Start hereDNA, genes, chromosomes — and the genotype/phenotype split
Before any of the arguments in this chapter make sense, four ideas have to be pinned down: DNA, the gene, the chromosome, and the difference between what you inherit and what you become.
Nearly every cell in the body carries the same instruction set, written in deoxyribonucleic acid (DNA) — a long molecule shaped like a twisted ladder whose rungs are pairs of four chemical bases (adenine with thymine, cytosine with guanine). The sequence of those base pairs is a code. A gene is a stretch of that code that specifies how to build a particular protein, and proteins are the molecular machines that run the body, from the enzymes that digest food to the receptors that let neurons talk to one another. Genes do not sit loose in the cell; they are packaged into chromosomes, tightly coiled bundles of DNA wound around protein spools. Humans carry 46 chromosomes arranged in 23 pairs, one member of each pair from each parent, with the twenty-third pair carrying the sex chromosomes (typically XX or XY). The full complement — every gene on every chromosome — is the genome, and sequencing it revealed a number that surprised almost everyone: only about 20,000 to 25,000 protein-coding genes, far too few to hold a separate instruction for each human trait (International Human Genome Sequencing Consortium, 2004).
That surprise is the doorway to the whole chapter. If there are only around twenty thousand genes and yet an essentially unlimited range of human bodies, temperaments, and abilities, then traits cannot map neatly onto single genes. The distinction that organizes everything downstream is between genotype — the specific set of alleles (gene variants) a person actually carries — and phenotype, the observable characteristic that emerges: height, eye color, sociability, blood pressure. Genotype is the recipe; phenotype is the finished dish, and the same recipe cooked in different kitchens does not produce the same meal. The genome is better read as a reaction range, a set of possibilities whose realization depends on everything the developing organism encounters.
The tidy cases everyone learns first are the exceptions rather than the rule. In dominant–recessive inheritance, a single gene comes in two versions and one masks the other: two recessive alleles are needed to express a recessive phenotype, while one dominant allele is enough to express the dominant one. Mendel's pea plants work this way, and so do a handful of human conditions — Huntington's disease from a single dominant allele, cystic fibrosis and sickle-cell disease from paired recessive ones. These single-gene patterns are real, medically important, and perfectly predictable with a Punnett square. They are also, as the next section argues, a small minority of what makes people differ.
The reframeAlmost nothing is a single-gene trait
The dominant/recessive Punnett-square world is real — but it covers a tiny minority of traits. Height, temperament, intelligence, risk for depression: these are polygenic, shaped by thousands of genetic variants each nudging the outcome a hair.
The evidence for this is now overwhelming. Genome-wide studies of height — perhaps the most heritable common trait, and one that is easy to measure — have identified more than twelve thousand associated variants, and together these common variants account for nearly all of the trait's estimated genetic variance only because they are so numerous, each one contributing an effect too small to matter on its own (Yengo et al., 2022). Adult height in a well-nourished population is roughly 80% heritable, yet no one has ever found a "height gene" in the sense a student might imagine; there are only many thousands of tiny contributions summed across the genome. The same architecture holds for the psychological traits this course cares about. Intelligence, personality, and risk for depression are massively polygenic, and the older search for single "genes for" behavior — a gene for intelligence, a gene for novelty-seeking — has largely been abandoned in favor of this many-small-effects picture (Chabris et al., 2015).
Two ideas make polygenic inheritance intuitive. First, because a polygenic trait is the sum of many independent contributions, it tends to be continuously distributed in the population — a smooth bell curve of height or sociability rather than a few discrete categories — exactly what a one-gene Punnett square could never produce. Second, the same logic that makes the trait continuous makes any single variant nearly powerless on its own. Knowing that a person carries a particular "risk allele" for depression shifts the odds only slightly, because it is one small voice in a chorus of thousands. This is why the leap from "we found genes associated with a trait" to "we can read a person's destiny from their DNA" is a leap the science does not license.
A genome-wide association study (GWAS) scans hundreds of thousands of people to find the many common variants statistically linked to a trait. Add those up and you get a polygenic score — a single number estimating someone's genetic propensity. What they can do: predict differences across large groups reasonably well for traits like height. What they can't do: reliably predict an individual child's outcome, or work well across ancestries (most GWAS samples are European, echoing the WEIRD problem from Unit 02). A polygenic score is a loaded tendency, never a verdict.
This is the single most misunderstood number in the course. Heritability describes variance within a population, not a person. If height is 80% heritable, it does not mean 80% of your height came from genes — that question is meaningless, like asking which is more important to a rectangle's area, length or width. It means that, in a given population and environment, about 80% of the differences among people trace to genetic differences. Two more traps: heritability is not fixed — it can rise when the environment is made more equal (everyone well-fed) and fall when environments vary wildly — and it says nothing about group differences or about how changeable a trait is. A highly heritable trait (like nearsightedness) can be corrected instantly with glasses.
The number everyone misreadsWhat heritability is — and Turkheimer's three laws
Heritability is a statistic with a precise and narrow meaning: within a particular population living in a particular range of environments, it is the proportion of the variance in a trait that can be attributed to genetic differences among people. The stress belongs on three words — variance, population, and differences. Heritability is not a property of a person, and it is not a property of a trait in the abstract; it is a property of a trait in a population at a time. The myth box above is worth restating because the error is so seductive: a heritability of .50 does not mean half of anyone's trait is genetic, any more than one can say which half of a drumbeat is the drum and which half is the drummer. The whole trait is built from both, every time.
Because heritability is about variance in a specific setting, it moves when the setting changes. When environments are made more uniform — say, when a whole population is well-fed — the environmental share of variance shrinks and heritability rises, since genetic differences become the main thing still varying. When environments become more unequal, heritability falls. A striking demonstration comes from cognitive ability in early childhood, where the heritability of IQ was found to be near zero among children in the poorest families and substantial among the affluent: in deprived environments the environment dominates the variance, leaving genetic potential less room to express itself (Turkheimer et al., 2003). Heritability is thus not a fixed biological constant but a moving readout of how a population is living — which is precisely why it cannot be used to argue that a trait is unchangeable or that differences between groups must be genetic.
The field's hard-won consensus was distilled into three deceptively simple statements (Turkheimer, 2000). The first law: all human behavioral traits are heritable — everything from cognitive ability to political attitudes shows some genetic influence, so a heritability greater than zero is the expected finding, not a discovery. The second law: the effect of being raised in the same family is smaller than the effect of genes — shared family environment, the thing most people assume is decisive, turns out to explain relatively little of why adult siblings differ. The third law, and the most consequential: a substantial portion of the variation in complex behavioral traits is not explained by genes or by families at all. That unexplained remainder — the nonshared environment together with chance, developmental noise, and the idiosyncratic accumulation of experience — is where much of the action lives, and it is exactly the part that a heritability statistic leaves in shadow.
The wrong questionNot nature vs. nurture — nature through nurture
Heredity–environment correlations are only half the story. The deeper idea is heredity–environment interaction: genes and environments don't just add up, they multiply, each changing what the other does.
Start with gene–environment correlation, the observation that genotypes and environments are not handed out independently. A child's genes and the child's surroundings tend to line up, and they do so in three ways (Scarr & McCartney, 1983). Passive correlation comes first: biological parents supply both the genes and the home, so a musically inclined child of musical parents grows up amid instruments she did not choose. Evocative correlation follows: a sociable, smiley infant draws warmth and conversation out of the people around her, so her genotype helps manufacture her social environment. Active correlation, or "niche-picking," grows dominant with age: as children gain freedom they seek out settings that suit their dispositions — the bookish child finds the library, the daring child finds the skate park — and the environment increasingly becomes an expression of the genotype. This is one reason the heritability of many traits increases across development rather than fading: with time, people build the very worlds that amplify their genetic tendencies. It is also why the clean separation of "genetic" and "environmental" causes is partly an illusion, since the environment is itself, in part, a genetic effect.
Diathesis–stress
A genetic vulnerability (diathesis) stays silent until a stressor triggers it. A gene variant may raise depression risk only for people who face severe adversity; in calm lives, it does nothing. Genes set a threshold; the environment decides whether it's crossed.
Differential susceptibility (Belsky & Pluess, 2009)
A richer idea: some kids are "orchids" — the same genes that make them wilt under harsh parenting make them flourish most under good parenting. Others are hardy "dandelions," doing okay anywhere. Sensitivity cuts both ways — for better AND worse — not just toward risk.
The famous early result — that a serotonin-transporter variant (5-HTTLPR) interacted with life stress to predict depression (Caspi et al., 2003) — became a textbook star. Then large, better-powered studies and meta-analyses often failed to replicate it. Being honest about this is the whole lesson from Unit 02: real science includes findings that don't hold up. The concept of gene × environment interaction is sound and important; that one specific candidate-gene result is shaky. Hold both.
The genuine version of the idea — gene–environment interaction, where the effect of an experience depends on genotype — is not in doubt, but its most famous demonstrations demand care. Two candidate-gene studies from the same research program defined the field's optimistic first decade. One reported that a variant of the MAOA gene predicted antisocial behavior only among boys who had been maltreated, sparing those with the same genotype who were not abused (Caspi et al., 2002). The other, described in the box above, reported that the short allele of the serotonin-transporter promoter predicted depression only in people exposed to stressful life events (Caspi et al., 2003). Both told a compelling story: the gene is not a fate but a sensitivity dial, turned up only when life turns harsh.
Then the replications came in, and the picture darkened. A large collaborative meta-analysis found no evidence that the 5-HTTLPR genotype, alone or in interaction with stress, was associated with depression risk (Risch et al., 2009), and a broader review concluded that the first ten years of candidate gene-by-environment research had produced a literature riddled with false positives — the product of small samples, flexible analytic choices, and publication bias, the very pathologies Unit 02 warned about (Duncan & Keller, 2011). The honest reading is layered. The specific claim that one measured gene interacts with stress to cause depression is, at best, unproven; the general principle that genetic differences shape sensitivity to the environment is supported by the whole of twin and adoption research. Modern behavioral genetics has largely moved from single candidate genes to genome-wide, polygenic methods precisely to escape the fragility of betting the story on one variant. Holding both halves at once — the concept sound, the celebrity finding shaky — is itself a lesson in how science self-corrects.
How we knowThe logic behind twin & adoption studies
Behavioral genetics has a clever reasoning engine at its core, and it is worth being able to explain rather than merely invoke. The problem it solves is that human genes and human environments are hopelessly tangled in ordinary families — the same parents supply both — so the field relies on natural experiments that pry the two apart. Twins and adoptees are the levers.
The estimates this machinery produces are remarkably consistent. A meta-analysis pooling virtually every twin study conducted over fifty years — nearly 15 million twin pairs across more than 17,000 traits — found an average heritability of about 49% across all human characteristics, with the balance attributable to environment (Polderman et al., 2015). The classic Minnesota Study of Twins Reared Apart reached a similar verdict for psychological traits specifically, reporting that identical twins separated in infancy and raised in different homes were nearly as alike in personality, interests, and measured intelligence as identical twins raised together (Bouchard et al., 1990). The number to carry away is not any single figure but the pattern: genes matter, roughly half the story, and never the whole of it.
Two cautions keep these designs honest. The equal-environments assumption — that identical and fraternal twins share their rearing environments to the same degree — is not perfectly true, since identical twins are often treated more alike, which can inflate heritability estimates. And twins reared apart share not only genes but nine months of a common prenatal environment, and are frequently placed in similar kinds of homes, both of which can masquerade as genetic similarity. None of this overturns the core findings; it simply means the estimates are approximations, not exact measurements.
The single most counterintuitive result to come out of this work concerns the environmental half. When behavioral geneticists decompose the environmental variance, most of it turns out to be nonshared — experiences that differ between siblings and make them less alike — rather than the shared family climate that dominates popular thinking about parenting (Plomin & Daniels, 1987). Two children raised in the same home by the same parents are, on many psychological traits, about as different as two children pulled at random from the population once their shared genes are accounted for. Whatever the family does that matters, much of it is experienced differently by each child. This finding, one of the most replicated in the field, reframes the parenting question: the environment is powerful, but its power runs largely through the particular, non-shared texture of each child's life (Plomin et al., 2016).
Above the genesEpigenetics, made concrete
Epigenetics really clicks once you see the mechanism and a case that makes it unforgettable.
The prefix says it all: epi- means "above" or "on top of," and epigenetics concerns the layer of regulation that sits on top of the DNA sequence and decides which genes are actually read in a given cell at a given time. Every cell in the body carries the same genome, yet a neuron and a liver cell could hardly be more different; the difference is epigenetic, a matter of which genes are switched on. The best-understood mechanism is DNA methylation, the attachment of small chemical tags (methyl groups) to the DNA that typically quiet a gene without altering a single base pair. Methylation patterns are laid down during development, can be influenced by experience, and — crucially for this chapter — can persist. The sequence is the hardware; methylation is part of the software that decides how the hardware runs.
The experiment that made this concrete in behavioral science examined mother rats and their pups (Weaver et al., 2004). Rat mothers differ naturally in how much they lick and groom their offspring, and pups of high-licking mothers grow into calmer, less stress-reactive adults. Maternal care was found to alter the methylation of a gene governing stress-hormone receptors in the pup's hippocampus: attentive mothering left the gene less methylated and more active, tuning the stress system for life. The clinching detail was that cross-fostering pups to high- or low-licking mothers transferred the phenotype along with the caregiving, showing the effect ran through the behavior of the rearing mother and not through inherited genes. Here, in one study, was a physical mechanism for how nurture gets "under the skin" and into the regulation of the genome — nature and nurture meeting at a chemical tag.
Epigenetics is about which genes get switched on or off without changing the DNA sequence — chemical tags like DNA methylation that act as dimmer switches on gene expression. The haunting evidence: during the Dutch Hunger Winter (1944–45), a Nazi blockade starved a population. Decades later, people conceived during that famine showed distinct methylation patterns on a growth-related gene (IGF2) and elevated metabolic risk — a prenatal environmental event chemically inscribed on the genome and detectable 60 years on (Heijmans et al., 2008). The environment doesn't just interact with genes; it can leave a mark on how they're read.
The tools have leapt ahead of the biology many of us learned. CRISPR-Cas9 lets scientists cut and rewrite DNA cheaply and precisely. In 2018, He Jiankui announced he had edited the embryos of twin girls born in China — the first gene-edited humans — igniting global condemnation and a prison sentence, because the edits were heritable (they'd pass to future generations), the medical justification was thin, and consent for the unborn is impossible. The science-literacy point: the germline-editing debate is no longer hypothetical, and "can we?" has outrun "should we?"
The synthesisWhy "nature vs. nurture" is the wrong question
Put the pieces together and the old debate collapses. If a genotype only ever expresses itself through an environment, if genes and environments are correlated because people build and evoke their own surroundings, if the effect of an experience can depend on genotype, and if experience can chemically alter how genes are read, then "nature versus nurture" is not a hard question with a missing answer — it is a badly formed question. Asking whether a person's temperament is due to genes or environment is like asking whether a river's course is due to the water or the banks; neither exists as a river without the other. The developmental view, sometimes summed up as "nature through nurture," treats every phenotype as the outcome of a genome and an environment in continuous, mutually shaping conversation from conception onward (Rutter, 2006).
This reframing carries a practical edge that matters for everything that follows in the course. Because heritability is not destiny, strong genetic influence on a trait still leaves enormous room for intervention — a highly heritable reading difficulty responds to teaching, just as highly heritable nearsightedness responds to glasses. Because the environmental variance is largely nonshared, helping children is less about providing a generic "good home" and more about the specific experiences each child actually has. And because the genome is responsive rather than fixed, the environments we build for developing humans are not merely the stage on which genes perform; they are among the forces that decide which genetic possibilities become real. The remaining units are, in a sense, a long elaboration of that single idea.
SourcesCited in APA 7
Belsky, J., & Pluess, M. (2009). Beyond diathesis stress: Differential susceptibility to environmental influences. Psychological Bulletin, 135(6), 885–908.
Bouchard, T. J., Jr., Lykken, D. T., McGue, M., Segal, N. L., & Tellegen, A. (1990). Sources of human psychological differences: The Minnesota Study of Twins Reared Apart. Science, 250(4978), 223–228.
Caspi, A., McClay, J., Moffitt, T. E., Mill, J., Martin, J., Craig, I. W., Taylor, A., & Poulton, R. (2002). Role of genotype in the cycle of violence in maltreated children. Science, 297(5582), 851–854.
Caspi, A., Sugden, K., Moffitt, T. E., Taylor, A., Craig, I. W., Harrington, H., McClay, J., Mill, J., Martin, J., Braithwaite, A., & Poulton, R. (2003). Influence of life stress on depression: Moderation by a polymorphism in the 5-HTT gene. Science, 301(5631), 386–389.
Chabris, C. F., Lee, J. J., Cesarini, D., Benjamin, D. J., & Laibson, D. I. (2015). The fourth law of behavior genetics. Current Directions in Psychological Science, 24(4), 304–312.
Duncan, L. E., & Keller, M. C. (2011). A critical review of the first 10 years of candidate gene-by-environment interaction research in psychiatry. American Journal of Psychiatry, 168(10), 1041–1049.
Heijmans, B. T., Tobi, E. W., Stein, A. D., Putter, H., Blauw, G. J., Susser, E. S., Slagboom, P. E., & Lumey, L. H. (2008). Persistent epigenetic differences associated with prenatal exposure to famine in humans. Proceedings of the National Academy of Sciences, 105(44), 17046–17049.
International Human Genome Sequencing Consortium. (2004). Finishing the euchromatic sequence of the human genome. Nature, 431(7011), 931–945.
Plomin, R., & Daniels, D. (1987). Why are children in the same family so different from one another? Behavioral and Brain Sciences, 10(1), 1–16.
Plomin, R., DeFries, J. C., Knopik, V. S., & Neiderhiser, J. M. (2016). Top 10 replicated findings from behavioral genetics. Perspectives on Psychological Science, 11(1), 3–23.
Polderman, T. J. C., Benyamin, B., de Leeuw, C. A., Sullivan, P. F., van Bochoven, A., Visscher, P. M., & Posthuma, D. (2015). Meta-analysis of the heritability of human traits based on fifty years of twin studies. Nature Genetics, 47(7), 702–709.
Risch, N., Herrell, R., Lehner, T., Liang, K.-Y., Eaves, L., Hoh, J., Griem, A., Kovacs, M., Ott, J., & Merikangas, K. R. (2009). Interaction between the serotonin transporter gene (5-HTTLPR), stressful life events, and risk of depression: A meta-analysis. JAMA, 301(23), 2462–2471.
Rutter, M. (2006). Genes and behavior: Nature–nurture interplay explained. Blackwell.
Scarr, S., & McCartney, K. (1983). How people make their own environments: A theory of genotype → environment effects. Child Development, 54(2), 424–435.
Turkheimer, E. (2000). Three laws of behavior genetics and what they mean. Current Directions in Psychological Science, 9(5), 160–164.
Turkheimer, E., Haley, A., Waldron, M., D'Onofrio, B., & Gottesman, I. I. (2003). Socioeconomic status modifies heritability of IQ in young children. Psychological Science, 14(6), 623–628.
Weaver, I. C. G., Cervoni, N., Champagne, F. A., D'Alessio, A. C., Sharma, S., Seckl, J. R., Dymov, S., Szyf, M., & Meaney, M. J. (2004). Epigenetic programming by maternal behavior. Nature Neuroscience, 7(8), 847–854.
Yengo, L., Vedantam, S., Marouli, E., Sidorenko, J., Bartell, E., Sakaue, S., … Lettre, G. (2022). A saturated map of common genetic variants associated with human height. Nature, 610(7933), 704–712.