Highlights: Blueprint: How DNA Makes Us Who We Are, by Robert Plomin


Raw Generation of Kindle Notes, up for Summary + Review

Location 34-38

Environmentalism – the view that we are what we learn – dominated psychology for decades. From Freud onwards, the family environment, or nurture, was assumed to be the key factor in determining who we are. In the 1960s geneticists began to challenge this view. Psychological traits from mental illness to mental abilities clearly run in families, but there was a gradual recognition that family resemblance could be due to nature, or genetics, rather than nurture alone, because children are 50 per cent similar genetically to their parents.

Location 54-60

most measures of the environment that are used in psychology – such as the quality of parenting, social support and life events – show significant genetic impact. How is this possible when environments have no DNA themselves? As we shall see, genetic influence slips in because these are not pure measures of the environment ‘out there’ independent of us and our behaviour. We select, modify and even create our experiences in part on the basis of our genetic propensities. This means that correlations between such so-called ‘environmental’ measures and psychological traits cannot be assumed to be caused by the environment itself. In fact, genetics is responsible for half of these correlations. For example, what appears to be the environmental effect of parenting on children’s psychological development actually involves parents responding to their children’s genetic differences.

Location 54-60

For example, one remarkable discovery is that even most measures of the environment that are used in psychology – such as the quality of parenting, social support and life events – show significant genetic impact. How is this possible when environments have no DNA themselves? As we shall see, genetic influence slips in because these are not pure measures of the environment ‘out there’ independent of us and our behaviour. We select, modify and even create our experiences in part on the basis of our genetic propensities. This means that correlations between such so-called ‘environmental’ measures and psychological traits cannot be assumed to be caused by the environment itself. In fact, genetics is responsible for half of these correlations. For example, what appears to be the environmental effect of parenting on children’s psychological development actually involves parents responding to their children’s genetic differences.

Location 64-68

In fact, the environment makes siblings reared in the same family as different as siblings reared in separate families. Family resemblance is due to our DNA rather than to our shared experiences like TLC, supportive parenting or a broken home. What makes us different environmentally are random experiences, not systematic forces like families. The implications of this finding are enormous. Such experiences affect us, but their effects do not last; after these environmental bumps we bounce back to our genetic trajectory. Moreover, what look like systematic long-lasting environmental effects are often reflections of genetic effects, caused by us creating experiences that match our genetic propensities.

Location 117-122

However, the problem with the mantra ‘nature and nurture’ is that it runs the risk of sliding back into the mistaken view that the effects of genes and environment cannot be disentangled. No one has trouble accepting that the environment we experience contributes to who we are, but few people realize how important DNA differences are. My reason for focusing on DNA as the blueprint for making us who we are is that we now know that DNA differences are the major systematic source of psychological differences between us. Environmental effects are important but what we have learned in recent years is that they are mostly random – unsystematic and unstable – which means that we cannot do much about them.

Location 118-118

nurture’ is that it runs the risk of sliding back into the mistaken

Location 118-122

nurture’ is that it runs the risk of sliding back into the mistaken view that the effects of genes and environment cannot be disentangled. No one has trouble accepting that the environment we experience contributes to who we are, but few people realize how important DNA differences are. My reason for focusing on DNA as the blueprint for making us who we are is that we now know that DNA differences are the major systematic source of psychological differences between us. Environmental effects are important but what we have learned in recent years is that they are mostly random – unsystematic and unstable – which means that we cannot do much about them.

Location 129-133

I hope this book conveys the excitement I feel about this historic moment in psychology. The message from earlier research has begun to sink in, that DNA is the major systematic force, the blueprint, that makes us who we are. The implications for our lives – for parenting, education and society – are enormous. However, this only sets the stage for what will be the main event: the ability to predict our psychological problems and promise from DNA. This is the turning point when DNA changes psychology – scientifically and clinically – and the impact of psychology on our lives. Our future is DNA.

Location 143-146

For most of the twentieth century it was assumed that psychological traits were caused by environmental factors. These environmental factors were called nurture because, from Freud onwards, their origins were thought to lie in the family environment. Because these traits run in families, it was reasonable to assume that the family environment is responsible for these traits.

Location 229-233

Heritability is frequently misunderstood. For example, it is not a constant like the speed of light or gravity. It is a statistic that describes a particular population at a particular time with that population’s particular mix of genetic and environmental influences. A simpler way of expressing this is that it describes what is but does not predict what could be. Another population, or the same population at a different time, could have a different mix of genetic and environmental influences. Heritability will reflect these differences.

Location 330-334

Weight can run in families for reasons of nature (genetics) or nurture (environment). For a century, genetic research has relied on two methods to disentangle nature and nurture: the adoption method and the twin method. The two methods have different assumptions, strengths and weaknesses. Despite the great differences in the two methods, the results of adoption and twin studies converge on the same conclusion about the importance of inherited DNA differences in the origins of psychological traits.

Location 338-341

Adoption studies also provide a direct test of nurture. If nurture is why weight runs in families, adopted children should resemble their adoptive parents, who are their ‘environmental’ parents. Just like parents who rear their genetic children, adoptive parents provide their children’s family environment, including the food they eat, and model healthy or unhealthy lifestyles.

Location 342-345

Therefore, an even better test of the influence of family environment is to study ‘environmental’ siblings. About a third of adoptive families adopt two children. These children have different biological parents and are not genetically related but they grow up in the same family. If nurture explains individual differences in weight, adoptive siblings should be just as similar as siblings who share both nature and nurture.

Location 350-354

The adoption design is particularly powerful in disentangling the influence of nature and nurture because it can include ‘genetic’ parents, ‘environmental’ parents and ‘genetic-plus-environmental’ parents. ‘Genetic’ parents are birth parents of adopted-away children, and ‘environmental’ parents are the adoptive parents of these children. ‘Genetic-plus-environmental parents’ refers to the usual situation in which parents share both nature and nurture with their children. This design enables powerful estimates of genetic and environmental influence.

Location 487-490

Like CAP, the most important contribution of TEDS is its role in discovering the ‘big findings’ described in the following chapters. For example, TEDS took the lead in showing that what we call disorders are not genetically distinct from the normal range of variation. Although it might not sound very exciting, this finding has far-reaching implications for clinical psychology because it means that there are no disorders, that the ‘abnormal is

Location 545-550

The heritability estimate from TEDS is about 60 per cent, but the estimate from all research is 80 per cent. Why do these two estimates of heritability differ? The answer is an example of another of the ‘big findings’ of genetic research: heritability increases during development. Twins in TEDS are adolescents, but most other twin studies involve adults. In an analysis of forty-five twin studies, the heritability of weight increases from about 40 per cent in early childhood to about 60 per cent in adolescence to about 80 per cent in adulthood. The heritability estimate of 60 per cent from the adolescent twins in TEDS is just what would be expected. When we study the TEDS twins later in adulthood, we will expect to find a heritability estimate closer to 80 per cent.

Location 611-615

A large effect explains 25 per cent of the variance, an effect so large you would stumble over it in the dark. There are very few large effect sizes in psychology. One example is that general intelligence accounts for about 25 per cent of the variance in educational achievement. On this scale from small (1%) to medium (10%) to large (25%) effect sizes, heritability of 50 per cent is literally way off the scale. Inherited DNA differences are by far the most important systematic force in making us who we are.

Location 676-679

When we think about nurture, images come to mind like parents cooing to and cuddling their babies. Freud thought that parenting is the essential ingredient in a child’s development. He focused on specific aspects of parenting, including breastfeeding and toilet-training, and how they affect sexual identity. He wrote persuasively about clinical case studies that supported his ideas, but he provided no real data.

Location 710-712

What if we analysed environmental measures in a genetic design like a twin study? It seemed like a silly thing when I first did this in the 1980s because environmental measures should not show any genetic influence – after all, they are environmental measures. Or are they? This was how the nature of nurture phenomenon was first discovered.

Location 727-731

How can stressful life events show genetic influence? The questionnaire used in this study combined perceptions of whether an event occurred and how you respond to the event. Genetic influence on personality can affect both these perceptions. People differ in what they are willing to call a serious illness or injury, financial difficulty or relationship breakdown. Personality is especially involved in how much they feel these events affected them. Optimists might see these experiences through rose-coloured glasses, while pessimists see them in shades of grey. What about

Location 727-730

How can stressful life events show genetic influence? The questionnaire used in this study combined perceptions of whether an event occurred and how you respond to the event. Genetic influence on personality can affect both these perceptions. People differ in what they are willing to call a serious illness or injury, financial difficulty or relationship breakdown. Personality is especially involved in how much they feel these events affected them. Optimists might see these experiences through rose-coloured glasses, while pessimists see them in shades of grey.

Location 731-737

What about stressful events themselves, free of perception? Divorce is an example of an objective event and one of the most stressful life events for most people. The first genetic study of divorce caused a stir. In a study of 1,500 pairs of adult twins, concordance for divorce was much greater for identical than for fraternal twins (55 per cent versus 16 per cent), suggesting substantial genetic influence on divorce. USA Today called this study ‘the epitome of asinine’ because it seemed preposterous to conclude that divorce is influenced by genetic factors. But is it ‘the epitome of asinine’ to think that the objective event of divorce could be influenced by our genetically rich differences in personality? To the contrary, I think it is unreasonable to assume that events like divorce are just things that happen to us, as if we have nothing to do with them.

Location 751-753

So, divorce doesn’t just happen by chance. We make or break our relationships. We are not just passive bystanders at the whim of events ‘out there’. As always, genetic influence means just that – influence, not hard-wired genetic determinism. There are no schlimazel (Yiddish for ‘crooked luck’) genes that attract life’s pies in the face.

Location 781-783

But even these parent–child results suggested substantial genetic influence. Non-adoptive parents and their children were significantly more similar (0.30) in how much television they watched than were adoptive parents and their adopted children (0.15).

Location 832-836

What’s new is that these studies have greatly extended the list of environmental measures that show genetic influence. For example, evidence for genetic influence has been found for home environments such as chaotic family environments, for classroom environments such as supportive teachers, peer characteristics such as being bullied, neighbourhood safety, being exposed to drugs, work environments and the quality of one’s marriage. Results showing genetic influence are not limited to the classic twin design. They also emerged from studies of twins reared apart, other adoption designs and, most recently, from DNA studies.

Location 836-840

Characteristics of adolescents’ peer groups are especially highly heritable, such as the peer group’s academic orientation or their delinquency. The reason for this high heritability may be that you can choose your friends but you cannot choose your family, as Harper Lee wrote in To Kill a Mockingbird. You passively share genes with your parents and siblings, which leads to correlations between genes and your family experiences. With friends, you can select individuals similar to you genetically, actively creating correlations between your genes and your experiences with friends.

Location 853-860

For quality of support, we found that a third of the differences between people could be explained by genetic factors, but quantity of support showed no significant genetic influence. Why would quality of support show genetic influence but not quantity of support? In our paper describing these results we suggested that the answer might be that quality seems more subjective than quantity. More subjective measures catch genetic influence as perceptions filter through people’s personality, memories and motivation. That’s just a guess, though, and we still don’t know why quality of support is more heritable than quantity of support. And this might be different now, with the prominence of social media, which seems more a matter of quantity than quality. A recent twin analysis showed that individual differences in the use of Facebook in young adults yielded a heritability of 25 per cent, although quantity and quality of social support were not distinguished.

Location 868-873

It is more useful to phrase the question in the language of individual differences, which is the bailiwick of behavioural genetics. Why do some people live in warm and sunny climes and others tolerate cold, wet places? One answer is that, although we cannot control the weather, we can control where we live. If you love being outside, or if you have seasonal affective disorder, you can consider moving to a climate that suits you. Being outdoorsy or being prone to depression is influenced in part by genetic factors. Moving to a climate that suits you is one way in which genetic differences could contribute to individual differences in responses to straightforward questions about the weather such as ‘How often does the sun shine where you live?’ You might live in a sunny place because you chose to live there.

Location 896-898

The importance of measures of the environment lies in their psychological impact. If genes affect environmental measures as well as psychological measures, this raises the possibility that genes also contribute to correlations between them.

Location 944-949

Despite the difficulties in summarizing these studies systematically, they tell a simple story. This story is the same as that told by the original 1985 CAP paper and the 1990 SATSA paper: Genetics typically accounts for about half of the correlation between environmental measures and psychological traits. This finding about the nature of nurture is one of the most unexpected and important examples of how DNA makes us who we are. Instead of assuming that correlations between the ‘environment’ and psychological traits are caused environmentally, it is safer to assume that half of the correlation is due to genetic differences between people. This research is also important because it shows how we can study ‘true’ environmental effects controlling for genetics. This will be a major direction for research as the DNA revolution takes hold.

Location 949-957

The nature of nurture suggests a new way of thinking about experience. In the past, psychologists assumed that the environment is what happens to us passively, but genetic research on the nature of nurture suggests a more active model of experience. Psychological environments are not ‘out there’, imposed on us passively. They are ‘in here’, experienced by us as we actively perceive, interpret, select, modify and even create environments correlated with our genetic propensities. Our genetically rich differences in personality, psychopathology and cognitive ability make us experience life differently. For example, genetic differences in children’s aptitudes and appetites affect the extent to which they take advantage of educational opportunities. Genetic differences in our vulnerability to depression affect the extent to which we interpret experiences positively or negatively. This is a general model for thinking about how we use the environment to get what our DNA blueprint whispers that it wants. This is the essence of the nature of nurture.

Location 950-956

about experience. In the past, psychologists assumed that the environment is what happens to us passively, but genetic research on the nature of nurture suggests a more active model of experience. Psychological environments are not ‘out there’, imposed on us passively. They are ‘in here’, experienced by us as we actively perceive, interpret, select, modify and even create environments correlated with our genetic propensities. Our genetically rich differences in personality, psychopathology and cognitive ability make us experience life differently. For example, genetic differences in children’s aptitudes and appetites affect the extent to which they take advantage of educational opportunities. Genetic differences in our vulnerability to depression affect the extent to which we interpret experiences positively or negatively. This is a general model for thinking about how we use the environment to get what our DNA blueprint whispers

Location 962-965

From this perspective, one of the big findings from behavioural genetic research is counterintuitive: genetic influences become more important as we grow older. No psychological trait shows less genetic influence with age, but the domain where heritability increases most dramatically during development is cognitive ability.

Location 965-968

There are many types of cognitive abilities – for example, verbal and spatial – but in fact you are more likely to have one if you have the other. People with higher ability for memory, say, tend to have higher ability for all the other forms of intelligence. People often think they are good at either literature or maths, for example, but in fact they are more likely to be good at both if they are naturally skilled in one, although there are exceptions.

Location 971-975

According to the majority view of intelligence researchers, the core of intelligence is ‘the ability to reason, plan, solve problems, think abstractly, comprehend complex ideas, learn quickly, and learn from experience’. Intelligence is important scientifically and socially. Scientifically, intelligence reflects how the brain works, not as specific modules that light up in brain-imaging studies, but as brain processes working in concert to solve problems. Socially, intelligence is one of the best predictors of educational achievement and occupational status.

Location 976-980

During the past century, genetic research on intelligence was in the eye of the storm of the nature–nurture debate in the social sciences. The debate was driven by misplaced fears about biological determinism, eugenics and racism. This controversy raised the threshold for acceptance of the importance of genetics. Genetic research exceeded this threshold with bigger and better studies stockpiling evidence consistently showing that genetic differences between people account for about half of their differences in tests of intelligence. This general estimate of 50 per cent heritability masks an intriguing finding, which is how heritability changes over the course of our lives.

Location 1003-1003

Location 1009-1011

The heritability of 50 per cent for intelligence is just the lifetime average across all studies. The impressive increase in heritability from 20 per cent in infancy to 40 per cent in childhood to 60 per cent in adulthood stands out from other traits that show little developmental change in heritability, most notably personality and school achievement.

Location 1039-1043

This type of analysis shows that genetic effects on intelligence are highly stable from age to age. For example, in TEDS, genetic effects on intelligence in Year 2 correlate 0.7 with genetic effects on intelligence in Year 4. Genetic correlations from age to age are even greater after childhood. A recent DNA study strongly supports these results from twin studies, finding 90 per cent overlap in the genes that affect intelligence in childhood and adulthood.

Location 1051-1053

like the idea that we grow into our genes. The older we get, the more we become who we are genetically. To some extent, especially for cognitive ability, this means we become more like our parents as we age. Perhaps this is why people, as they get older, often seem to fear that they are becoming just like their parents.

Location 1061-1064

Genetic research shows that the medical model is all wrong when it comes to psychological problems. What we call disorders are merely the extremes of the same genes that work throughout the normal distribution. That is, there are no genes ‘for’ any psychological disorder. Instead, we all have many of the DNA differences that are related to disorders. The salient question is how many of these we have. The genetic spectrum runs from a few to a lot, and the more we have, the more likely we are to have problems.

Location 1079-1083

This research indicates that the same genes are responsible for reading disability and reading ability. Similar results have been found for other psychological disorders, suggesting that there are no genes for psychological disorders – they are the same genes responsible for heritability throughout the normal distribution, from those few people with very low genetic risk to the many people with average genetic risk to the few people with very high genetic risk.

Location 1113-1120

This is exactly the way genetic influence works for all common disorders. Later, we will consider polygenic scores comprised of thousands of DNA differences identified by their association with psychological disorders. The point here is that these polygenic scores are always perfectly normally distributed, meaning that they predict variation throughout the distribution – from people who are hardly ever depressed to those who sometimes get depressed to people who are chronically depressed. These polygenic scores predict whether someone is diagnosed as depressed or not only because these people are at the extreme of the normal distribution of genetic risk. The abnormal is normal in the sense that we all have many of the DNA differences that contribute to the heritability of any psychological disorder. Whether or not we reach some arbitrary diagnostic cut-off depends on how many of these DNA differences we have.

Location 1120-1125

This genetic research leads to a momentous conclusion: There are no qualitative disorders, only quantitative dimensions. Psychological problems like depression, alcohol dependence and reading disability are serious. The more extreme the problem, the more likely it is to affect the individual, their family and society. But because the genetic risk is continuous, it makes no sense to try to reach a decision about whether someone ‘has’ the disorder or not. There is no disorder – just the extremes of quantitative dimensions. People differ in how depressed they are, how much alcohol they consume and how well they read, but these problems are part of the normal distribution. A shift in vocabulary is needed so that we talk about ‘dimensions’ rather than ‘disorders’.

Location 1169-1171

As we will see later, the DNA revolution will put this issue front and centre in clinical psychology and psychiatry. The polygenic scores that predict genetic liability for ‘disorders’ are perfectly normally distributed. Therefore, we can, for the first time, investigate individuals at the ‘other end’ of the normal distribution of polygenic scores to find out who they are.

Location 1188-1193

One of the first hints came from research that showed that generalized anxiety disorder and major depressive disorder are the same thing genetically. Inherited DNA differences contribute substantially to your risk of being anxious or depressed but they do not specify whether you will be diagnosed as anxious or depressed. Whether you become anxious or you become depressed is caused by environmental factors. In other words, genetic risks are general across disorders; environmental risks are specific to a disorder. Generalist genes are not limited to cases diagnosed with disorders. The same result emerged from two dozen twin studies that looked at the genetic overlap between dimensions of anxiety symptoms and dimensions of depression symptoms.

Location 1193-1198

Hundreds of studies later, the genetic architecture of psychopathology suggests just three broad genetic clusters, in contrast to the dozens of disorders in psychologists’ diagnostic manuals. One cluster includes problems like anxiety and depression, which are called internalizing problems because they are directed inward. The second genetic cluster, externalizing problems, includes problems in conduct and aggressiveness in childhood, and, in adulthood, antisocial behaviour, alcohol dependence and other substance abuse. Psychotic experiences such as hallucinations and other extreme thought disorders form the third genetic cluster, which includes schizophrenia, bipolar disorder and major depression.

Location 1203-1209

Most severe mental illnesses, or psychoses, show the effects of generalist genes. The first branching point in diagnosing psychoses separates schizophrenia and depressive disorders. This dividing point is so enshrined in diagnoses that the two diagnoses were until recently viewed as mutually exclusive. That is, if you were diagnosed as schizophrenic, you could not be diagnosed with bipolar disorder, a severe form of depression that alternates with mania. For this reason, it was a complete surprise to find that most DNA differences found to be associated with schizophrenia also showed associations with bipolar disorder, as well as with major depression and other disorders. Even though schizophrenia, bipolar disorder and major depressive disorder are the oldest and most consistently diagnosed disorders, they show the greatest genetic overlap. This means that we are going to have to tear up our diagnostic manuals based on symptoms.

Location 1213-1218

Generalist genes are not limited to the domain of psychopathology. Most genetic effects are also general across cognitive abilities. For example, cognitive abilities such as vocabulary, spatial ability and abstract reasoning yield genetic correlations greater than 0.5, even though these abilities are thought to involve very different neurocognitive processes. That is, when we find a DNA difference associated with one cognitive ability, there is a greater than 50 per cent chance that it will also be associated with other cognitive abilities. Some genetic effects are specific to each cognitive ability, but the surprise is that most genetic effects are general to all cognitive abilities.

Location 1220-1221

Education-related skills such as reading, mathematics and science show even higher genetic correlations: about 0.7.

Location 1243-1248

The generalist-genes model makes more sense genetically and evolutionarily than the traditional modularity model. There are two great principles of genetics as they affect complex psychological traits like psychopathology and cognitive abilities as well as neurocognitive traits involving brain structure and function. First, genetic influence is caused by thousands of DNA differences of extremely small effect size; this is called polygenicity. Second, each DNA difference affects many traits; this is called pleiotropy. Given polygenicity and pleiotropy, it seems likely that generalist genes result in generalist brains. It also makes sense

Location 1243-1248

The generalist-genes model makes more sense genetically and evolutionarily than the traditional modularity model. There are two great principles of genetics as they affect complex psychological traits like psychopathology and cognitive abilities as well as neurocognitive traits involving brain structure and function. First, genetic influence is caused by thousands of DNA differences of extremely small effect size; this is called polygenicity. Second, each DNA difference affects many traits; this is called pleiotropy. Given polygenicity and pleiotropy, it seems likely that generalist genes result in generalist brains.

Location 1261-1265

The second big finding about the environment also began by bumping into something odd: Why are children who grow up in the same family so different? One sibling might be an extravert, the other withdrawn; one may be better at school than the other. We now know that genetics makes siblings 50 per cent similar, which means it also makes them 50 per cent different. But before genetics was taken seriously, it was a puzzle why children growing up in the same family, with the same parents, living in the same neighbourhood and going to the same school should be so different.

Location 1288-1290

The astonishing implication from this research is that we would be just as similar to our parents and our siblings even if we had been adopted apart at birth and reared in different families. As unbelievable as this might seem, as we shall see, adoption research shows that this is literally true.

Location 1301-1305

Twin studies come to the same conclusion that growing up in the same family does not make family members similar in weight, unless they share genes. Twin studies estimate heritability of weight as 80 per cent, even though all the genetic data together estimate heritability as 70 per cent. Identical twins correlate 0.8, which means that genetic similarity completely accounts for their similarity in weight. Fraternal twins correlate 0.4, which is exactly what would be expected if heritability is 80 per cent, because fraternal twins are only 50 per cent similar genetically.

Location 1316-1319

Model-fitting analyses that put all the data together consistently find that experiences shared by family members have no effect on individual differences. Family members are similar for all psychological traits but for genetic reasons. Growing up with a sibling does not make you similar to them beyond the similarity due to genetics.

Location 1316-1318

Model-fitting analyses that put all the data together consistently find that experiences shared by family members have no effect on individual differences. Family members are similar for all psychological traits but for genetic reasons. Growing up with a sibling does not

Location 1328-1333

This finding about the importance of non-shared environment was ignored when it was first noted in 1976 in relation to personality. It was controversial when I first reviewed the genetic research pointing to this phenomenon in 1987, and again in 1998, when a popular book tackled the topic. But now the finding is so widely accepted, at least among behavioural geneticists, that attention has switched to finding any shared environmental influence at all. For instance, delinquency in adolescence shows some shared environmental influence, meaning that you might be more likely to get into bad behaviour if your sibling does, although even here most of the environmental influence is non-shared.

Location 1338-1341

Could it be that the importance of shared environment for intelligence drops out by adolescence? Subsequent studies of older adoptive siblings have found similarly low correlations for intelligence. The most impressive evidence comes from a ten-year longitudinal follow-up study of adoptive siblings. At the average age of eight, the adoptive siblings correlated 0.25 for intelligence. Ten years later, the same adoptive siblings correlated 0.

Location 1345-1352

School achievement is another apparent exception to the rule. Tests of school achievement in all subjects from science to the humanities typically estimate that 20 per cent of the variance in performance can be explained by shared environment. Does the effect of shared environment on school achievement diminish after adolescence, as it does for intelligence? The first genetic research on educational achievement at university suggests this might be the case. Shared environment had no effect on performance in STEM subjects (science, technology, engineering and mathematics) and accounted for only 10 per cent of the variance on performance in humanities subjects. The only other exceptions from the hundreds of traits that have been investigated are some religious and political beliefs, for which shared environment accounts for about 20 per cent of the variance.

Location 1361-1365

Summarizing a huge literature, siblings living in the same family have very different experiences. It is almost as if siblings are living in different families, especially when it comes to their perceptions of how differently they are treated by their parents. Early research focused on parents and siblings. In retrospect, it seems odd that so much research looking for factors that make family members different would focus on the family. Looking outside the family – school, peers, friends, for example – would seem a better bet for finding factors that make siblings different.

Location 1369-1373

The third step takes on board the nature-of-nurture phenomenon. ‘Environmental’ measures show genetic influence, and genetics typically accounts for about half of the correlation between environmental measures and psychological traits. In other words, siblings might be treated differently because they differ genetically. For example, differences in how negative parents are towards their children might be an effect rather than a cause of a child’s depression. Very few candidates for non-shared environment are left at this third step.

Location 1378-1380

Differences in parental negativity towards their children related to the children’s differences in depression as well as to differences in antisocial behaviour.

Location 1381-1384

The first report of this phenomenon came from NEAD, showing that genetics was largely responsible for the association between differences in parental negativity towards their children and the children’s differences in their likelihood of becoming depressed or engaging in antisocial behaviour. In other words, parents’ negativity was a response to, rather than a cause of, their children’s depression and antisocial behaviour.

Location 1382-1384

association between differences in parental negativity towards their children and the children’s differences in their likelihood of becoming depressed or engaging in antisocial behaviour. In other words, parents’ negativity was a response to, rather than a cause of, their children’s depression and antisocial behaviour.

Location 1397-1401

It seems likely that the influence of non-shared environment comes from many experiences that each have a small effect. There might be so many experiences of such small effect that they are essentially idiosyncratic, meaning that it comes down to chance. Sometimes chance is writ large, as in the case of major illnesses or accidents or war experiences that dramatically alter the course of an individual’s development. More surprising are the often seemingly trivial chance events that launch lives in slightly different directions with cascading effects as time goes by.

Location 1397-1400

It seems likely that the influence of non-shared environment comes from many experiences that each have a small effect. There might be so many experiences of such small effect that they are essentially idiosyncratic, meaning that it comes down to chance. Sometimes chance is writ large, as in the case of major illnesses or accidents or war experiences that dramatically alter the course of an individual’s development. More surprising are the often seemingly trivial chance events that launch lives in slightly

Location 1397-1400

It seems likely that the influence of non-shared environment comes from many experiences that each have a small effect. There might be so many experiences of such small effect that they are essentially idiosyncratic, meaning that it comes down to chance. Sometimes chance is writ large, as in the case of major illnesses or accidents or war experiences that dramatically alter the course of an individual’s development. More surprising are the often seemingly trivial chance events that launch lives in slightly

Location 1419-1421

No identical twin differences have been shown to be stable over several years, which would be necessary if non-shared environment had enduring effects. This means that the non-shared environmental factors that make identical twins different are not stable. They are like random noise.

Location 1449-1453

The environment accounts for the other half of the variance but, as we saw in the previous chapter, it is not the environment as we had understood it that was important. We know that the environment makes a difference because heritability is only about 50 per cent, but the salient environmental influences are not the shared, systematic and stable effects psychologists had assumed were important in development. Moreover, research on the nature of nurture has demonstrated that what look like environmental effects are to a large extent really reflections of genetic differences.

Location 1456-1463

Parents obviously matter tremendously in their children’s lives. They provide the essential physical and psychological ingredients for children’s development. But if genetics provides most of the systematic variance and environmental effects are unsystematic and unstable, this implies that parents don’t make much of a difference in their children’s outcomes beyond the genes they provide at conception. We saw in the previous chapter that shared environmental influence hardly affects personality, mental health or cognitive abilities after adolescence. This even includes personality traits that seem especially susceptible to parental influence such as altruism, kindness and conscientiousness. The only exception from hundreds of traits that shows some evidence of shared environmental influence is religious and political beliefs. As a parent, you can make a difference to your child’s beliefs, but even here shared environmental influence accounts for only 20 per cent of the variance.

Location 1484-1492

The first caveat is that genetic research describes what is, not what could be. Parents can make a difference to their child but, on average in the population, parenting differences don’t make a difference in children’s outcomes beyond the genes they share. Parents differ in how much they guide their children in all aspects of development. They differ in how much they push their children’s cognitive development, for example in language and reading. Parents also differ in how much they help or hinder their children’s self-esteem, self-confidence and determination, as well as more traditional aspects of personality such as emotionality and sociability. But in the population, these parenting differences don’t make much of a difference in their children’s outcomes once genetics is taken into account. Over half of children’s psychological differences are caused by inherited DNA differences between them. The rest of the differences are largely due to chance experiences. These environmental factors are beyond our control as parents. As we saw in the previous chapter, we don’t even know what these factors are.

Location 1492-1495

The second caveat is that genetic research describes the normal range of variation, genetically and environmentally. Its results do not apply outside this normal range. Severe genetic problems such as single-gene or chromosomal problems or severe environmental problems such as neglect or abuse can have devastating effects on children’s cognitive and emotional development. But these devastating genetic and environmental events are, fortunately, rare and do not account for much variance in the population.

Location 1504-1508

In the tumult of daily life parents mostly respond to genetically driven differences in their children. This is the source of most correlations between parenting and children’s outcomes. We read to children who like us to read to them. If they want to learn to play a musical instrument or play a particular sport, we foster their appetites and aptitudes. We can try to force our dreams on them, for example, that they become a world-class musician or a star athlete. But we are unlikely to be successful unless we go with the genetic grain. If we go against the grain, we run the risk of damaging our relationship with our children.

Location 1509-1510

Genetics provides an opportunity for thinking about parenting in a different way. Instead of trying to mould children in our image, we can help them find out what they like to do and what they do well. In other words, we can help them become who they are.

Location 1510-1511

Remember that your children are 50 per cent similar to you.

Location 1515-1517

Most importantly, parents are neither carpenters nor gardeners. Parenting is not a means to an end. It is a relationship, one of the longest lasting in our lives. Just as with our partner and friends, our relationship with our children should be based on being with them, not trying to change them.

Location 1525-1528

But our focus is on individual differences. Children differ a lot in how well they do at school. How much do differences in children’s school achievement depend on which school they go to? The answer is not much. This conclusion follows from direct analyses of the effect of schools on differences in students’ achievement and is especially true when we control for genetic effects.

Location 1538-1540

This does not mean that the quality of teaching and support offered by schools is unimportant. It matters a lot for the quality of life of students, but it doesn’t make a difference in their educational achievement.

Location 1541-1545

This reflects the confusion between average differences and individual differences. Average differences between schools in the league tables mask a wide range of individual differences within schools, meaning that there is considerable overlap in the range of performance between children in the best and worst schools. In other words, some children in the worst schools outperform most children in the best schools. The biggest average difference in achievement is between selective and non-selective schools. We will look at this issue later, when we examine selection in education and occupation, which raises issues of meritocracy and social mobility.

Location 1550-1552

Environmental influence shared by children attending the same schools as well as growing up in the same family accounts for only 20 per cent of the variance of achievement in the school years and less than 10 per cent of academic performance at university.

Location 1556-1563

No specific policy implications necessarily follow from finding that inherited DNA differences are by far the most important source of individual differences in school achievement and that schools make so little difference. Similar to the message for parents, genetic research suggests that teachers are not carpenters or gardeners in the sense of changing children’s school performance. Rather than frenetic teaching in an attempt to make pupils pass the tests that will improve their standing in league tables, schools should be supportive places for children to spend more than a decade of their lives, places where they can learn basic skills like literacy and numeracy but also learn to enjoy learning. To paraphrase John Dewey, the major American educational reformer of the twentieth century, education is not just preparation for life – education is a big chunk of life itself.

Location 1568-1570

But they don’t change who we are psychologically – our personality, our mental health and our cognitive abilities. Life experiences matter and can affect us profoundly, but they don’t make a difference in terms of who we are.

Location 1571-1576

Individual differences in stressful life events were among the first environmental measures for which genetic influence was found. Most research on life events used self-report measures of stressful events and their effects. However, we saw that even objectively measured events such as divorce show genetic influence. Parental divorce is the best predictor of children’s divorce, but this correlation, easily interpreted as environmental, is entirely due to genetics. Quality of social support is another major aspect of life experiences that has been assumed to be a source of environmental influence

Location 1576-1578

Finding genetic influence on individual differences in ‘environmental’ measures led to research that showed that genetics accounts for about half of the correlations between life experiences and psychological traits, such as the correlation between perceptions of life events and depression. This is another example of the nature of nurture.

Location 1582-1589

The importance of non-shared environment has major implications as well for understanding why life experiences don’t make a difference psychologically. The heritability of life experiences is about 25 per cent, which means that most of the individual differences in life experiences are environmental in origin. But these environmental influences are not shared by our siblings, even if our sibling is our identical twin. Our parents cannot take much credit or blame for how we turned out, other than via the genes they gave us. No one can take credit or blame because these non-shared environmental influences are unsystematic and unstable. Beyond the systematic and stable force of genetics, good and bad things just happen. As mentioned earlier in relation to parenting, the good news is that these random experiences don’t matter much in the long run because their impact is not long-lasting. We eventually rebound to our genetic trajectory. To the extent that our experiences appear shared, systematic and stable, they reflect our genetic propensities.

Location 1589-1592

In summary, parents matter, schools matter and life experiences matter, but they don’t make a difference in shaping who we are. DNA is the only thing that makes a substantial systematic difference, accounting for 50 per cent of the variance in psychological traits. The rest comes down to chance environmental experiences that do not have long-term effects.

Location 1595-1597

Anecdotes are not enough, and it’s not enough to show a statistically significant effect – the issue is whether these things explain more than 1 or 2 per cent of the variance. I am not worried about the conclusion being falsified, because there is a century of research behind it. One general

Location 1595-1597

Anecdotes are not enough, and it’s not enough to show a statistically significant effect – the issue is whether these things explain more than 1 or 2 per cent of the variance. I am not worried about the conclusion being falsified, because there is a century of research behind it.

Location 1597-1601

One general message that should emerge from these discoveries is tolerance for others – and for ourselves. Rather than blaming other people and ourselves for being depressed, slow to learn or overweight, we should recognize and respect the huge impact of genetics on individual differences. Genetics, not lack of willpower, makes some people more prone to problems such as depression, learning disabilities and obesity. Genetics also makes it harder for some people to mitigate their problems. Success and failure – and credit and blame – in overcoming problems should be calibrated relative to genetic strengths and weaknesses.

Location 1601-1605

Going even further out on this limb, I’d argue that understanding the importance of genetics and the random nature of environmental influences could lead to greater acceptance and even enjoyment of who we are genetically. Rather than striving for an ideal self that sits on an impossibly tall pedestal, it might be worth trying to look for your genetic self and to feel comfortable in your own skin. Moreover, as we have seen, with age, as genetic influence increases, the more we become who we are.

Location 1606-1609

Heritability describes what is but does not predict what could be, as I have emphasized several times. High heritability of weight does not mean there is nothing you can do about your weight. Nor does heritability mean that we must succumb to our genetic propensities to depression, learning disabilities or alcohol abuse. Genes are not destiny. You can change.

Location 1615-1618

Anyone who is influenced by these maxims should understand that, to the contrary, genetics is the main systematic force in life. Again, this is not to say that genes are destiny. It just seems more sensible, when possible, to go with the genetic flow rather than trying to swim upstream. As W. C. Fields said, ‘If at first you don’t succeed, try, try again. Then quit. There’s no use being a damn fool about it.’

Location 1620-1623

If schools, parenting and our life experiences do not change who we are, what does this mean for society, especially for equality of opportunity and meritocracy? In particular, does it mean that the genetically rich will get richer and the poor poorer? Are genetic castes inevitable? What does this say about inequality? In this chapter, we will explore the implications of the counterintuitive findings discussed in the previous chapters.

Location 1626-1632

Although meritocracy sounds like an irresistibly good idea, both parts of the neologism ‘meritocracy’ are loaded with unpalatable connotations. The noun ‘merit’ refers to ability and effort but it also connotes value and worth. It is derived from the Latin word meritum meaning ‘worthy of praise’. The ‘-ocracy’ part of ‘meritocracy’ refers to power and governance. Putting these two components of meritocracy together with genetics implies that we are governed by a genetic elite whose status is justified by their ability and effort. Instead, it could be argued that people who got lucky by drawing a good genetic hand do not merit anything. Their luck at learning easily and getting satisfying jobs is its own reward. And who says we should be governed by genetic elites? The populist strain of politics around the world suggests a desire for the opposite.

Location 1636-1641

At first glance, genetics seems antithetical to equality of opportunity, violating the principle enshrined in the second sentence of the 1776 United States Declaration of Independence that all people are created equal. However, the American founders did not mean that all people are created identical. They were referring to ‘unalienable rights’, which include ‘life, liberty and the pursuit of happiness’. In less lofty terms, this means equal protection before the law and equal opportunity. But ‘equal’ does not mean identical. If everyone were identical, there would be no need to worry about equal rights or equal opportunity. The essence of democracy is that people are treated fairly despite their differences.

Location 1641-1644

The most important point about equality of opportunity from a genetic perspective is that equality of opportunity does not translate to equality of outcome. If educational opportunities were the same for all children, would their outcomes be the same in terms of school achievement? The answer is clearly ‘no’ because even if environmental differences were eliminated genetic differences would remain.

Location 1644-1651

What follows from this point is one of the most extraordinary implications of genetics. Instead of genetics being antithetical to equal opportunity, heritability of outcomes can be seen as an index of equality of opportunity. Equal opportunity means that environmental advantages and disadvantages such as privilege and prejudice have little effect on outcomes. Individual differences in outcomes that remain after systematic environmental biases are diminished are to a greater extent due to genetic differences. In this way, greater educational equality of opportunity results in greater heritability of school achievement. The higher the heritability of school achievement, the less the impact of environmental advantages and disadvantages. If nothing but environmental differences were important, heritability would be 0. Finding that heritability of school achievement is higher than for most traits, about 60 per cent, suggests that there is substantial equality of opportunity.

Location 1652-1656

To the extent that environmental influences are non-shared, this means that they are not caused by systematic inequalities of opportunity. However, as we have seen, genetic research on primary- and secondary-school achievement is an exception to the rule that environmental influences are non-shared. For school achievement, half of the environmental influence – 20 per cent of the total variance – is shared by children attending the same school. This finding implies that up to 20 per cent of the variance in school achievement could be due to inequalities in school or home environments, although this effect mostly washes out by the time children go to university.

Location 1658-1666

For example, the socioeconomic status of parents is correlated with their children’s educational and occupational outcomes. This correlation has been interpreted as if it is caused environmentally. That is, better-educated, wealthier parents are assumed to pass on privilege, creating environmentally driven inequality in educational opportunity and stifling what is called intergenerational educational mobility. Genetics turns the interpretation of this correlation upside down. Socioeconomic status of parents is a measure of their educational and occupational outcomes, which are both substantially heritable. This means that the correlation between parents’ socioeconomic status and their children’s outcomes is actually about parent–offspring resemblance in education and occupation. Phrased as ‘parent–offspring resemblance’, it should come as no surprise that genetics largely mediates the correlation. Parent–offspring resemblance is an index of heritability, and heritability is an index of equal opportunity. So, parent–offspring resemblance for education and occupation indicates social mobility rather than social inertia.

Location 1672-1676

The most important type is active gene–environment correlation. Children actively select, modify and create environments correlated with their genetic propensities. For example, genetic differences in children’s aptitudes and appetites affect the extent to which they take advantage of educational opportunities. This is why equal opportunities cannot be imposed on children to create equal outcomes. Genetic differences in aptitudes and appetites influence the extent to which children take advantage of opportunities. To a large extent, opportunities are taken, not given.

Location 1684-1688

It is worth reiterating that this genetic research describes the mix of genetic and environmental influences on individual differences in school achievement in specific samples at specific times. Most of the research comes from developed countries, especially Europe and the US, in the twentieth century. The results could be different for different countries in different times. Our focus here is on the effects of equal opportunity on individual differences in school achievement. As access to education broadens, heritability would be expected to increase.

Location 1708-1709

So it should come as no surprise that students in selective schools perform better than students in non-selective schools, because it is a self-fulfilling prophecy that the students selected by selective schools for their school achievement have higher GCSE scores.

Location 1709-1712

When we control for the factors that are used to select students the average difference in GCSE scores is negligible and overall GCSE variance explained by school type shrinks to less than 1 per cent. In other words, selective schools do not improve students’ achievement once we take into account the fact that these schools preselect students with the best chance of success.

Location 1728-1730

I hope it will help parents who cannot afford to pay for private schooling or move house to know that it doesn’t make much of a difference in children’s school achievement. Expensive schooling cannot survive a cost–benefit analysis on the basis of school achievement itself.

Location 1748-1753

Both occupational status and income are substantially heritable, about 40 per cent in more than a dozen twin studies in developed countries. This should not be surprising, because occupational status and income are related to educational attainment and intelligence, which are heritable traits. Similar to the argument we made for education, heritability is an index of meritocratic selection for occupational status and income, so we can conclude on the basis of substantial heritability that selection is considerably meritocratic. Unlike education, shared environmental influence for occupational status is negligible, which means that environmental influences are random and that most of the systematic effects on occupational status and income can be attributed to genetics.

Location 1770-1773

At first thought, it might seem that, given free rein, genetics will limit social mobility and calcify society into genetic castes, as happened in India, where for thousands of years mating was limited to members of the same caste. I would argue that this is not a problem in modern societies for two reasons. The first is simple: a lot of the environmental variation between us is not systematic. Random effects will not create stable castes.

Location 1773-1776

The second reason is that parents and offspring are only 50 per cent similar genetically. Their genetic similarity means that, on average, brighter parents have brighter children. But their 50 per cent genetic dissimilarity means that children of brighter parents will show a wide range of ability, including some children of lower-than-average ability. If you take pairs of individuals

Location 1773-1777

The second reason is that parents and offspring are only 50 per cent similar genetically. Their genetic similarity means that, on average, brighter parents have brighter children. But their 50 per cent genetic dissimilarity means that children of brighter parents will show a wide range of ability, including some children of lower-than-average ability. If you take pairs of individuals randomly, their average difference will be seventeen IQ points. First-degree relatives – parents and their offspring or siblings – differ by thirteen IQ points on average. This allows plenty of room to go down as well as up the ladder.

Location 1783-1787

Because children are 50 per cent similar genetically to their parents, genetics predicts that the children’s average IQ will regress halfway from their parents’ IQ to the population average. For example, parents with an average IQ of 130 are expected to have children whose average IQ is 115, regressing halfway back to the population average of 100. This reshuffling of DNA differences in the genetic lottery prevents the evolution of a rigid genetic caste system.

Location 1790-1791

As long as downward social mobility as well as upward social mobility occurs, we do not need to fear that genetics will lead to a rigid caste system.

Location 1791-1795

Even though most of the systematic differences between people are genetic in origin, this does not mean that we need to be fatalistic and accept the status quo. One reason, emphasized earlier, is that genetics describes what is – it does not predict what could be. You can beat the genetic odds. But it is not fatalistic to recognize that DNA matters and to appreciate genetic differences between our children and between ourselves. It seems only sane to suggest that, when you can, try to go with the grain of genetics rather than fight against

Location 1850-1856

In 1961 Francis Crick and Sydney Brenner began to crack the genetic code by showing that the code consists of a sequence of three rungs on the rope ladder (e.g., A-A-A or C-A-G or G-T-T), which is like a three-letter ‘word’. The four letters (A, C, G, T) taken three at a time yield sixty-four possible combinations. In the next few years the meaning of all sixty-four words in the DNA dictionary was gradually worked out. For example, A-A-A is one word, C-A-G is another and G-T-T is another. These words code for one of twenty amino acids. There are hundreds of amino acids but only twenty are produced from scratch by our DNA. For example, A-A-A codes for phenylalanine, C-A-G for valine, G-T-T for glutamine. Some three-letter words code for the same amino acid and some provide punctuation such as start and stop signals, using up all sixty-four words in the DNA dictionary.

Location 1856-1860

Why amino acids? Amino acids are the building blocks of proteins, which are integral to all that we are. Proteins are essential for the structure, function and regulation of our bodies, including neurons and neurotransmitters, which are the basic elements of our brain and who we are psychologically. The average protein contains a unique sequence of the 20 amino acids, varying from 50 to 2,000 amino acids in length. With 20 amino acids in any order in such long strings, there is a limitless variety of proteins. On average, each of our cells produces 2,000 different proteins.

Location 1863-1866

This model of DNA coding for amino acids is what the word ‘gene’ classically meant. However, we now know that DNA does much more than code for amino-acid sequences. Only 2 per cent of the human DNA sequence works like this; there are only 20,000 classical ‘genes’. The other 98 per cent of DNA was thought to be junk but is now known to have important functions, as I will describe later.

Location 1871-1873

We have 3 billion rungs in the double helix of DNA, which is called the genome. But the genome is not one continuous rope ladder with 3 billion rungs. It is broken up into twenty-three segments, or chromosomes, which vary in length from 50 million to 250 million rungs.

Location 1879-1882

For each pair of chromosomes, your sibling has a fifty-fifty chance of getting the same chromosome as you, which is why siblings are, on average, 50 per cent similar. The exception is identical twins, who have exactly the same chromosomes because they come from the same fertilized egg. This is why siblings are similar but also different in terms of psychological traits and why identical twins are more similar than other siblings.

Location 1885-1887

As new cells are formed, the double helix unzips and each strand of the rope ladder seeks its complement for each rung. This duplication process is incredibly reliable, but mistakes are made – mutations – which are like typos in the genetic code. When a mutation occurs in egg or sperm, it is passed on to offspring, who then pass it on to their offspring.

Location 1887-1892

All kinds of differences in DNA sequence can occur, but the most common is when a single rung differs between people. A change in one of the 3 billion rungs in the double helix of DNA is called a single-nucleotide polymorphism (SNP, pronounced ‘snip’). You and I have about 4 million SNPs but many of these are present only in a few people, which means that we do not have the same 4 million SNPs. There may be as many as 80 million SNPs in the world. Any particular population – the UK, for example – has about 10 million SNPs. The rest of this book focuses on SNPs, because they have played a central role in the DNA revolution.

Location 1892-1896

All we inherit is the DNA sequence in the single cell with which our lives began, with its unique combination of DNA differences. Although all cells have the same DNA, cells express only a small portion of all DNA. Different types of cell – for example, brain, blood, skin, liver and bone cells – express different bits of DNA. DNA sequence is transcribed by a messenger molecule called RNA. RNA is then translated into amino-acid sequences according to the genetic code. This process is what is meant by the term gene expression.

Location 1896-1900

Many mechanisms affect gene expression. Some are long-term mechanisms (called epigenetic) that involve adding molecules to the DNA that prevent its transcription. Other mechanisms for expression have shorter-term effects. For example, proteins that interact with DNA regulate transcription in response to cues from the environment. You are changing the expression of many genes that code for neurotransmitters in your brain as you read this sentence. As the neural processes involved in reading deplete these neurotransmitters, you express the genes that code for these neurotransmitters in order to replenish them.

Location 1896-1900

Many mechanisms affect gene expression. Some are long-term mechanisms (called epigenetic) that involve adding molecules to the DNA that prevent its transcription. Other mechanisms for expression have shorter-term effects. For example, proteins that interact with DNA regulate transcription in response to cues from the environment. You are changing the expression of many genes that code for neurotransmitters in your brain as you read this sentence. As the neural processes involved in reading deplete these neurotransmitters, you express the genes that code for these neurotransmitters in order to replenish them.

Location 1904-1912

Let’s zoom in on one of the 10 million SNPs in the human genome. For reasons that will become clear, let’s focus on one SNP that happens to be in the middle of chromosome 16. Chromosome 16 has 90 million rungs in the double helix, and this SNP is at rung number 53,767,042. This mutation could have been A, C, T or G – but it happened to be T, until a mutation occurred long ago that changed T to A in one individual. The person with this mutation passed on this new A nucleotide to half of their offspring, who then passed it on to half of their offspring. After several generations, the new A nucleotide spread in the population. Perhaps its frequency increased because it conveyed some slight advantage evolutionarily, which is the case for this particular mutation, as we shall see. More often, its frequency increased because it didn’t have any effect and it just spread from generation to generation, following Mendel’s laws of inheritance. Today 40 per cent of all chromosomes have the A nucleotide at this spot on chromosome 16. The other 60 per cent has the original T nucleotide. These alternate forms of DNA sequence are called alleles.

Location 1912-1918

Because we inherit a pair of chromosomes, one from each parent, we have one allele from each parent. The pair of alleles is called our genotype. For the SNP on chromosome 16, we could inherit either an A allele or a T allele from our mother and an A or a T from our father. If we inherit an A allele from both parents, our genotype is AA. If we inherit an A from one parent and a T from the other, our genotype is AT. The third possibility results in a TT genotype. For this spot on chromosome 16, 15 per cent of us are AA, 50 per cent are AT and 35 per cent TT. Genotypes are just alleles considered two at a time, the way they are packaged in individuals. If you count the alleles in these genotype frequencies, you get the allele frequencies of 40 per cent A and 60 per cent T.

Location 1942-1947

In fact, our SNP is in a stretch of DNA in the FTO gene that does not code for proteins. It turns out that less than 2 per cent of the genome’s DNA sequence codes for proteins. These are the 20,000 classical genes mentioned earlier. Most mutations are in the other 98 per cent of DNA that does not code for a change in amino-acid sequence and used to be called ‘junk DNA’ because it is not translated into amino-acid sequences. Even within genes like the FTO gene, most of the DNA does not code for proteins. These non-coding stretches of genes, or introns, are spliced out of the RNA code before the RNA is translated into proteins. The remaining RNA segments, or exons, are spliced back together and proceed to be translated into amino-acid sequences.

Location 1949-1951

What we do know is that they do make a difference. Some research suggests that as much as 80 per cent of this non-coding DNA is functional, in that it regulates the transcription of other genes. This distinction is important because most DNA associations with psychological traits involve SNPs in non-coding regions of DNA rather than in classical genes.

Location 1957-1958

Pleiotropy and polygenicity mean that many DNA differences of small effect are likely to affect psychological traits – which is the case, as we shall see.

Location 1978-1982

Because each of us has two genomes, one from each parent, we have 6 billion nucleotide bases in our genome. If we knew the sequence of these 6 billion bases for many individuals, we could identify all of the inherited DNA differences, not just SNPs, that make a difference in psychological traits. This is now happening; it is called whole-genome sequencing. Rather than ‘just’ genotyping millions of SNPs, whole-genome sequencing works out the sequence of all 6 billion nucleotide bases.

Location 1989-1990

Rather than laboriously and expensively sequencing the whole genome of individuals, SNP microarrays were developed that focused on genotyping SNPs rather than sequencing the entire genome.

Location 1999-2001

SNP chips are now cheap, costing less than £50, and have been used to genotype millions of people for hundreds of thousands of SNPs across the genome. Until SNP chips became available, attempts to find DNA differences associated with psychological traits were limited to laboriously genotyping SNPs in a few ‘candidate’ genes thought to be important for a particular trait.

Location 2003-2006

SNP chips made it possible to scan the entire genome to identify SNPs associated with complex traits and common disorders, rather than just looking at a few candidate genes. This systematic approach is called genome-wide association (GWA). Genome-wide association studies kicked off the DNA revolution by providing the first effective tool to hunt for genes responsible for the heritability of psychological traits. We will join the hunt in the next chapter.

Location 2030-2034

The new approach was genome-wide association, which is the opposite of the candidate-gene approach. The dream was to look systematically across the genome rather than picking a few, somewhat arbitrary, candidate genes. To do this would require tens of thousands of SNPs genotyped for each of thousands of individuals. Although genotyping costs had gone down by then, it still cost about ten pence to genotype one DNA marker for one individual. So, genotyping ‘just’ 10,000 DNA markers one by one for 1,000 individuals would cost almost £1 million and a lot of time.

Location 2071-2074

This visionary big-science study, funded with £10 million from the Wellcome Trust and a dozen other UK agencies, was called the Wellcome Trust Case Control Consortium. Across the seven disorders, twenty-four genome-wide significant SNP associations were found, mostly for Type 2 diabetes and Crohn’s disease.

Location 2071-2077

This visionary big-science study, funded with £10 million from the Wellcome Trust and a dozen other UK agencies, was called the Wellcome Trust Case Control Consortium. Across the seven disorders, twenty-four genome-wide significant SNP associations were found, mostly for Type 2 diabetes and Crohn’s disease. This Wellcome Trust study was a cause for celebration because it showed that GWA studies with large sample sizes could be successful even for common disorders influenced by many DNA differences of small effect. One index of the importance of this paper is that it has been cited more than 5,000 times in other scientific papers. In addition, GWA won the ‘Breakthrough of the Year’ in 2007 awarded by Science.

Location 2091-2096

This new threshold of 80,000 cases motivated more researchers to collaborate, because they knew that their individual studies, usually with sample sizes of fewer than a thousand cases, had no power to detect associations of the size we now knew could be expected. In the biological and medical sciences more than a thousand GWA studies were reported in the five years following the Wellcome Trust study. Great progress was made during these five years, going from the twenty-four significant associations for seven traits from the Wellcome Trust study to more than 2,000 SNP associations for more than 200 traits. After five more years, in 2017, the number of genome-wide significant SNP associations had reached 10,000.

Location 2112-2116

What this means is that GWA hits are beginning to appear as studies of psychological disorders reach the power afforded by tens of thousands of cases. The results of GWA studies of psychological disorders confirm the daunting predictions from analyses of statistical power. With 10,000 cases, no significant associations are found. Significant associations begin to appear with 20,000 cases. Doubling the number of cases to 40,000 quadruples the number of significant hits. Doubling the sample size again to 80,000 shows another large increase in significant hits as power is reached to scoop up many of the smaller effects.

Location 2121-2123

As we shall see, our passport to this new world was the ability to aggregate the effects of many tiny associations to predict psychological differences, or polygenic scores. For schizophrenia, DNA differences packaged as polygenic scores are now the best predictor we have of who will become schizophrenic. The rest of this book is about these polygenic scores and their impact on psychology and society.

Location 2138-2139

Effects of this size are seen when the allele frequency for a SNP differs just slightly between cases and controls, for example, 45 per cent versus 40 per cent.

Location 2143-2147

One way around this problem is to study dimensions rather than disorders. Dimensions provide more power in GWA studies than disorders because every individual counts, whether they are low, middle or high in the distribution. In contrast, GWA studies of disorders look for average DNA differences between two groups, cases who are diagnosed with the disorder versus controls who do not have the disorder. This assumes that disorders are real, but this assumption clashes with one of the big findings of genetic research – that the abnormal is normal, meaning that there are no qualitative disorders, just quantitative dimensions.

Location 2163-2167

In the past two years there has been a surge of successful GWA studies of psychological dimensions. The first breakthrough was for an unlikely variable: years of education. In developed countries heritability of years of education is 50 per cent. Many psychological traits contribute to this heritability, such as previous achievement at school and cognitive abilities, which correlate 0.5 with years of education. The variable years of education is also affected by personality traits such as perseverance and conscientiousness, and mental health such as the absence of debilitating depression.

Location 2182-2188

The first successes have come in GWA studies of the two major dimensions of personality, extraversion and neuroticism, which twin studies indicate are about 40 per cent heritable. Extraversion includes sociability, impulsiveness and liveliness. Neuroticism, which refers to emotional instability rather than being neurotic, involves moodiness, anxiousness and irritability. For extraversion, a GWA study of 100,000 individuals found 5 hits. For neuroticism, over 100 hits were reported in a GWA study with a sample size of 300,000. A newer focus of personality research is a sense of well-being, basically happiness, which shows a similar heritability of 40 per cent in twin studies. In a GWA study of nearly 200,000 individuals, 3 hits were found.

Location 2191-2195

This is just the beginning of the DNA revolution. By the time you read this there will be dozens of bigger and better GWA studies of these and many other traits. An important source of new information will come from the biggest direct-to-consumer genomics company, 23andMe, with nearly 2 million paying customers. Eighty per cent of its customers have agreed to have their genotypes used in research and to consider follow-up requests for research. The average customer contributes to more than 200 brief studies, many of which are psychological studies.

Location 2200-2202

The GWA results tell a very different story. For complex traits, no genes have been found that account for 5 per cent of the variance, not even 0.5 per cent of the variance. The average effect sizes are in the order of 0.01 per cent of the variance, which means that thousands of SNP associations will be needed to account for heritabilities of 50 per cent.

Location 2204-2206

One certain boost will come from genotyping all DNA differences, not just those currently on SNP chips. SNP chips used in GWA studies rely on common SNPs, those with allele frequencies greater than 1 per cent in the population, whereas the vast majority of DNA differences in the genome are much less frequent than 1 per cent. Many inherited DNA differences are unique to an individual.

Location 2223-2225

Further complicating this bottom-up pathways approach from DNA to behaviour is pleiotropy, which, as we have seen, means that any DNA difference affects many traits. Pleiotropy guarantees that there is no clear path from genes to brain to behaviour.

Location 2233-2237

In contrast to the biologists’ bottom-up-pathways game plan is the psychologists’ top-down approach. For biologists, the ultimate goal of genetics is to understand every path between inherited DNA differences and individual differences in behavioural traits, a bottom-up approach. However, psychologists focus on behaviour and use genetics to understand behaviour. This top-down psychological perspective begins with prediction. We can use inherited DNA differences to predict individual differences in psychological traits without knowing anything about the myriad pathways connecting genes and behaviour.

Location 2239-2241

Although the effects of individual SNPs are tiny, these effects can be added like we add items on a test to create a composite score. In 2005 I called these SNP sets. There are now at least a dozen names for these composite scores, but they are generally called polygenic scores.

Location 2258-2263

Polygenic scores, based on DNA rather than crystal balls, are fortune tellers. As we shall see, prediction is crucial because it is the key to the prevention of psychological problems and the promotion of promise. This is the new world of personal genomics, which begins with the ability to use inherited DNA differences across the genome to predict psychological differences. For psychological dimensions and disorders, some polygenic scores have already reached impressive levels of predictive power. This chapter shows what a polygenic score is and describes the power of polygenic scores created in the past two years. It reveals some of my own polygenic scores to glimpse the future of psychological personal genomics.

Location 2273-2279

This is exactly how polygenic scores are created, except that, instead of items on a questionnaire, we add up SNP genotypes. Like the three-point rating scale for shyness, SNP genotypes are scored as 0, 1 or 2, indicating the number of ‘increasing’ alleles, as in the example of the FTO SNP. In the same way that we can add up alleles for one SNP to create a genotypic score, we can also add up alleles for many SNPs to create a polygenic score, just as we add questionnaire items to create a shyness score. The results from genome-wide association studies are used to select SNPs and to assign weights to each SNP. For example, in the GWA analysis of weight, the FTO SNP accounts for much more variance than other SNPs, so it should count for much more in a polygenic score for weight.

Location 2296-2303

Using only genome-wide significant hits is like demanding that each item in our shyness scale predicts significantly on its own. We don’t do this for other psychological scores because it is unrealistic to expect each item to stand on its own. The goal is to have a composite scale that is as useful as possible. A better idea is to do what we do when we create other psychological scores: keep adding items as long as they add to the reliability and validity of the composite in independent samples. For polygenic scores, the key criterion is prediction. The new approach to polygenic scores is to keep adding SNPs as long as they add to the predictive power of the polygenic score in independent samples. This is the strategy that has paid off in the last two years in producing powerful polygenic scores for psychological traits. Some false positives will be included in the polygenic score but that is acceptable as long as the signal increases relative to the noise, in the sense that the polygenic score predicts more variance.

Location 2360-2362

The big question is the extent to which polygenic scores will be able to predict all the heritable variance of traits. This gap is called missing heritability, and is described in the Notes section at the end of this book.