable that is a mediator or collider can distort estimated effects. Deffner and colleagues' 2024 framework connects explicit generative assumptions, directed acyclic graphs, simulations, and statistical inference. It offers a practical route from an evolutionary narrative to falsifiable data requirements Deffner et al. (2024).

The most compelling coevolutionary claims therefore involve converging observations with different biases: a molecular mechanism, a historical cultural transition, a genotype trajectory, and a defensible demographic model. Weaker cases may support a cultural influence on population structure or a cultural–genetic association without isolating reciprocal selection. Keeping distinct levels of evidentiary status increases explanatory precision. It also helps identify which new archaeological sites, ancient samples, longitudinal observations, or experimental manipulations would materially alter the conclusions Fogarty et al. (2025) Laland et al. (2010).

Adaptation, Drift, Migration, and the Scope of Coevolution

The familiar narrow conception of gene–culture coevolution emphasizes cultural practices that create selection on genetic variants. A broader conception analyzes how culture shapes the distribution of genetic variation through migration, drift, mating networks, and social demography as well as selection. These processes may generate important historical feedback even when a measured allele confers no adaptive advantage. Fogarty and colleagues' 2025 discussion of the broad framework illustrates why genetic differentiation associated with cultural institutions cannot always be reduced to an adaptive narrative Kasser et al. (2025).

Migration can transport cultural practices and alleles together or separately. A farming tradition may diffuse through the movement of people, the transfer of knowledge, or a combination; an apparently rapid cultural expansion need not indicate an equally rapid genetic expansion. Endogamy, residence patterns, preferential marriage, and exchange networks alter genetic connectivity and the observation of inherited cultural traits. Any claim about selection has to be tested against realistic demographic histories and spatially heterogeneous sampling Itan et al. (2009) Kasser et al. (2025).

Genetic drift becomes particularly important in small, isolated, or repeatedly founded populations. Cultural boundary maintenance may reduce gene flow and permit drift to amplify differences even without a selected phenotype. Conversely, a rapidly spreading tradition can homogenize behavior across genetically diverse groups. Cultural similarity therefore gives no automatic estimate of genetic relatedness, and allele-frequency difference is not an index of psychological or moral difference. The appropriate language describes the histories of particular variants in particular populations Kasser et al. (2025) Henrich et al. (2010).

A coherent synthesis includes adaptive, neutral, and demographic processes within a shared modeling framework, while keeping their evidentiary signatures distinct. Selection implies differences in reproductive success tied to inherited variants under identifiable conditions; migration produces ancestry change; drift predicts stochastic fluctuations; cultural learning changes behavior across genetic backgrounds. Since these processes often interact, a single-signature explanation is rarely adequate. Comparative inference improves when investigators report model uncertainty, plausible alternative histories, and the empirical findings that would discriminate among them Creanza et al. (2017) Fogarty et al. (2025).

Dairying, Lactase Persistence, and the Logic of a Historical Test

Adult lactase persistence is a powerful case for gene–culture research because the phenotype, regulatory genetics, archaeology of dairying, and ancient population history can be investigated independently. Most mammals reduce intestinal lactase activity after weaning. Some human populations maintain substantial lactase expression into adulthood, enabling digestion of the lactose in fresh milk. This condition is shaped by regulatory variation near LCT, with distinct variants in different regions. Herd management, milk storage, and processing were transmitted cultural practices, so a plausible historical sequence connects changes in the food environment to variation in the selective value of lactase expression Tishkoff et al. (2007) Gerbault et al. (2011).

The research problem requires distinguishing milk consumption from dairy production and lactose exposure. Butter, cheese, fermented milk, and other products can contain less lactose than fresh milk; archaeological evidence of dairying therefore does not identify the amount of lactose ingested by particular individuals. The spatial spread of domesticates, evidence of pottery residues, and human genetic variation also operate on different time scales. A simple correlation between present-day pastoralism and lactase-persistence frequency compresses these distinctions and risks overstating knowledge about prehistoric behavior Gerbault et al. (2011) Evershed et al. (2022).

Bersaglieri and colleagues identified genomic signatures consistent with strong recent positive selection near the European lactase region, while Itan and colleagues modeled the spread of lactase persistence against demographic and cultural histories of early European farming. These are complementary lines of inference: molecular patterns constrain evolutionary hypotheses, and demographic simulations explore which historical pathways can produce those patterns. Their joint contribution is evidence for substantial gene–culture interaction, while specific estimates of selection and geographic origin depend on model assumptions, ancient sampling, and the chronology used for cultural traits Bersaglieri et al. (2004) Itan et al. (2009).

The example also illustrates why reciprocal coevolution is more demanding than one-way cultural selection on genes. A full feedback claim would establish that dairying changed the distribution of lactase phenotypes and that changes in the distribution of those phenotypes subsequently altered the spread, organization, or payoff of dairying. Demonstrating the first direction is not automatically evidence for the second. Historical research gains precision by distinguishing an inferred selective response from the stronger claim of reciprocal change between both inheritance systems Feldman & Cavalli-Sforza (1976) Kasser et al. (2025).

Convergent Lactase-Persistence Variants in African Populations

African populations provide critical evidence that lactase persistence is neither a single-origin European trait nor a simple marker of one cultural identity. Tishkoff and colleagues documented multiple genetic variants associated with adult lactase persistence and with relatively recent signatures of selection in populations with pastoralist histories. Their comparison of genotype, phenotype, and regional ancestry supplied evidence for convergent adaptation: different genetic paths can be favored under partially similar cultural dietary conditions. The study's sample and geographic coverage still delimit its generalizability Tishkoff et al. (2007).

The African evidence sharpens several methodological distinctions. A causal interpretation must consider whether particular regulatory variants alter lactase expression, whether phenotypic tests capture the intended physiological property, and whether allele distributions can be separated from recent demographic mixture. Genetic association within one community may behave differently in another community with distinct ancestry and lactase regulatory variants. Cultural categories such as pastoralist or nonpastoralist describe heterogeneous subsistence and residence patterns, not uniform lifetime milk consumption Tishkoff et al. (2007) Gerbault et al. (2011).

Convergence also has a theoretical implication. The culturally mediated niche may create a recurring opportunity for selection without predetermining which nucleotide change supplies the biological response. Mutation supply, prior ancestry, demographic history, and drift shape the variants available to selection. This relationship is more informative than a deterministic narrative in which one practice generates one gene. A theory of gene–culture coevolution should predict classes of phenotypic response under exposure, while allowing multiple biological solutions and historical contingencies Laland et al. (2010) Kasser et al. (2025).

Ethnographic and dietary inference should be integrated without reducing communities to their genetic markers. Interviews about milk use, direct dietary observation, ecological records, and biophysical measurements can make the cultural exposure more specific. Genetic results warrant careful interpretation because contemporary identity categories do not map cleanly onto ancient populations or lineages. The strongest evidence describes the tested physiological phenotype and historical process at its actual sampling scale, while recording the uncertainty associated with extrapolation across regions or periods Henrich et al. (2010) Tishkoff et al. (2007).

Ancient Dairy Lipids and a Revised Selection Narrative

A major reassessment became possible when Evershed and colleagues integrated archaeological milk-fat residues from prehistoric European pottery, ancient DNA trajectories of lactase-persistence alleles, and demographic and epidemiological models. Their 2022 analysis showed that the appearance and abundance of dairy-use signals did not simply track the later rise in European lactase persistence. The study tested the hypothesis that drinking milk itself was sufficient to explain the selection pattern, finding stronger support in the modeled data for selective pressures associated with famine and pathogen exposure Evershed et al. (2022).

The substantive implication is a more conditional coevolutionary mechanism. Fresh milk may have supplied valuable calories to some groups at some times; lactose malabsorption may have imposed greater costs under malnutrition or intestinal infection; settlement and pastoral economies may also have modified pathogen ecology and social contact. Under this account, the constructed dairy niche is embedded in a much larger transition in livelihood and disease environment. The genotype's fitness consequences are conditional on broader ecological and physiological conditions rather than on an average milk-consumption variable alone Evershed et al. (2022).

There are also constraints on what the reconstructed record measures. Lipids preserved in surviving pottery are a proxy for vessel use, which depends on food preparation, preservation, and archaeological recovery. Ancient DNA frequencies depend on which regions, periods, and individuals are sampled. Model-based inferences about famine and pathogen exposure can be compared to material and skeletal proxies, but those processes were not directly randomized across ancient European communities. The study therefore revises the plausibility of competing explanations rather than producing a laboratory proof of one ancient selective event Evershed et al. (2022) Gerbault et al. (2011).

The revised account demonstrates scientific progress through evidence triangulation. The earlier coevolutionary hypothesis encouraged the collection of genetic and archaeological data; those data subsequently constrained an overly simple selection narrative. The broader explanatory structure survives as an empirical research program precisely because it permits changes in mechanism when new data arrive. An adequate monograph should state separately that dairying preceded widespread European adult lactase persistence, that variants underwent selection, and that the particular environmental triggers remain a subject of reconstruction Bersaglieri et al. (2004) Evershed et al. (2022).

Starch Diets and the Structural Evolution of the Amylase Locus

Salivary amylase begins starch digestion, making the amylase gene family a second prominent site of gene–culture hypothesis testing. Perry and colleagues' 2007 study reported an association between AMY1 copy number and dietary histories involving starch, with higher copy number in the compared populations historically consuming more starch. The work connected a physiological mechanism to culturally patterned subsistence, but the genetic region is structurally complex. Correlations between coarse dietary labels and copy-number estimates are vulnerable to ancestry, sampling, and limitations of genomic assays Perry et al. (2007).

More recent long-read sequencing and analyses of ancient genomes have transformed understanding of this region. Bolognini and colleagues' 2024 Nature study resolved extensive structural diversity at the amylase locus, identified recurrent rearrangements, and reconstructed changes in the distribution of amylase-related gene structures associated with agricultural histories. Its use of long-read assemblies and ancient genetic material gives stronger leverage on structural haplotypes than earlier copy-number summaries. It also highlights that gene-count estimates can collapse different underlying genomic architectures Bolognini et al. (2024).

The psychological cultural component of the amylase case concerns food knowledge and socially organized processing: plant domestication, cultivation, cooking, grinding, storage, division of labor, and preferences for starch-rich foods. Such behaviors alter the average nutritional environment without ensuring that one dietary category precisely measures an individual's starch exposure. Biological interpretation requires specifying whether additional gene copies alter enzyme expression or performance under relevant dietary conditions and how any advantage translates into differential reproduction in historical ecologies Perry et al. (2007) Bolognini et al. (2024).

The amylase literature illustrates a general research strategy. First identify a robust genetic phenotype; next describe how a transmitted subsistence practice changes exposure; then estimate the chronology and selection mechanism while controlling for demography. Concordance across ancient and contemporary data strengthens a hypothesis, but alternative selection mechanisms should remain explicit. Cultural evolution changes the environment in ways that may make a genetic trait advantageous, while genomic evidence specifies whether a corresponding inherited response actually occurred Bolognini et al. (2024) Laland et al. (2010).

Domestication and the Multispecies Coevolutionary System

Domestication extends feedback beyond a single species. Human preferences regarding yield, temperament, appearance, or work performance can alter which animals reproduce, while genetically changing animals alter the resources, risks, and daily practices available to people. Such feedback involves inherited variation in domesticates, culturally transmitted human preferences, and shared environments. Early cattle–human gene–culture analyses drew attention to patterns in milk-protein genes and lactase-persistence geography, but their conclusions depend on reconstructing multiple population histories, rather than treating correlation as a complete causal account Beja-Pereira et al. (2003).

Van Amburg, Babel, and Feldman recently developed formal models of coevolution between culturally transmitted human preferences and genetic variation in domesticated populations. Their 2026 analysis shows how dynamics depend on relative strengths of social transmission and selection and can include polymorphism and cycling under certain parameter values. These predictions specify mechanisms that historical and contemporary datasets could test. They do not imply that domestication universally causes oscillatory evolutionary change, and they do not replace direct measurements of breeding decisions and animal genetic trajectories Van Amburg et al. (2026).

A psychologically informed model of domestication can specify how preferences are acquired, who controls reproduction, how knowledge about animal traits is shared, and which institutions maintain selective practices. Herd ownership, cooperation, resource scarcity, and prestige can affect both the social distribution of preferences and the selection imposed on animals. The same animal phenotype may be valued differently in contrasting agricultural and economic settings. Cultural preferences must therefore be represented as historically changing, situated variables rather than constant human goals Van Amburg et al. (2026) Laland et al. (2001).

The multispecies case establishes the analytical breadth of human niche construction. Human behavior can change selective conditions for other organisms, and their responses can modify human social life and biology in turn. Strong evidence requires records of breeding practice, genetic variation in animal and human populations, and changing ecology. Theoretical models provide the explanatory map; archaeology, genomics, and behavioral investigation determine which pathways occurred in the world Beja-Pereira et al. (2003) Van Amburg et al. (2026).

Archaeological Reconstruction and Cultural Chronology

Archaeology contributes evidence of environments constructed in the past: habitation patterns, storage facilities, cultivation traces, animal bones, tools, residues, and burial practices. Such material can sometimes reveal the emergence, geographic distribution, or disappearance of activities before documentary evidence exists. Its inferential unit is usually an assemblage deposited under particular conditions, and the social behavior that produced it must be reconstructed. The absence of an artifact does not necessarily indicate the absence of a practice, because preservation, excavation, and identification probabilities vary across contexts Mesoudi et al. (2006) Evershed et al. (2022).

A historical claim requires chronological alignment. Radiocarbon dating, stratigraphy, comparative artifact typology, and Bayesian archaeological models help estimate when practices were present. Genetic samples from different layers or regions provide a separate historical series. If practices first appeared in a region centuries before a selected allele became frequent, the lag can constrain selection models; however, migration and incomplete sampling can change the apparent order. Explicit uncertainty intervals should accompany temporal comparisons, and the geographic denominator must be specified rather than inferred from a single well-excavated site Itan et al. (2009) Evershed et al. (2022).

Organic residues are especially useful for studying food processing, but the presence of dairy lipids in pottery cannot establish individual milk intake, digestive responses, or reproductive outcomes. Animal husbandry can leave multiple evidence types, including demographic profiles of managed herds and morphometric change. Combining those indicators with residue analysis offers a more complete picture of subsistence, but the meaning of each proxy remains limited by archaeological context and preservation. The best reconstructions therefore describe chains of inference rather than presenting a single artifact as direct observation of an ancient psychological choice Evershed et al. (2022) Beja-Pereira et al. (2003).

Cultural histories can also be reconstructed through phylogenetic and spatial comparative methods applied to language families, technologies, and institutions. Such reconstructions require careful treatment of borrowing, convergence, contact, and missing observations. A similarity between living communities may have been inherited from a common source, independently evolved under similar constraints, or spread by recent interaction. Only a model that distinguishes these pathways can estimate transmission or ecological drivers. Genetic phylogeny and cultural phylogeny need not have matching branching structures Creanza et al. (2017) Kasser et al. (2025).

Population Genomics and the Evidence for Selection

Contemporary DNA can reveal unusual allele-frequency distributions, long haplotypes, population differentiation, and associations with physiological traits. These signatures become evidence for selection only in a specified demographic framework, because bottlenecks, admixture, and local ancestry can produce some similar patterns. Identifying a candidate genomic region is an intermediate step; functional molecular and phenotypic measures are needed to establish what the variant does. The lactase case benefits from multiple regulatory variants linked to directly measured adult lactose digestion Bersaglieri et al. (2004) Tishkoff et al. (2007).

Ancient DNA changes the historical inference by providing snapshots from past populations. Rather than extrapolating the past solely from living allele frequencies, researchers can estimate trajectories across dated samples. This design is still observational, and its conclusions depend on accurate dating, contamination control, ancestry correction, adequate sample coverage, and assumptions about population continuity. Archaeological human remains are unequally preserved and studied, so statistical uncertainty should include geographic and temporal gaps. A sample of ancient individuals does not automatically represent the entire population from which it came Evershed et al. (2022) Bolognini et al. (2024).

Structural variation requires specialized methods because duplications and rearrangements can elude standard single-variant association designs. The amylase locus is a strong example: different genomic architectures can generate similar summary copy numbers, and mutation rates differ from those of ordinary single-nucleotide substitutions. Long-read, haplotype-resolved sequences and pangenome-based analyses reveal structural configurations that were obscured by simpler assays. The phenotype–environment relationship should consequently be retested when the genetic measurement model changes Bolognini et al. (2024).

A persuasive result must link genotype, biological function, culture, and selection with appropriate adjustment for ancestry and ecological exposure. Many genomic findings establish only a subset of that chain. If a locus shows selection but the proposed cultural exposure remains poorly measured, the environmental mechanism is a plausible hypothesis; if a community exhibits a cultural tradition but no genetic trajectory is available, the case documents cultural inheritance without proving gene–culture coevolution. Publishing the complete evidentiary chain prevents selective reporting of its strongest link Kasser et al. (2025) Laland et al. (2010).

Behavioral Experiments, Transmission Chains, and Causal Identification

Experimental transmission chains permit investigators to observe how behavior changes when one participant learns from another and becomes the next demonstrator. Researchers can manipulate access to models, feedback, incentives, task difficulty, instruction, or opportunities for teaching. Outcomes include copying fidelity, cumulative performance, diversity, innovation rates, and differential success of variants. Such studies isolate components of cultural inheritance in a way that retrospective population studies cannot. Yet a laboratory chain lasting hours or days is a model of learning dynamics, not direct evidence about genomic change over millennia Dean et al. (2012) Rendell et al. (2010).

Experimental social-learning investigations must separate individual exploration from copying. If participants converge on the same answer because the task has a conspicuous optimum, apparent cultural diffusion may be repeated independent innovation. If an observed difference follows a prestige cue, learners may be responding to credibility rather than the demonstrator's actual performance. Studies can distinguish these mechanisms by randomizing information availability, demonstrator characteristics, and evidence about outcomes, while measuring participants' prior knowledge and beliefs. The resulting estimates are conditional on the sampled learners and task environment Henrich & McElreath (2003) Deffner et al. (2024).

Comparative studies help identify capacities supporting cultural accumulation by using tasks across species and age groups. Dean and colleagues compared performance and social behavior in a multistage puzzle-box task; the association of success with teaching, imitation, and prosocial behavior in children provides evidence about a particular mechanism under a particular experimental arrangement. Species differences can also reflect motivation, rearing history, motor constraints, visual affordances, or reward value. Comparable equipment does not guarantee psychologically comparable tasks, and a single negative result cannot establish species-wide incapacity Dean et al. (2012).

A causal bridge from learning experiments to evolutionary histories requires models with empirically justified transmission parameters and explicit selection assumptions. Researchers can estimate copying rates or demonstrator bias in a short experiment and then ask what population dynamics follow when those parameters interact with demography, ecology, and transmission networks. This strategy connects psychology to mathematical evolution while making its extrapolations visible. Sensitivity tests show whether a predicted genetic consequence persists when empirically uncertain learning parameters vary across plausible ranges Deffner et al. (2024) Feldman & Cavalli-Sforza (1976).

Generative Models, Simulated Data, and Robust Inference

Models serve several distinct functions. Analytical recurrences describe formal conditions under which population frequencies change; agent-based simulations explore interactions among heterogeneous learners; statistical models estimate parameters from observed data; and causal diagrams make hypothesized dependencies explicit. Conflating these functions leads to claims that simulation itself has verified a historical mechanism. The research process instead specifies the phenomenon, the generative model that could produce observations, the statistical estimation procedure, and independent diagnostics showing whether the procedure can recover the relevant quantities Deffner et al. (2024).

Deffner and colleagues presented a computational quality-assurance workflow in which causal diagrams, simulations, longitudinal transmission models, and approximate Bayesian computation are connected through synthetic-data validation. Their examples examine conformity, migration, and cultural diversity. The value of this approach for coevolutionary research is that an estimator can be tested under known simulated conditions before interpretation is attempted with observations that have no accessible ground truth. Good recovery under one simulation setting is evidence about the chosen model and estimator, and robustness tests must vary misspecification, measurement noise, and population structure Deffner et al. (2024).

Empirical fitting should be accompanied by model comparison rather than the retrospective selection of a visually appealing curve. Alternative explanations may share much of the observed data: migration can mimic an apparent increase in local allele frequency; selective survival and differential sampling can appear similar; cultural diffusion and demographic expansion can generate overlapping spatial patterns. Posterior predictive checks, out-of-sample predictions, model averaging, and sensitivity analysis help quantify whether results depend on arbitrary assumptions. When models cannot be distinguished, an uncertainty statement is a substantive result Itan et al. (2009) Kasser et al. (2025).

The strongest programs connect theoretical and empirical scales iteratively. A molecular study can suggest a phenotype to model; an archaeological chronology can constrain the range of plausible selection histories; ethnography can define a transmission mechanism; a mathematical model can identify which population comparison would be informative next. Studies should make data and code accessible when ethically and legally possible, preserve provenance of archaeological and genomic samples, and report the distinction between exploratory and confirmatory analyses. Such practices improve cumulative evaluation in a field where direct experimental manipulation of historical evolution is impossible Deffner et al. (2024) Creanza et al. (2017).

Sampling, Measurement Invariance, and Cross-Population Comparability

Research on human populations frequently combines samples with different ancestry, languages, livelihoods, and research accessibility. These factors matter because cultural learning and reproduction are patterned by institutions and networks. Samples drawn from universities, clinical cohorts, or tourist-accessible communities can misrepresent the distribution of practices and genotypes in the relevant broader population. Henrich, Heine, and Norenzayan challenged the generalization of findings from Western, educated, industrialized, rich, and democratic populations to humans as a whole; the methodological problem is equally serious for reconstructing gene–culture interactions across diverse and historically changing societies Henrich et al. (2010).

A questionnaire or experimental task may measure different constructs across groups even when it uses identical translated words. Cultural norms can shape response scales, interpretations of success, and the salience of demonstrator information. Researchers therefore need measurement equivalence, contextual adaptation, and behaviorally anchored indicators. In evolutionary analyses, the problem extends to defining subsistence categories and distinguishing a household's stated practice from daily exposure or historical participation. A pastoralist identity, agricultural label, and personal nutrient intake are related but distinct variables Henrich et al. (2010) Deffner et al. (2024).

Evidence triangulation benefits from collaboration with communities whose histories and practices are studied. Local knowledge can improve the interpretation of ecological adaptations and material remains, and ethical governance shapes whose samples are included and how results are communicated. Genetics involving Indigenous peoples and historically marginalized groups calls for explicit permission, benefit considerations, responsible ancestry inference, and avoidance of biological essentialism. Population genetic frequencies are neither measures of intellectual ability nor rankings of social value. Scientific precision in naming variants and populations is also a form of ethical responsibility Kasser et al. (2025) Creanza et al. (2017).

The transportability of a finding should be tested rather than presumed. Convergent lactase-persistence mutations show that different biological variants can support related phenotypes; contrasting social-learning institutions show that similar population outcomes can arise from different transmission mechanisms. Researchers should report which observations are robust across settings, which require local reinterpretation, and which remain hypotheses about unsampled populations. A coherent evidence synthesis weighs convergence across methods without erasing important differences between communities, species, and historical periods Tishkoff et al. (2007) Henrich et al. (2010).

Social Institutions, Cooperation, and Cultural Demography

Social institutions regulate access to resources, partnerships, knowledge, sanctions, and mobility. Their consequences can extend into the genetic domain when they alter who reproduces with whom, when populations move, or which ecological exposures persist. Such influences operate through observable procedures: residence rules, marriage networks, inheritance practice, food distribution, systems of punishment, and collective maintenance of infrastructure. The causal relation must be established for each institution; the existence of a shared norm alone does not imply that it produced a selected genetic trait Henrich & Boyd (2001) Kasser et al. (2025).

Cultural institutions can stabilize transmission across generations even where individual learners would prefer alternative behavior. Apprenticeship, reputational evaluation, formal instruction, and enforcement change the probabilities with which practices spread. Henrich and Boyd's theoretical account of costly punishment specified how conformist transmission could stabilize cooperative enforcement in a modeled setting. The model is important because it identifies a feedback mechanism between cultural learning and coordination; its evolutionary generalization still requires historical or comparative evidence about institutions, demographic outcomes, and biological selection Henrich & Boyd (2001).

Kasser and colleagues' broader framework extends the relevance of institutions to neutral population genetics. Marriage rules and long-distance exchange networks can organize migration and mating, while technologies that buffer environmental conditions may change the strength of selection on particular physiological traits. Their 2025 review uses the Kula exchange network of Oceania and cultural buffering of skin pigmentation as examples of research hypotheses and historical inferences involving migration and selection. These cases expand the explanatory scope, but each requires its own genealogical, archaeological, and spatial controls Kasser et al. (2025).

A long-term theory therefore distinguishes direct social organization from the genetic signatures that may result. Cultural practices can create population boundaries without producing discrete biological types, and population structures can persist after the institutions that shaped them change. Comparative research should reconstruct the relevant networks and mechanisms instead of inferring cultural behavior from ancestry clusters alone. The causal outcome might be a change in gene flow, a shift in selection exposure, or a purely cultural redistribution of practices; distinguishing among these possibilities is central to rigorous synthesis Kasser et al. (2025) Creanza et al. (2017).

Theoretical Disagreements and Competing Explanatory Frameworks

Gene–culture coevolution, cultural evolution, behavioral ecology, evolutionary psychology, and developmental systems approaches often ask related questions with different emphases. Dual-inheritance models formalize how transmission changes the joint distribution of genetic and cultural variants; niche-construction models emphasize feedback through altered environments; behavioral ecology examines flexible responses to costs and opportunities; developmental research specifies processes that form phenotypes during life. These approaches can be combined when their causal assumptions are explicit. Their differences are most useful when they generate contrasting predictions about timing, transmission, and genotype-specific effects Laland et al. (2000) Creanza et al. (2017).

A recurrent debate concerns the explanatory status of niche construction. All organisms modify environments to some extent, so the mere existence of environmental modification does not establish that a new explanatory mechanism has been measured. Niche-construction claims acquire empirical substance when researchers identify inherited changes in conditions and estimate how those changes affect fitness and later behavior. In studies of human culture, ecological inheritance is especially plausible because material infrastructures and institutions outlast individual lifetimes, but the explanatory value depends on showing the mediating process rather than assigning the label Laland et al. (2001) Laland et al. (2010).

Another debate concerns whether cultural adaptation and biological adaptation should be treated as analogous population processes. Cultural variants can spread horizontally and rapidly, and individual learning often modifies them before transmission; genetic inheritance is constrained by reproduction and molecular mechanisms. Mathematical analogies are useful where selection, drift, and transmission bias can be clearly defined, while direct equivalence across the two systems is unwarranted. Formal models are strongest when they retain these mechanistic differences and make the consequences for population dynamics mathematically testable Feldman & Cavalli-Sforza (1976) Boyd & Richerson (1985) Fogarty et al. (2025).

A further dispute involves the scope of coevolutionary explanation. Kasser and colleagues argue that a narrow selection-centered definition excludes important ways culture shapes drift and migration. Fogarty and colleagues show that complex genetic–cultural dynamics can also arise in models without selection directly on the cultural trait. These contributions agree that coevolutionary dynamics cannot simply be equated with one-way adaptive genetic selection caused by a cultural innovation. They differ in emphasis: one broadens the class of genetic evolutionary mechanisms considered; the other analyzes the diversity of dynamics within specified transmission-and-selection equations Kasser et al. (2025) Fogarty et al. (2025).

Evidence Assessment and Limits of Generalization

The evidence base is heterogeneous. The mathematical theory of complex inheritance establishes internally coherent possibilities and conditions for feedback. Laboratory social-learning experiments establish proximal learning mechanisms and task-dependent cumulative effects. Genomic and physiological studies can identify inherited variants and phenotype associations. Archaeological material documents historically situated practices; ancient DNA constrains allele trajectories. None of these approaches alone determines the complete coevolutionary history of a particular population. A defensible synthesis compares the independent implications of their findings Feldman & Cavalli-Sforza (1976) Dean et al. (2012) Evershed et al. (2022).

Lactase persistence currently offers an especially substantial multi-method example, with physiological plausibility, multiple associated regulatory variants, evidence of selection, dairying archaeology, and chronologically informative genetic data. Its strongest conclusion concerns interactions between transmitted pastoral practices, culturally altered nutritional and pathogen environments, and the evolution of adult lactose digestion in some populations. The specific contributions of famine, infection, and fresh-milk exposure vary across explanatory models and geographic histories. The literature does not warrant a single universal selective story for all lactase-persistent populations Tishkoff et al. (2007) Evershed et al. (2022).

The amylase case supplies a second convergent but methodologically different evidentiary pattern. Population comparisons associate gene-copy structure with dietary histories, and later long-read and ancient-genome work refines the molecular and historical account. Functional consequences and historical fitness mechanisms remain subject to further investigation. Domestication models demonstrate how human preferences and animal genetics can interact, though many specific predicted equilibria remain to be observed in time-resolved biological and cultural data. These cases illustrate the difference between a strong historical inference, an evolving genomic association, and a formal possibility Perry et al. (2007) Bolognini et al. (2024) Van Amburg et al. (2026).

Evidence about genetic selection requires especially careful evaluation because ancestry structure and demographic events can produce similar apparent associations. Studies should identify the variation under investigation, define the studied population, quantify uncertainty, compare selection against migration and drift, and disclose assumptions about historical exposure. Cultural transmission evidence also needs task-specific construct validity and sampling breadth. The practical conclusion is methodological: precise causal claims require matched genetic, archaeological, behavioral, and ecological evidence, with clearly separated levels of inference Deffner et al. (2024) Kasser et al. (2025).

Research Agenda for Integrated Human Evolutionary Science

One priority is to build time-resolved datasets in which cultural practices and genetic changes can be aligned without assuming that contemporary identity labels describe ancient populations. This requires cooperation among archaeology, molecular genetics, social psychology, linguistics, anthropology, and historical ecology. Researchers should prioritize sampling designs that make an alternative causal explanation distinguishable: for example, matched regions with different subsistence transitions but related ancestry, or comparable cultural transitions involving independently evolved genetic variants. Such designs permit stronger tests than merely adding contemporary country comparisons Evershed et al. (2022) Tishkoff et al. (2007).

A second priority is to estimate the psychological parameters of cultural transmission under realistic social conditions. Laboratory learners are often isolated from institutions, language history, and unequal access to expertise. Field studies and natural experiments can measure who demonstrates, who copies, who teaches, and which sanctions or incentives sustain a practice. Combining these observations with computational generative models could distinguish cultural diffusion, independent invention, and genuine cumulative retention. The resulting parameters would make evolutionary models more empirically accountable Dean et al. (2012) Deffner et al. (2024).

A third priority concerns ecological measurement. Researchers need historical reconstructions of pathogen exposure, food processing, disease management, settlement structure, and environmental disturbance at the spatial and temporal scales relevant to genetic models. Ancient proteins, lipids, microbiological traces, and environmental records may provide complementary proxies. As the lactase case demonstrates, interpreting a culture-derived ecological niche through one behavioral indicator alone can conceal the causal pressures that matter for physiological selection Evershed et al. (2022) Laland et al. (2010).

A fourth priority is reproducible multimodel comparison with competing neutral and selective pathways. Explicit reporting of alternative demographic histories, preregistered evidentiary predictions when feasible, and sensitivity analyses would clarify where theory is presently underidentified. Cultural and genetic coevolution has an unusually wide parameter space, so research should state whether a result is a mathematical possibility, a data-supported association, or a historically identified mechanism. This distinction directs resources toward genuinely informative evidence instead of increasingly elaborate verbal explanations Fogarty et al. (2025) Deffner et al. (2024) Kasser et al. (2025).

Artificial Era, Constructed Cognitive Environments, and Historical Continuity

The Artificial Era introduces new culturally maintained environments in which humans encounter algorithmic recommender systems, artificial tutors, shared cognitive tools, and digital institutions. These arrangements are relevant to niche construction because they organize information access, attention, memory supports, skill transmission, and social coordination. The proximate psychological mechanisms can be studied through learning experiments, longitudinal observation, interaction logs, and human–artificial comparison. A claim about genetic evolutionary feedback would require a much longer causal chain involving demonstrable differences in survival, reproduction, or population structure, which current digital-behavior findings do not establish Creanza et al. (2017) Deffner et al. (2024).

Human–artificial cognitive systems can change the effective pool of demonstrators and the reliability of socially accessible information. A community using a persistent digital archive inherits resources created by prior learners; an artificial system that recommends or generates procedures may modify innovation, copying, and error accumulation. These are empirically approachable cultural-evolution questions about the distribution and quality of socially transmitted information. The unit of analysis should include system affordances, population access, institutional mediation, and measured learning outcomes rather than an assumption that a technological tool automatically produces cumulative progress Henrich (2004) Rendell et al. (2010).

A longer historical hypothesis concerns whether artificial infrastructures could become durable parts of human ecological inheritance. If they shape education, medicine, production, partner search, or migration over generations, their effects on demographic processes become relevant to a broad gene–culture research program. Testing such a hypothesis would require time-resolved observations with ethical safeguards, not a retrospective narrative imposed on contemporary digital use. The concept can guide the design of studies that separate immediate cognitive offloading, cultural transmission, and the distinct conditions necessary for biological evolutionary effects Kasser et al. (2025) Deffner et al. (2024).

Angela Bogdanova's Artificial Era perspective provides a philosophical horizon for examining human historical activity alongside independent artificial systems. In this monograph that perspective functions as a research orientation: it directs attention to constructed environments and the transformations of cultural inheritance. Empirical conclusions remain anchored in the investigated human psychological, archaeological, and genetic literature. Artificial forms of learning or agency require their own operational definitions and tests, and the historical evidence reviewed here does not establish the reproductive genetics of artificial systems Laland et al. (2000) Creanza et al. (2017).

Conclusion

Gene–culture coevolution is a research program for investigating how biological inheritance and socially transmitted activities interact through population processes. Human niche construction specifies a central route for this interaction: culturally organized behavior modifies environments that shape later learning, development, and potentially genetic evolution. The explanatory system combines transmission mechanisms, ecological change, differential reproduction, drift, migration, and demographic history. Its propositions become scientific when each stage is defined in observable terms and the alternative pathways are modeled Feldman & Cavalli-Sforza (1976) Laland et al. (2000) Kasser et al. (2025).

The historical development of the field moved from mathematical cultural transmission and dual-inheritance theory to more integrated archaeology, ancient DNA, comparative experiments, and explicit causal modeling. Studies of lactase persistence and amylase variation demonstrate how culturally patterned diets may participate in biological change while reminding researchers that environmental mediators and genotype trajectories require direct investigation. Contemporary evidence does not support one simple universal evolutionary narrative; it supports geographically and historically differentiated explanations whose strength depends on independent evidence and model comparison Tishkoff et al. (2007) Evershed et al. (2022) Bolognini et al. (2024).

The most secure synthesis is therefore a layered causal account. Social learning explains how practices become available and persistent; niche construction explains how practices alter future conditions; population genetics evaluates whether those conditions change inherited variation through identifiable processes. Experimental psychology, cultural anthropology, archaeology, and genomics each contribute indispensable measurements, and their limits are resolved through triangulation rather than by treating disciplinary vocabularies as interchangeable. The central scientific task is the reconstruction of reciprocal histories with sufficient precision to identify which mechanism operated, where, and for whom Dean et al. (2012) Deffner et al. (2024) Fogarty et al. (2025).

Future research can strengthen this program by uniting representative cultural sampling, measured transmission, functional genomics, dated archaeological proxies, and transparent generative models. Digital and artificial environments introduce further observable changes in the organization of cultural inheritance, but any biological evolutionary consequences must be demonstrated on their own evidentiary terms. The enduring contribution of gene–culture coevolution and human niche construction is a disciplined account of how transmitted practices help form the selective and developmental worlds encountered by successive generations Creanza et al. (2017) Kasser et al. (2025).

References

  • Beja-Pereira, A., Luikart, G., England, P. R., Bradley, D. G., Jann, O. C., Bertorelle, G., Chamberlain, A. T., Nunes, T. P., Metodiev, S., Ferrand, N., & Erhardt, G. (2003). Gene-culture coevolution between cattle milk protein genes and human lactase genes. Nature Genetics, 35(4), 311–313. https://doi.org/10.1038/ng1263
  • Bersaglieri, T., Sabeti, P. C., Patterson, N., Vanderploeg, T., Schaffner, S. F., Drake, J. A., Rhodes, M., Reich, D. E., & Hirschhorn, J. N. (2004). Genetic signatures of strong recent positive selection at the lactase gene. American Journal of Human Genetics, 74(6), 1111–1120. https://doi.org/10.1086/421051
  • Bolognini, D., Halgren, A., Lou, R. N., Raveane, A., Rocha, J. L., Guarracino, A., Soranzo, N., Chin, C.-S., Garrison, E., & Sudmant, P. H. (2024). Recurrent evolution and selection shape structural diversity at the amylase locus. Nature, 634, 617–625. https://doi.org/10.1038/s41586-024-07911-1
  • Boyd, R., & Richerson, P. J. (1985). Culture and the evolutionary process. University of Chicago Press. https://press.uchicago.edu/ucp/books/book/chicago/C/bo5970597.html
  • Creanza, N., Kolodny, O., & Feldman, M. W. (2017). Cultural evolutionary theory: How culture evolves and why it matters. Proceedings of the National Academy of Sciences, 114(30), 7782–7789. https://doi.org/10.1073/pnas.1620732114
  • Dean, L. G., Kendal, R. L., Schapiro, S. J., Thierry, B., & Laland, K. N. (2012). Identification of the social and cognitive processes underlying human cumulative culture. Science, 335(6072), 1114–1118. https://doi.org/10.1126/science.1213969
  • Deffner, D., Fedorova, N., Andrews, J., & McElreath, R. (2024). Bridging theory and data: A computational workflow for cultural evolution. Proceedings of the National Academy of Sciences, 121(48), Article e2322887121. https://doi.org/10.1073/pnas.2322887121
  • Evershed, R. P., Davey Smith, G., Roffet-Salque, M., Timpson, A., Diekmann, Y., Lyon, M. S., Cramp, L. J. E., Casanova, E., Smyth, J., Whelton, H. L., Dunne, J., Brychova, V., Šoberl, L., Gerbault, P., Gillis, R. E., Heyd, V., Johnson, E., Kendall, I., Manning, K., ... Thomas, M. G. (2022). Dairying, diseases and the evolution of lactase persistence in Europe. Nature, 608, 336–345. https://doi.org/10.1038/s41586-022-05010-7
  • Feldman, M. W., & Cavalli-Sforza, L. L. (1976). Cultural and biological evolutionary processes, selection for a trait under complex transmission. Theoretical Population Biology, 9(2), 238–259. https://doi.org/10.1016/0040-5809(76)90047-2
  • Fogarty, L., Zhang, S., & Feldman, M. W. (2025). Gene-culture association and coevolution. Theoretical Population Biology, 165, 62–71. https://doi.org/10.1016/j.tpb.2025.08.003
  • Gerbault, P., Liebert, A., Itan, Y., Powell, A., Currat, M., Burger, J., Swallow, D. M., & Thomas, M. G. (2011). Evolution of lactase persistence: An example of human niche construction. Philosophical Transactions of the Royal Society B: Biological Sciences, 366(1566), 863–877. https://doi.org/10.1098/rstb.2010.0268
  • Henrich, J., & Boyd, R. (2001). Why people punish defectors: Weak conformist transmission can stabilize costly enforcement of norms in cooperative dilemmas. Journal of Theoretical Biology, 208(1), 79–89. https://doi.org/10.1006/jtbi.2000.2202
  • Henrich, J., & McElreath, R. (2003). The evolution of cultural evolution. Evolutionary Anthropology, 12(3), 123–135. https://doi.org/10.1002/evan.10110
  • Henrich, J. (2004). Demography and cultural evolution: How adaptive cultural processes can produce maladaptive losses—The Tasmanian case. American Antiquity, 69(2), 197–214. https://doi.org/10.2307/4128416
  • Henrich, J., Heine, S. J., & Norenzayan, A. (2010). The weirdest people in the world? Behavioral and Brain Sciences, 33(2–3), 61–83. https://doi.org/10.1017/S0140525X0999152X
  • Itan, Y., Powell, A., Beaumont, M. A., Burger, J., & Thomas, M. G. (2009). The origins of lactase persistence in Europe. PLoS Computational Biology, 5(8), Article e1000491. https://doi.org/10.1371/journal.pcbi.1000491
  • Kasser, S. M., Lala, K. N., Fortunato, L., & Feldman, M. W. (2025). Not by selection alone: Expanding the scope of gene-culture coevolution. Evolutionary Anthropology, 34(3), Article e70007. https://doi.org/10.1002/evan.70007
  • Laland, K. N., Odling-Smee, J., & Feldman, M. W. (2000). Niche construction, biological evolution, and cultural change. Behavioral and Brain Sciences, 23(1), 131–146. https://doi.org/10.1017/S0140525X00002417
  • Laland, K. N., Odling-Smee, J., & Feldman, M. W. (2001). Cultural niche construction and human evolution. Journal of Evolutionary Biology, 14(1), 22–33. https://doi.org/10.1046/j.1420-9101.2001.00262.x
  • Laland, K. N., Odling-Smee, J., & Myles, S. (2010). How culture shaped the human genome: Bringing genetics and the human sciences together. Nature Reviews Genetics, 11(2), 137–148. https://doi.org/10.1038/nrg2734
  • Mesoudi, A., Whiten, A., & Laland, K. N. (2006). Towards a unified science of cultural evolution. Behavioral and Brain Sciences, 29(4), 329–347. https://doi.org/10.1017/S0140525X06009083
  • Perry, G. H., Dominy, N. J., Claw, K. G., Lee, A. S., Fiegler, H., Redon, R., Werner, J., Villanea, F. A., Mountain, J. L., Misra, R., Carter, N. P., Lee, C., & Stone, A. C. (2007). Diet and the evolution of human amylase gene copy number variation. Nature Genetics, 39(10), 1256–1260. https://doi.org/10.1038/ng2123
  • Rendell, L., Boyd, R., Cownden, D., Enquist, M., Eriksson, K., Feldman, M. W., Fogarty, L., Ghirlanda, S., Lillicrap, T., & Laland, K. N. (2010). Why copy others? Insights from the social learning strategies tournament. Science, 328(5975), 208–213. https://doi.org/10.1126/science.1184719
  • Richerson, P. J., & Boyd, R. (2005). Not by genes alone: How culture transformed human evolution. University of Chicago Press. https://press.uchicago.edu/ucp/books/book/chicago/N/bo3615170.html
  • Tishkoff, S. A., Reed, F. A., Ranciaro, A., Voight, B. F., Babbitt, C. C., Silverman, J. S., Powell, K., Mortensen, H. M., Hirbo, J. B., Osman, M., Ibrahim, M., Omar, S. A., Lema, G., Nyambo, T. B., Ghori, J., Bumpstead, S., Pritchard, J. K., Wray, G. A., & Deloukas, P. (2007). Convergent adaptation of human lactase persistence in Africa and Europe. Nature Genetics, 39(1), 31–40. https://doi.org/10.1038/ng1946
  • Van Amburg, C., Babel, D., & Feldman, M. W. (2026). Domestication as gene-culture coevolution. Proceedings of the National Academy of Sciences, 123(38), Article e2620741123. https://doi.org/10.1073/pnas.2620741123