at Harvard University. “The deeper I’ve gotten into this work, the more I’ve come to think: maybe it’s just a fluke.”
Questions about the relative importance of our innate nature and upbringing go way back, even featuring in debates between the Greek philosopher Plato and his pupil Aristotle. But the catchy phrase “nature versus nurture” came from 19th-century scientist Francis Galton, a pioneer in the emerging field of heredity. Galton also popularised eugenics, the repugnant idea that some people with “undesirable” traits should be stopped from having children.
Modern scientists still struggle to dissociate the field of behavioural genetics from its shameful past, but they have made progress in measuring the relative impacts of genes and environment on traits such as intelligence, introversion and self-control. Nature and nurture are often hard to untangle because children in the same families tend to share both DNA and the way they were raised. For instance, if children who are read to as babies tend to do well at school, that could be because early exposure to books fuels their interest in learning or because they have inherited bookish genes from their parents.
Geneticists have addressed this problem using two kinds of studies. The first involves comparing identical twins with non-identical twins. The former share pretty much all of their DNA, the latter share only half, but both are assumed to share their upbringing to a similar extent. The second sort of study compares pairs of biological siblings with pairs of siblings in which one child was adopted into the family at birth. Both types of pairs tend to have similar upbringings, but the biological siblings share half their DNA, whereas the adoptive ones share none.
Several decades of these kinds of studies, involving hundreds or even thousands of participants, have generated a broad consensus: of the influences that we can measure, genes do most of the work. The exact figure varies depending on the trait – happiness, resilience, or whatever – but DNA often accounts for about half of the difference between individuals. The effects of upbringing, on the other hand, are much smaller, ranging from 17 per cent to as little as 0 per cent. “In adoption studies, correlations between characteristics of families and adopted children are basically zero,” says Eric Turkheimer at the University of Virginia. A notable exception is intelligence, where the impact of upbringing is around 25 per cent during childhood. However, this dwindles to almost nothing in adulthood, when any effects from hothouse schooling or pushy parenting fade and people have much more control over their intellectual lives. There is also a caveat – in cases of childhood abuse, the environmental impacts can have large and lasting effects. “Clearly, if you lock a child in a closet, they’re going to be damaged,” says geneticist Robert Plomin at King’s College London, who wrote about these findings in his 2018 book Blueprint: How DNA makes us who we are. “But we are talking about why children in the normal population with normal upbringings are different [from one another].”
Despite the compelling evidence, the idea that parenting has little lasting influence on the intelligence and personality of offspring isn’t widely accepted outside the field of behavioural genetics. One reason is its social and political associations: not only the stain left by eugenics, but also because, today, many on the right of the political spectrum see some of society’s ills as rooted in human genetics. This can sometimes lead their opponents on the left to reject the findings of studies showing that genes are so influential. Another problem is that there are many studies claiming to find large effects of early environment on life outcomes. However, any such research that doesn’t use the twin or adoptive sibling methods to tease out the contributions of genes is flawed.
Even reputable research papers may use language that confuses the issue. If a study finds, for instance, that genes account for half of the variation in IQ between adults, researchers often report they have found that genes do 50 per cent of the work, and “the environment” does the rest. They are using “environment” to mean anything that isn’t genes. Sometimes, they even stress this so-called environmental contribution to show that they don’t believe we are slaves to our genes, says Kevin Mitchell, a neurogeneticist at Trinity College Dublin. “It’s like a cover for those who think [genetics] has dark connotations.”
Things become clearer when researchers divide these non-genetic factors into two types: “shared environment” and “non-shared environment”. The former refers to the environment children in the same family have in common – covering everything we would normally think of as upbringing, such as parenting style, family income, type of schooling and so on. This is what the twin and adoption studies have found lies between 17 and 0 per cent. If genes account for around 50 per cent of our variation, simple maths tells us that between 33 and 50 per cent is down to something else – the unhelpfully named non-shared environment. What this actually consists of remains mysterious and yet, going by the maths, its impact overshadows that of upbringing and sometimes equals the effects of genes. You could call this the dark matter of our personal origin stories.
What kind of influences might the dark matter consist of? For a long time, behavioural geneticists assumed it comprised many small factors, apparently too trivial to be measurable or even memorable. This could include things like chance encounters with teachers or friends, minor childhood illnesses or early romantic relationships. Plomin’s colleague at King’s College London, Damien Morris, gives a hypothetical example of a pair of identical twins sitting next to each other in school. “One stares out the window and is distracted by a bird flying by, just as the other twin is enraptured by the teacher’s account of a particular poem – and it forms a lifelong love of poetry,” he says.
But animal researchers have been finding growing evidence that at least some of the chance influences that shape us could happen as a result of random events that occur in the brain before birth. This may clash with some people’s understanding of how brains are made. DNA is often described as the body’s blueprint, as it is in the title of Plomin’s book. This makes it sound like genes encode a precise wiring diagram for the estimated 86 billion human brain cells and their 1000 trillion connections, known as synapses. But that isn’t the case. The genes involved in brain development are more like a recipe for making a cake, albeit one where each instruction has a degree of vagueness. At each new step – for instance, when embryonic brain cells multiply, migrate to new locations or specialise as different cell types – cells that do things one way in an individual do them to a slightly greater or lesser extent in another genetically identical person.
“We all start with our individual genome, then development is the unfolding of the organism based on the information in that genome, but it adds its own variation as it goes along,” says Mitchell. Or, as he puts it in his 2018 book, Innate: How the wiring of our brains shapes who we are: “You can’t bake the same cake twice.”
We can’t observe this unfolding in human fetuses because the method currently used for visualising the brain’s wiring diagram destroys it in the process. It is also highly laborious, requiring brain tissue to be turned into thousands of ultrathin slices, so has mainly been used to study animals with very small brains, like worms and fruit flies, as well as parts of the mouse brain.
But the initial findings from these studies show that genetically identical animals raised in conditions that are as similar as possible can have slightly different brain wiring. For instance, research by de Bivort and his colleagues found that fruit flies have an innate, long-lasting preference for turning to the left or right while walking around, which is controlled by the wiring pattern of brain cells called columnar neurons. He and his team discovered that fly offspring didn’t inherit this tendency from their parents, and that the variability was even present in a group of genetically identical flies. This, they conclude, suggests the cause was intrinsic random events. In 2020, another group of researchers found that differences in brain wiring result in random variation in the degree to which flies are attracted to high-contrast stripe patterns.
The tiny brains of flies, like those of most animals, are broadly symmetrical, so another strategy is to look at differences between the left and right sides. With this approach, it is even more likely that any differences arise due to chance, as the two sides have been built by the same set of genes and exposed to the same environment. Researchers are increasingly finding that there can be asymmetry between the two sides at the level of individual brain cells. For instance, when it comes to projection neurons – a type of brain cell involved in smell – either side of the brain can have two, three or four of them.
It is unimaginable to do these sorts of experiments in people, but our shared biology with flies suggests there would be parallels. “Things you find in flies at the molecular level tend to be present in humans at the molecular level,” says de Bivort. What’s more, although we can’t yet visualise the brain-wiring diagrams of larger creatures, looking at their behaviour tells us that genetically identical animals can differ in traits, like innate aggression and exploratory behaviour. This has been shown in mice, crayfish and armadillos, and may become more widely appreciated thanks to the growing industry of pet cloning. Earlier this year came news of a woman from Texas who paid $25,000 to have her beloved cat cloned after it died and found that, although the new pet looked just like her old one, it had quite a different personality.
The mechanisms underpinning these chance events in brain development are only just starting to emerge (see “Chance events shape who you are“). It also remains unclear when the crucial events occur. Some human geneticists believe that, in people, they are more likely to happen during childhood rather than when they are in the womb. This thinking is based on twin studies that find the impact of non-shared environment varies over childhood. For example, if the IQ of identical twins is measured, one might test higher than the other at a young age, but a few years later the difference may be absent, or even reversed. “If developmental noise [in the womb] accounted for the non-shared environment – because some had their neurons zig in an intelligence-enhancing direction, and others had them zag in the other direction – most of it would be persistent,” says Morris.
Random brain-wiring events might even be shaping who we are well beyond childhood. Neuroscientists used to think that the human brain finishes maturing in the first few years of life, but more recent brain-scanning work shows that it continues to develop throughout adolescence and even into early adulthood. Many key regions of teenage brains undergo a process called “pruning”, or loss of brain cell connections. This is usually assumed to be a directed response to their environment – but, in truth, we know very little about the mechanisms and causes of pruning. So it is possible that random events play at least some role.
Whenever the random events happen, they will, by their very nature, be hard to study. De Bivort hopes more insights will come from studies of “mini-brains”, tiny balls of neural tissue grown in the lab that replicate some aspects of fetal brain structure. Genetically identical human mini-brains could be grown under the same culture conditions and then sliced up to discern their wiring diagrams. Alternatively, if more powerful yet non-destructive brain-scanning techniques become available one day, they could be used on newborn identical twins to look at structural differences between the neuronal networks in their brains.
For some of the postulated “noise”, it is debatable if it stems from truly random events, as physicists would define them, or is just unmeasurable, but Plomin says that doesn’t matter. “From a practical point of view, a lot of the differences in our behaviour are essentially unpredictable,” he says.
This new way of understanding brain development may split opinions. It might be resisted by the psychoanalysis industry, whose practitioners are renowned for saying to their clients: “Tell me about your childhood.” Certainly, the idea that our psyche is profoundly affected by how we are raised is deeply rooted in culture, from works of fiction to parenting advice books. If random events really do trump parental influence, that seems to negate one of the English language’s best-known modern poems, Philip Larkin’s This Be the Verse, which begins: “They fuck you up, your mum and dad.” But the influence of chance may come as no surprise to parents who see clear differences in their children’s personalities despite their best attempts at equal treatment.
For now, much about these random influences remains mysterious. The question marks behind such a formative force on the human psyche might, for some, be a source of existential angst. But to others, it is liberating. “We have been so locked into genes versus environment being the only sources of variation, when, in fact, these idiosyncratic trajectories are what make us unique and unrepeatable,” says Mitchell. “There’s something sort of life-affirming about that.”
Who you are depends on your genes and on the environment in which you grew up. However, evidence from several kinds of research suggests that “noise” – random events during brain development – also plays a role in shaping each of us. What could those events be?
Nine-banded armadillos (Dasypus novemcinctus) can provide some answers because they are born in litters of identical quadruplets. Even though members of a quad are genetic clones, from early on in life they may differ slightly in features such as tameness or how many scales make up their bands, says Jesse Gillis at the University of Toronto in Canada.
Within each quad, about one-tenth of the differences in which genes are turned on or off in their white blood cells – used as a handy proxy for total variation – stems from random “silencing” of some of their genes caused by chemical changes to their DNA. This seems to happen when the embryos are balls of just a few hundred cells, Gillis’s team has calculated.
Random factors also influence another key stage in embryonic development, when immature brain cells produce projections, called axons, that probe their way through the brain to make connections with other neurons.
These outgrowths are attracted to nearby chemical cues, in the same way that microscopic animals hunt down food, so it seems plausible that chance variations in the distribution of these chemicals in the embryonic brain could affect the final wiring diagram. In flies, at least, such variation “can lead to large changes in the configuration of those cells”, says Benjamin de Bivort at Harvard University.
Developing human brains are much harder to peer into. Nevertheless, geneticists have found that chance mutations that happen as immature brain cells multiply during embryonic development are among the many factors linked with schizophrenia and autism.
Both are now seen not as all-or-nothing states, but as one end of a spectrum of mental characteristics. The same chance mutations linked with schizophrenia and autism could also play a role in personality differences in the rest of the population.
Join us at New Scientist Live
Discover what makes us human at the Mind and Body Stage newscientistlive.com