genes, environment and behaviour

There was considerable kerfuffle on Twitter last week following a blog post by David Didau entitled ‘What causes behaviour?’  The ensuing discussion resulted in a series of five further posts from David culminating in an explanation of why his views weren’t racist. I think David created problems for himself through lack of clarity about gene-environment interactions and through ambiguous wording. Here’s my two-pennyworth.


Genes are regions of DNA that hold information about (mainly) protein production. As far as we know, that’s all they do. The process of using this information to produce proteins is referred to as genetic expression.


The context in which genes are expressed. Before birth, the immediate environment in which human genes are expressed is limited, and is largely a chemical and biological one. After birth, the environment gets more complex as Urie Bronfenbrenner demonstrated.  Remote environmental effects can have a significant impact on immediate ones. Whether a mother smokes or drinks is influenced by genetic and social factors, and the health of both parents is often affected by factors beyond their control.


Epigenetic factors are environmental factors that can directly change the expression of genes; in some cases they can be effectively ‘switched’ on or off.   Some epigenetic changes can be inherited.


Behaviour is a term that’s been the subject of much discussion by psychologists. There’s a useful review by Levitis et al here. A definition of behaviour the authors decided reflected consensus is:

Behaviour is: the internally coordinated responses (actions or inactions) of whole living organisms (individuals or groups) to internal and/or external stimuli, excluding responses more easily understood as developmental changes.

traits and states

Trait is a term used to describe a consistent pattern in behaviour, personality etc. State is used to describe transient behaviours or feelings.

David Didau’s argument

David begins with the point that behavioural traits in adulthood are influenced far more by genes than by shared environments during childhood. He says: “Contrary to much popular wishing thinking, shared environmental effects like parenting have (almost) no effect on adult’s behaviour, characteristics, values or beliefs.* The reason we are like our parents and siblings is because we share their genes. *Footnote: There are some obvious exceptions to this. Extreme neglect or abuse before the age of 5 will likely cause permanent developmental damage as will hitting someone in the head with a hammer at any age.”

In support he cites a paper by Thomas Bouchard, a survey of research (mainly twin studies) about genetic influence on psychological traits; personality, intelligence, psychological interests, psychiatric illnesses and social attitudes. David rightly concludes that it’s futile for schools to try to teach ‘character’ because character (whatever you take it to mean) is a stable trait.

traits, states and outcomes

But he also refers to children’s behaviour in school, and behaviour encompasses traits and states; stable patterns of behaviour and one-off specific behaviours. For David, school expectations can “mediate these genetic forces”, but only within school; “an individual’s behaviour will be, for the most part, unaffected by this experience when outside the school environment”.

He also refers to “how we turn out”, and how we turn out can be affected by one-off, even uncharacteristic behaviours (on the part of children, parents and teachers and/or government).   One-off actions can have a hugely beneficial or detrimental impact on long-term outcomes for children.

genes, environment and interactions

It’s easy to get the impression from the post that genetic influences (David calls them genetic ‘forces’ – I don’t know what that means) and environmental influences are distinct and act in parallel. He refers, for example, to “genetic causes for behaviour as opposed to environmental ones” (my emphasis), but concedes “there’s definitely some sort of interaction between the two”.

Obviously, genes and environment influence behaviour. What’s emerged from research is that the interactions between genetic expression and environmental factors are pretty complex. From conception, gene expression produces proteins; cells form, divide and differentiate, the child’s body develops and grows. Genetic expression obviously plays a major role in pre-natal development, but the proteins expressed by the genes very quickly form a complex biochemical, physiological and anatomical environment that impacts on the products of later genetic expression. This environment is internal to the mother’s body, but external environmental factors are also involved in the form of nutrients, toxins, activities, stressors etc. After birth, genes continue to be expressed, but the influence of the external environment on the child’s development increases.

Three points to bear in mind: 1) A person’s genome remains pretty stable throughout their lifetime. 2) The external environment doesn’t remain stable – for most people it changes constantly.  Some of the changes will counteract others; rest and good nutrition can overcome the effects of illness, beneficial events can mitigate the impact of adverse ones. So it’s hardly surprising that shared childhood environments have comparatively little, if any, effect on adult traits.   3) Genetic and environmental influences are difficult to untangle due to their complex interactions from the get-go. Annette Karmiloff-Smith* highlights the importance of gene-environment-behaviour interactions here.

Clearly, if you’re a kid with drive, enthusiasm and aspirations, but grow up on a sink estate in an area of high social and economic deprivation where the only wealthy people with high social status are drug dealers, you’re far more likely to end up with rather dodgy career prospects than a child with similar character traits who lives in a leafy suburb and attends Eton. (I’ve blogged elsewhere about the impact of life events on child development and long-term outcomes, in a series of posts starting here.)

In other words, parents and teachers might have little influence over behavioural traits, but they can make a huge difference to the outcomes for a child, by equipping them (or not) with the knowledge and strategies they need to make the most of what they’ve got. From other things that David has written, I don’t think he’d disagree.  I think what he is trying to do in this post is to put paid to the popular idea that parents (and teachers) have a significant long-term influence on children’s behavioural traits.  They clearly don’t.  But in this post he doesn’t make a clear distinction between behavioural traits and outcomes. I suggest that’s one reason his post resulted in so much heated discussion.

genes, environment and the scientific method

I’m not sure where his argument goes after he makes the point about character education. He goes on to suggest that anyone who queries his conclusions about the twin studies is dismissing the scientific method, which seems a bit of a stretch, and finishes the post with a series of ‘empirical questions’ that appear to reflect some pet peeves about current educational practices, rather than testing hypotheses about behaviour per se.

So it’s not surprising that some people got hold of the wrong end of the stick. The behavioural framework including traits, states and outcomes is an important one and I wish, instead of going off at tangents, he’d explored it in more detail.

*If you’re interested,  Neuroconstructivism by Mareschal et al and Rethinking Innateness by Elman et al. are well worth reading on gene-environment interactions during children’s development.  Not exactly easy reads, but both reward effort.


Bouchard, T. (2004).  Genetic influence on human psychological traits.  Current Directions in Psychological Science, 13, 148-151.

Elman, J. L., Bates, E.A., Johnson, M., Karmiloff-Smith, A., Parisi, D., & Plunkett, K. (1996). Rethinking Innateness: A Connectionist Perspective on Development.  Cambridge, MA: MIT Press.

Karmiloff-Smith A (1998). Development itself is the key to understanding developmental disorders. Trends in Cognitive Sciences, 2, 389-398.

Levitis, D.A., Lidicker, W.Z., & Freund, G. (2009).  Behavioural biologists don’t agree on what constitutes behaviour.  Animal Behaviour, 78 (1) 103-110.

Mareschal, D., Johnson, M., Sirois, S., Spratling, M.W., Thomas, M.S.C. & Westermann, G. (2007). Neuroconstructivism: How the brain constructs cognition, Vol. I. Oxford University Press.







All snowflakes are unique: comments on ‘What every teacher needs to know about psychology’ (David Didau & Nick Rose)

This book and I didn’t get off to a good start. The first sentence of Part 1 (Learning and Thinking) raised a couple of red flags: “Learning and thinking are terms that are used carelessly in education.” The second sentence raised another one: “If we are to discuss the psychology of learning then it makes sense to begin with precise definitions.”   I’ll get back to the red flags later.

Undeterred, I pressed on, and I’m glad I did. Apart from the red flags and a few quibbles, I thought the rest of the book was great.  The scope is wide and the research is up-to-date but set in historical context. The three parts – Learning and Thinking, Motivation and Behaviour, and Controversies – provide a comprehensive introduction to psychology for teachers or, for that matter, anyone else. Each of the 26 chapters is short, clearly focussed, has a summary “what every teacher needs to know about…”, and is well-referenced.   The voice is right too; David Didau and Nick Rose have provided a psychology-for-beginners, written for grown-ups.

The quibbles? References that were in the text but not in the references section, or vice versa. A rather basic index. And I couldn’t make sense of the example on p.193 about energy conservation, until it dawned on me that a ‘re’ was missing from ‘reuse’. All easily addressed in a second edition, which this book deserves. A bigger quibble was the underlying conceptual framework adopted by the authors. This is where the red flags come in.

The authors are clear about why they’ve written the book and what they hope it will achieve. What they are less clear about is the implicit assumptions they make as a result of their underlying conceptual framework. I want to look at three implicit assumptions about; precise definitions, the school population and psychological theory.

precise definitions

The first two sentences of Part 1 are;

Learning and thinking are terms that are used carelessly in education. If we are to discuss the psychology of learning then it makes sense to begin with precise definitions.” (p.14)

What the authors imply (or at least what I inferred) is that there are precise definitions of learning and thinking. They reinforce their point by providing some. Now, ‘carelessly’ is a somewhat pejorative term. It might be fair to use it if there is a precise definition of learning and there is a precise definition of thinking, but people just can’t be bothered to use them. But if there isn’t a single precise definition of either…

I’d say terms such as ‘learning’, ‘thinking’, ‘teaching’, ‘education’ etc. (the list is a long one) are used loosely rather than carelessly. ‘Learning’ and ‘thinking’ are constructs that are more complex and fuzzier than say, metres or molar solutions. In marked contrast to the way ‘metre’ and ‘molar solution’ are used, people use ‘learning’ and ‘thinking’ to refer to different things in different contexts.   What they’re referring to is usually made clear by the context. For example, most people would consider it reasonable to talk about “what children learn in schools” even if much of the material taught in schools doesn’t meet Didau and Rose’s criterion of retention, transfer and change (p.14). Similarly, it would be considered fair use of the word ‘thinking’ for someone to say “I was thinking about swimming”, if what they were referring to was pleasant mental images of them floating in the Med, rather than the authors’ definition of a conscious, active, deliberative, cognitive “struggle to get from A to B”.

Clearly, there are situations where context isn’t enough, and a precise definition of terms such as ‘learning’ and ‘thinking’ are required; empirical research is a case in point. And researchers in most knowledge domains (maybe education is an exception) usually address this requirement by stating explicitly how they have used particular terms; “by learning we mean…” or “we use thinking to refer to…”.  Or they avoid the use of umbrella terms entirely. In short, for many terms there isn’t one precise definition. The authors acknowledge this when they refer to “two common usages of the term ‘thinking’”, but still try to come up with one precise definition (p.15).

Why does this matter? It matters because if it’s assumed there is a precise definition for labels representing multi-faceted, multi-component processes, that people use in different ways in different circumstances, a great deal of time can be wasted arguing about what that precise definition is. It would make far more sense simply to be explicit how we’re using the term for a particular purpose, or exactly which facet or component we’re referring to.

Exactly this problem arises in the discussion about restorative justice programmes (p.181). The authors complain that restorative justice programmes are “difficult to define and frequently implemented under a variety of different names…” Those challenges could be avoided by not trying to define restorative justice at all, but by people being explicit about how they use the term – or by using different terms for different programmes.

Another example is ‘zero tolerance’ (p.157). This term is usually used to refer to strict, inflexible sanctions applied in response to even the most minor infringements of rules; the authors cite as examples schools using ‘no excuses’ policies. However, zero tolerance is also associated with the broken windows theory of crime (Wilson & Kelling, 1982); that if minor misdemeanours are overlooked, antisocial behaviour will escalate. The broken windows theory does not advocate strict, inflexible sanctions for minor infringements, but rather a range of preventative measures and proportionate sanctions to avoid escalation. Historically, evidence for the effectiveness of both approaches is mixed, so the authors are right to be cautious in their conclusions.

What I want to emphasise is that there isn’t a single precise definition of learning, thinking, restorative justice, zero tolerance, or many other terms used in the education system, so trying to develop one is like trying define apples-and-oranges. To avoid going down that path, we simply need to be explicit about what we’re actually talking about. As Didau and Rose themselves point out “simply lumping things together and giving them the same name doesn’t actually make them the same” (p.266).

all snowflakes are unique

Another implicit assumption emerges in chapter 25, about individual differences;

Although it’s true that all snowflakes are unique, this tells us nothing about how to build a snowman or design a better snowplough. For all their individuality, useful applications depend on the underlying physical and chemical similarities of snowflakes. The same applies to teaching children. Of course all children are unique…however, for all their individuality and any application of psychology to teaching is typically best informed by understanding the underlying similarities in the way children learn and develop, rather than trying to apply ill-fitting labels to define their differences. (p. 254)

For me, this analogy begged the question of what the authors see as the purpose of education, and completely ignores the nomothetic/idiographic (tendency to generalise vs tendency to specify) tension that’s been a challenge for psychology since its inception. It’s true that education contributes to building communities of individuals who have many similarities, but our evolution as a species, and our success at colonising such a wide range of environments hinges on our differences. And the purpose of education doesn’t stop at the community level. It’s also about the education of individuals; this is recognised in the 1996 Education Act (borrowing from the 1944 Education Act), which expects a child’s education to be suitable to them as an individual.  For the simple reason that if it isn’t suitable, it won’t be effective.  Children are people who are part of communities, not units to be built into an edifice of their teachers’ making, or to be shovelled aside if they get in the way of the education system’s progress.

what’s the big idea?

Another major niggle for me was how the authors evaluate theory. I don’t mean the specific theories tested by the psychological research they cite; that would be beyond the scope of the book. Also, if research has been peer-reviewed and there’s no huge controversy over it, there’s no reason why teachers shouldn’t go ahead and apply the findings. My concern is about the broader psychological theories that frame psychologists’ thinking and influence what research is carried out (or not) and how. Didau and Rose demonstrate they’re capable of evaluating theoretical frameworks, but their evaluation looked a bit uneven to me.

For example, they note “there are many questions” relating to Jean Piaget’s theory of cognitive development (pp.221-223), but BF Skinner’s behaviourist model (pp.152-155) has been “much misunderstood, and often unfairly maligned”. Both observations are true, but because there are pros and cons to each of the theories, I felt the authors’ biases were showing. And David Geary’s somewhat speculative model of biologically primary and secondary knowledge and ability, is cited uncritically at least a dozen times, overlooking the controversy surrounding two of its major assumptions –  modularity and intelligence. The authors are up-front about their “admittedly biased view” Continue reading