2 main lessons: 1. personality, behaviour, attitudes (i.e.psychology) is highly influenced, though not determined, by genetics, 2. all these matters are on a spectrum not either/or. Both are things I suspected but this gives it scientific cred. DNA does not fix trait absolutely but makes for higher probabilities, which only show up when sampling is on a large enough scale. Clearly written with some personal touches, though the details of research processes sometimes a bit hard-going. ( )

Interesting but not convinced. ( )

(My) book of the year. The few bad policy recommendations doesn't nearly dent the amount of good contained here. ( )

"Robert Plomin is a pioneer of modern behaviour genetics and Blueprint is unabashedly an exercise in cheerleading for the field. His enthusiasm can be contagious and his exposition of the surprising and sometimes seemingly paradoxical discoveries in his discipline over the last three decades or so can be fascinating. But that enthusiasm sometimes gets the better of him: the book glosses over too many of the weaknesses of human behaviour genetics as it is currently practised and Plomin sometimes makes claims that, even if technically true, are at the very least deeply misleading.

In his prologue, Plomin asks us to imagine a ‘completely reliable and unbiased’ ‘fortune-telling device’ that can ‘predict psychological traits like depression, schizophrenia and school achievement’ from ‘the moment of your birth’. He then assures us that contemporary behaviour genetics will provide the same insights as such a device. A moment’s reflection, however, points towards a problem. Identical twins, who share the same DNA, are not always concordant for psychological traits. If one twin, for instance, has schizophrenia, the chance that the other has it is around 50 per cent (there is some controversy over the exact figure). So no analysis of our DNA will be able to predict in a ‘completely reliable’ way whether or not an individual will be diagnosed with schizophrenia at some point in their life. Plomin knows this: in the chapters devoted to making predictions on the basis of DNA, he is careful to explain that these are statistical in nature, based on studies of particular populations, and may not always apply to individuals. That’s fine, but I hope it is obvious that such an acknowledgement is at odds with the promise of the prologue.

Plomin’s main concern in Blueprint is with heritability. This is a technical notion and it is easy to be confused or even misled by it. Roughly, heritability is a measure of the extent to which the variation in a trait, within a particular population, is associated with genetic variation. It is not a measure of how ‘genetic’ a trait is. As Plomin makes clear, in the case of many traits that don’t vary (think, for example, of our basic bilateral symmetry, or the way that our bodies are structured from top to bottom), heritability is undefined or zero, but our genetic endowments are key to their development.

Plomin favours genetic explanations for human behaviour. For instance, he notes that studies of the environmental influences on academic achievement find at best very small effects – generally less than 1 per cent of the variance is associated with specific, identifiable environmental variations – and then compares that to the total heritability of academic performance (perhaps around 50 per cent), noting that this is an enormous effect size. But he fails to note that the total variance associated with environmental differences is also about 50 per cent, and that when one looks for particular regions of the genome associated with differences, specific loci are associated with less than 1 per cent of the variance. It is unfair to play down the importance of environmental variation because only a small number of specific variations can be identified when the same is true for genetic explanations.

More generally, while Plomin states multiple times that genetic research, including analysis of heritability, ‘describes what is, not what could be’, his interpretation of this catchphrase seems rather limited. He claims that while classroom size is associated with student achievement, it is linked to only 1 per cent of the total variance, but this requires an important caveat: that even if it is true, it says nothing about how big a difference changing classroom size can make. Plomin is simply noting that in one particular set of schools that has been studied little of the total variation in academic achievement is in fact associated with classroom size. But we do know, or at least have very good reason to believe, that classroom size can make a big difference in student learning outcomes. What is still considered by many to be one of the most rigorous studies of the effect of classroom size on student learning in the USA, the Tennessee STAR study, found that relatively large reductions in class size (from about twenty-two to about fifteen students) led to relatively large improvements in student learning – on average, the equivalent of an extra three months of schooling in a four-year period, with larger gains going to students from disadvantaged backgrounds.

And this points towards the most important weakness of the book: Plomin’s relative lack of attention to the broader political realm. He seems, for example, completely to ignore the research on the effects of poverty on cognitive development and cognitive performance. The constant refrain that ‘parents matter, but they don’t make a difference’ is rather misleading here – it makes a big difference to one’s cognitive development (say) whether one grows up in a household that is relatively economically secure or one that is struggling (and the same adoption studies that show that cognitive performance is heritable also make clear the importance of family income). If everyone were reasonably well-to-do and white, Plomin’s more sweeping claims would be less misleading. But that isn’t the case (around one in five children in the USA grows up in poverty), and the profound effects of poverty and racism are surely an important cause of the kinds of ‘between-population’ differences that Plomin rather casually dismisses.

In the end, we can take the high heritability of bodyweight as a reason not to attribute obesity to the personal failings of individuals, while also imagining a time to come when we can ask people with high polygenic scores (calculations of genetic risk) for obesity to be particularly careful about their diets. Alternatively, we can remark (as Plomin does in the book’s endnotes) that obesity rates have risen drastically over the past decades and ask what has happened to our environment that has caused this and what changes we can make that might reverse the shift. In every case like this, Plomin’s interest is in the first type of response, but the latter is likely to be where we need to look for real improvements in our wellbeing."

Literary Review: Jonathan Kaplan, November 2018 ( )

Popular science book about DNA. Fascinating material, often based on twin studies. No doubt DNA is powerful, however, one is left with a few question. For instance, it is not clear how much variance the estimated genetic propensities explain, even if they are predictive. And there are many unqualified statements about the importance of genes, with the point that it is in narrowly defined populations and environments only underlying. Should be kept in mind regarding statements about parents/schools/X "matter, but do not make a difference". And the social sciences have found many environmental and institutional effects. The book has also been forcefully criticized by fellow scientist Eric Turkheimer on is blog - it is by now unclear how that will unfold. ( )