An age-old question

What makes us the way we are? Where do our capacities for happiness, criminality, illness, height or sexual orientation come from? Complex traits do, of course, arise from a combination of our genes and our environment, but attempts to understand how these factors combine rarely provide a satisfactory or quantitative answer.

In his book, psychologist Eric Turkheimer describes the history of the nature-nurture debate, including its origins, technical and conceptual developments, and the controversies surrounding it.

He is suspicious of attempts to make bold genetic claims about complex human traits given that nature and nurture are “irredeemably confounded” and that designing the sorts of experiments that would allow for more precise partitioning of causation is difficult for human behavioural traits: while we can perform breeding experiments on plants and non-human animals, we are unable to perform such experiments on human beings for obvious ethical reasons.

Scientists studying the genetics of human behaviour develop methodological workarounds, such as twin and genetic association studies, but these are also imperfect. Any associations that they give rise to are correlational, but not necessarily explanatory. Most descriptions of ‘genetic causes’ for complex behaviours do not elicit the causal pathways from genotype to complex phenotype, so are not really explaining the link between genotype and behavioural trait.

Genetics of complex traits

Understanding the Nature-Nurture Debate explores the long history of attempts to ground complex human traits in genetic differences, through twin and adoption studies, candidate-gene studies, genome-wide association studies (GWAS) and polygenic score. While Turkheimer does acknowledge that genes are part of the causal pathway for every human trait, he queries attempts to put a number on this in terms of heritability estimates (how much of a given outcome difference is associated with genetic differences) for complex traits.

He is also critical of more recent research using GWAS, such as investigations of ‘educational attainment’ (which in most studies is the number of years of schooling completed). Turkheimer describes how the number of SNP (single-nucleotide polymorphism) ‘hits’ increased as these studies acquired much larger samples, but the individual genetic differences were associated with smaller and smaller variations in outcomes. Turkheimer is sober about the significance of these findings, believing that these are not discoveries “in the usual scientific sense” and that for some of the associations “no science has been built on top of it, and it has no application in the real world”.

Turkheimer is suspicious of attempts to make bold genetic claims about complex human traits

Turkheimer does not share the enthusiasm for genetic studies that many behavioural scientists have: “Despite universal heritability, there are still no genes for being smart, happy or successful […] the magnitude of that (scientifically) disappointing outcome cannot be overstated.” As modern genomic methods have advanced, estimates of heritability have decreased from where they stood a century ago in the twin-study era, causing him to reflect on his own career: “Everything is still heritable and we still don’t have anything resembling a genetic explanation for differences in behaviour.”

Turkheimer describes the contested legacies of figures such as Cyril Burt and Francis Galton, as well as the misapplication of genetics in the eugenic policies around World War II, plus the controversial applications of genetic prediction, such as ‘intelligence’ testing. He rightly takes issue with research that seeks to identify racial differences in intelligence and critiques the notion of biological races, describing how there is more genetic variation within what are commonly conceived of as ‘races’ than between them.

The intelligence question

For Turkheimer, IQ tests are essentially elaborate and more sophisticated spelling tests. He says this not to disparage intelligence tests or spelling tests, acknowledging that both have utility, but to highlight that it is very easy to “slip from sensible estimation of task scores into an undefinable notion of ‘innate intelligence’”. IQ tests are not like x-rays – they cannot find invisible innate capacities – so for Turkheimer there is “no such thing as innate intelligence, there is just a tendency to get test items correct in the here and now”.

Psychologists such as Arthur Jensen saw intelligence as significantly heritable, but later work by James Flynn described large improvements in population IQ scores across comparatively short periods of time in what became known as the ‘Flynn effect’. Turkheimer shows how environmental interventions can produce changes in IQ scores, but that once environmental interventions are removed the improvements in IQ disappear. This is consistent with the notion that environment affects the outcome as long as the environmental change persists – if you want a long-term effect on test scores, then you need to have long-term environmental change.

The book closes with a useful summary of its key arguments, and a defence of individual human agency and our capacity to shape our own lives. Genes, family, environment, random events – they all have some contribution to how we turn out. As Turkheimer says, “we live a life” and, in the service of science, we must be modest about what we can say about any of the causal pathways in the most complex of human traits.

Conor McCrory is a science teacher who has written on topics related to genetics and education