The nature and nurture of education

Environmental and genetic causes of individual differences in educational achievement

Blog written by Professor Michael Thomas


The nature-nurture issue is well-known amongst teachers: children differ, and some of these differences are due to the children’s nature, some due to the environment they are raised in, and some a combination of children’s different natural reactions to the environments they are raised in.

Gaps in educational achievement between children are a key issue for society, and their causes have been a focus for research in the social sciences. These gaps have spurred educational neuroscience investigations into possible underlying brain mechanisms, with separate work in the areas of environmental influences and genetic mechanisms.


With respect to environmental influences, research has focused on the effects of socioeconomic status (SES), one of the environmental measures showing most predictive power on cognitive and educational outcomes. SES is usually measured by parental levels of education and income. Notably, SES differences in cognitive abilities are already observable when children start school, and tend not to narrow with age (e.g., Hackman et al., 2015), and may even widen. For example, von Stumm and Plomin (2015) reported that in a large sample of almost 15,000 UK children followed from infancy through adolescence, children from low SES families scored on average 6 IQ points lower at age 2 than children from high SES backgrounds but by age 16, this difference had almost tripled. Notably, the differences in cognitive ability in school-age children associated with SES tend to be larger than the effects produced by different schooling (e.g., Walker, Petrill & Plomin, 2005). The greater effects of the home than the school on cognitive ability suggest that educational neuroscience must retain a focus on developmental factors beyond the classroom.

SES as a measure of the environment is only a proxy for the actual experiences of the child and therefore the actual causal mechanisms. These mechanisms may operate along multiple pathways, and those pathways may differ depending both on the absolute level of poverty across countries and the level of inequality. These pathways include prenatal factors, such as maternal diet, smoking, alcohol consumption, and stress; postnatal nurturing factors, such as early caregiver sensitivity; and the richness of postnatal cognitive stimulation (Hackman, Farah & Meaney, 2010; Sheridan & McLaughlin, 2016). Notably, the effects of SES are uneven across cognitive profiles, with relatively greater effects on language development and executive functions, and weaker on visuo-spatial cognition (Farah et al., 2006), although the reasons for this unevenness are unknown.

In a large US sample, Noble et al. (2015) demonstrated differences in structural brain measures of cortical thickness and cortical area correlated with SES measures, in regions that fitted with the cognitive abilities showing largest effects (temporal regions for language, prefrontal regions for executive functions). Investigations of causal mechanisms are compromised by the fact that so many correlated environmental factors are associated with differences in SES (Hackman et al., 2015); but in theory, an understanding of mechanism would suggest most effective points to intervene to alleviate the effects of poverty and deprivation on educational outcomes (Thomas, 2017).


With respect to genetics, work in behaviour genetics has begun to impact on education. Most obviously, evidence points to the heritability of differences in educational achievement. Heritability is defined as the amount of variation in behaviour explained by genetic similarity. Heritability is either inferred on the basis that achievement is more similar the more genetically similar individuals are (e.g., the idea that ability runs in families); or it is measured directly by correlating variation in individual letters of DNA (single nucleotide polymorphisms or SNPs) to educational outcomes in large samples (though the variation in behaviour explained by direct measures of DNA variation is typically much smaller, perhaps only ~10% instead of the ~50% explained by observing patterns that run in families; e.g., Okbay et al., 2016). For example, within the last five years, papers have been published showing that the heritability of educational achievement aged 16 is up to 60%, and that controlling for IQ, it appears to be the same genes that explain variability in different academic disciplines (Krapohl et al., 2014; Rimfeld et al., 2015); that social mobility (measured as children achieving higher educational levels than their parents) is itself around 50% heritable (Ayorech et al., 2017); that, on genetic grounds, there is little evidence that selective schools produce better educational outcomes than non-selective schools (Smith-Woolley et al., 2018); and that educational achievement may be a causal protective factor against Alzheimer’s disease (Zhu et al., 2018).

There is insufficient space here to evaluate these types of claims (see Thomas et al., 2013; Ashbury & Plomin, 2014; Meaburn, in press). However, we can simply note that a genetic perspective on education emphasises that not all differences between children are environmental in origin; but as yet, does not point to specific implications for teaching. Somewhat counter-intuitively, environments that optimise learning will increase the extent to which educational success is heritable. For example, greater heritability is observed in classes with better teachers (Taylor et al., 2010), as it is in affluent families compared to impoverished ones (Turkheimer et al., 2003). This is because heritability acts as an indicator of where limiting factors lie: when better school environments reduce limits on learning, variability between children is more readily explained by their genetic make-up (Asbury, 2015; Thomas, Kovas, Meaburn & Tolmie, 2015). Increasing heritability rates could therefore be a useful indicator of reducing educational inequality.

Effects linked to genes or changes in the brain are not inevitable

The risk with both neuroscience evidence of the effects of poverty on brain development, and of genetic effects on educational achievement, is that these will be construed by educators as ‘deterministic’ or inevitable outcomes. Yet many cognitive effects of poverty can be ameliorated by intervention, such as training executive functions, because the brain is plastic (Neville et al., 2013); and previously measured genetic effects may disappear when environments are altered. Indeed, the future potential of genetics would be to predict the best, tailored environment for each child to reach his or her genetic potential. However, currently, genetic studies are focused on maximising predictive power (such as by the use of polygenic risk scores, which combine the effects of all measured DNA variation to predict differences in behaviour) – rather than informing a neuroscientific understanding of the brain mechanisms of learning that would enable such tailoring of environments.



Asbury, K., & Plomin, R. (2014). G is for Genes: The Impact of Genetics on Education and Achievement. Oxford, UK: Wiley Blackwell.

Ayorech, Z., Krapohl, E., Plomin, R., & von Stumm, S. (2017). Genetic Influence on Intergenerational Educational Attainment      . Psychological Science,28(9), 1302 – 1310.

Farah, M. J., Shera, D. M., Savage, J. H., Betancourt, L., Giannetta, J. M., Brodsky, N. L., …  & Hurt, H. (2006). Childhood poverty: Specific associations with neurocognitive development. Brain Research, 1110, 166 –174.

Hackman D.A., Gallop, R., Evans, G.W. & Farah, M.J. (2015). Socioeconomic status and executive function: Developmental trajectories and mediation. Developmental Science, 18(5), 686–702.

Hackman, D.A., Farah, M.J. & Meaney, M.J. (2010). Socioeconomic status and the brain. Nature Reviews Neuroscience, 11, 651– 659.

Meaburn, E., (in press). Genetics and education. In: M. S. C. Thomas, D. Mareschal, & I. Dumontheil (Eds.), Educational Neuroscience: Development Across the Lifespan. London, UK: Routledge.

Noble, K.G., Houston, S.M., Brito, N.H. et al. (2015). Family income, parental education and brain structure in children and adolescents. Nature Neuroscience, 18, 773–778.

Okbay, A., Beauchamp, J. P., Fontana, M. A., Lee, J. J., Pers, T. H., Rietveld, C. A. et al. (2016). Genome-wide association study identifies 74 loci associated with educational attainment. Nature, 533, 539–542.

Rimfeld, K., Kovas, Y., Dale, P. S., & Plomin, R. (2015). Pleiotropy across academic subjects at the end of compulsory education. Scientific Reports, 5, 11713

Smith-Woolley, E., Pingault, J-B., Selzam, S., Rimfeld, K., … Plomin, R. (2018). Differences in exam performance between pupils attending selective and non-selective schools mirror the genetic differences between them. npjScience of Learning, 3, Article number: 3 (2018).

Thomas, M. S. C. (2017). A scientific strategy for life chances. The Psychologist, 30, 22-26.

Thomas, M. S. C., Kovas, Y., Meaburn, E., & Tolmie, A. (2015). What can the study of genetics offer to educators? Mind, Brain & Education, 9(2), 72-80.

von Stumm, S. & Plomin, R. (2015). Socioeconomic status and the growth of intelligence from infancy through adolescence. Intelligence, 48, 30–36.

Walker, S. O., Petrill, S. A., & Plomin, R. A. (2005). A genetically sensitive investigation of the effects of the school environment and socioeconomic status on academic achievement in seven-year-olds. Educational Psychology, 25(1), 55-73.

Using analogies in the classroom

apples-and-orangesBy Matt Slocombe

In my talk at the Centre for Educational Neuroscience, I discussed my research looking at how children learn using analogies. Using analogies can be a very powerful teaching and learning strategy in the classroom since analogies allow children to rapidly learn new knowledge by seeing how something new is similar to something they already know about. In a biology lesson, they might hear the analogy ‘mitochondria are the power source for a cell’. By using this analogy, children can use their knowledge of the causal relationships between something they know about – perhaps a battery and a mobile phone – to quickly understand the causal relationships between mitochondria and a biological cell.

The study I discussed investigated how the strength of children’s existing knowledge affects their ability to use it in an analogy. To use the mitochondria-power source example, the results indicate that just knowing about power sources is not enough; children need a strong abstract concept of power source to successfully make the analogy. When presented with the mitochondria-power source analogy, children with weaker abstract concepts may well be thinking of irrelevant features of a concrete example (such as how a battery looks or the sound of an engine) rather than the necessary abstract power source concept.

For teachers, one way to help children with weaker abstract concepts use analogies may be to briefly discuss the existing knowledge you want to use in an analogy first. For example, discussing the functional role of batteries in phones and engines in cars prior to introducing the mitochondria-power source analogy would highlight the relevant causal features of power sources necessary for understanding the functional role of mitochondria in cells.


For more information or to stay up to date with Matt’s research, please see his website at  or follow him on twitter @matthewslocombe

Stress and learning in children: evidence from neuroscience and relevance for teachers

In yesterday’s CEN seminar, Dr Sue Whiting talked about the complexities of the human stress response. She explained how each individual’s stress response – the strength and duration of each person’s response, their base stress levels, what constitutes a ‘stressor’ – is highly variable, presenting challenges for teachers to pitch lessons in a way that will be effective for all learners. At the end of her talk, Sue gave some tips for teachers which might help moderate the effects of children vulnerable to high levels of stress:

Supervised breakfast clubs could potentially serve a dual purpose: in addition to providing good nutrition, they could also allow a longer period of time for delayed cortisol effects, arising from stressors outside the school environment, to dissipate.
Deep diaphragmatic breathing, which appears to balance the nervous system by stimulating the parasympathetic system, sometimes characterised as the ‘rest and digest’ system. This could be used with whole classes on a regular basis to help control all the children’s stress levels.

If teachers learn to recognise signs that children are beginning to be overwhelmed, then they could reassure them and take some of the pressure away by reducing the complexity of a task, particularly reducing the load on working memory.

For some children, rewarding effort may help. Others may respond to adopting a mindset which repositions stress as an enhancing agent. Some children find it helpful simply to say ‘I am excited’, as a way of recognising their stress and casting it as something positive.

Simple interventions to capture interest can improve reading comprehension

In our last seminar, Lisa Fridkin and Jane Hurry presented their work investigating situational interest in children, asking whether improving intrinsic motivation through simple manipulations such as the arrival of a novel person in the classroom or being offered (apparent) choice in a reading book, could improve reading comprehension. Below, Lisa talks a bit more about the research:


Situational interest (SI) is described as the first stage of interest development (see Hidi & Renninger, 2006; 2011) and is potentially of high value to practitioners because it is an intrinsic motivator that can be externally manipulated. It is theorised to raise attention, effort and enjoyment in a task or activity and act as a forerunner to personal interest.

A series of three experimental studies were conducted to investigate the effects of three potential triggers of SI on reading comprehension performance and reported task enjoyment with children aged 8-9 years old. Interaction effects for gender and reading ability were also investigated. Approx. 100 children participated in each study which used a repeated measures, cross-over design, and the same central materials (two short stories, comprehension questions, enjoyment questionnaire).

Study 1 investigated choice, offering children a perceived but meaningful choice in storybook (control condition: allocated story); Study 2 investigated novelty where a story prologue was read aloud by a visitor (control condition: task administered by class teacher, without the prologue read aloud; Study 3 also investigated novelty, through non-textual features – scratch and sniff stickers interspersed in the text, (control condition: no stickers).

All three studies found significant effects on reading comprehension performance (medium to large effect sizes) and on reported task enjoyment (small effect sizes), suggesting that for this age group, these triggers are effective in changing the way the children interact with a reading comprehension task. Effects by gender were found for Study 3 only, where girls achieved higher scores on the reading comprehension task compared to boys. No effects were found by reading ability.

These results suggest that quite slight manipulations of a task can have a relatively large and immediate impact on comprehension task performance and enjoyment and suggest that SI may signal an effective way to introduce intrinsic motivation in the classroom. There is emerging evidence linking motivation to reward circuitry, indicating that our understanding of these processes could be supported with further research.

Susanne de Mooij talks about the potential for personalised learning in an online maths game for children

The examples below show how the eye and mouse tracking work with a real example.

Susanne says, “While answering a maths question, we measure where the child looks on the screen and how they move their mouse, as an indirect measure of their thought process before making the choice.
The figures display the eye (black line) and mouse tracking (orange line) of two opposite strategies.
The first figure (6+6) suggests mental calculation where the child’s eyes have first fixated on the question, followed directly after by fixating on the correct answer. In the second example (7×4), the child alternates between the question and several possible answer options, before ultimately giving a wrong answer.
The information that can be gleaned from such examples is clearly much more detailed than simply whether a child got a particular question correct or incorrect.”example1_strategyuse_mouseeyemovements example2_strategyuse_mouseeyemovements

EARLI SIG22 Neuroscience and Education – conference June 2018

by Jessica Massonnié –

Between the 4thand 6thof June, The Wellcome Trust hosted the EARLI SIG22 conference on Neuroscience and Education, organised by Lia Commissar and Annie Brookman-Byrne.

Over two and a half days, various forums and opportunities were provided for teachers and researchers to communicate their work, reflect on their practice, and push the field forward. It is hard to summarise such a dynamic conference in a few words without going into over-heard generalisations, so here are some key moments. The full program as well as the list of attendees is available here.

The keynote talks reflected the interdisciplinarity and diversity of the topics addressed throughout the conference. Heidi Johansen-Berg talked about one of the neuroscience in education projects funded by the Wellcome Trust and Education Endowment Fund – studying the effect of physical activity on the brain and cognition. She also highlighted challenges the team have faced during the process of planning, piloting, and carrying out the study. Nadine Gaab, who develops models to understand the development of the reading brain, demonstrated how neuroscience can be used to inform and develop a reading skills screening tool enabling earlier intervention for children. Robert Plomin gave an interesting update of how the field of genetics has progressed, and how a genetic approach could be used to understand behaviour.

One key line of thought however was to further the collaboration between University research and educational practice. Paul Howard-Jones presented a MOOC, the Science of Learning, aiming to introduce key concepts about cognitive science to teachers. Creating such a platform raises several challenges, such as the understanding of teachers’ needs, the awareness of official guidelines and curriculum, as well as the need to find a compromise between the always evolving and controversial research findings and the delimitation of useful and consensual concepts. This session also highlighted the importance of incorporating the science of learning into initial teacher education, so that new teachers can further understand “why” things may (or may not) be effective for learning.


Slide from Paul Howard-Jones, giving examples of academic papers which have messages for classroom practice.

Echoing his talk, a symposium was focused on how to make science accessible to the public, teachers and policy makers. Yana Weinstein and Megan Sumeracki from the Learning Scientists, Steve Cross, founder of the comedy network Bright Club, as well as Ben Bleasdale, policy advisor from Wellcome, provided some inspiring advice to remain flexible and creative in the dissemination process. Youtube videos, podcasts or stand-up comedy can all help to reach broader audiences and help researchers to reflect on their work by avoiding jargon and thinking about practical applications. Key messages here included being clear about who the audience is that you are aiming to reach, and ensure you plan opportunities that suit that audience (rather than suiting the researcher).


From left to right: Yana Weinstein, Megan Sumeracki, Steve Cross, and Ben Bleasdale

As dissemination and educational applications should be more than a “potential” or “desirable” by-product, a second symposium presented teacher-led research. The booklet “Evidence that counts: 12 teacher-led randomized controlled trials andother styles of experimental research”, edited by the Education Endowment Trust, provides concrete examples of such initiatives. One advantage of teachers carrying out research in their own setting is that it empowers them to design specific research that’s relevant to them. It’s important to remember that there is no “1 size fits all” in education, and teachers should feel empowered to try things out, evaluate, and make changes as appropriate.


From left to right: Richard Churches, Sharon Baker, Charlotte Hindley, Daria Makarova, and Glenn Whitman

Discussions about how to move the field forward culminated on the third day of the conference, offering attendees the opportunity to create group discussions on their topic of interest. The open space sessions fostered the creation of partnerships and the elaboration of concrete action points. Some specific research topics, such as motivation, sleep, or inhibition, were addressed. Discussions about the process of connecting researchers and teachers were also brought to the table / floor. Among others: Cognitive neuroscience into initial teacher education; the creation of a database for researchers and schools who want to work with each other; and ways to measure the effects of teacher neuroscience knowledge on their practice.

In sum, the engaging mix of posters, oral presentations, and discussions means that this conference will remain very memorable, but will also hopefully lead to concrete actions for the future, closer collaboration, and greater understanding about the role of educational neuroscience in teaching and learning.



We were delighted to welcome Rae Snape, headteacher at The Spinney Primary School in Cambridge, and Dr Sara Baker, University of Cambridge, to the CEN yesterday. They shared their experience of working together, and what they’ve learnt from that process about developing projects together as researchers and educators. Here is a short summary of their talk:

What if… we were able to say more about how the brain learns?

By Su Morris, PhD student

Debate at UCL IoE on 15th May 2018


Becky Allen, Director of the Centre for Education Improvement Science, UCL Institute of Education (IOE)

Steven Rose,Emeritus Professor of Neuroscience, the Open University

Cat Sebastian,Reader in the Department of Psychology, Royal Holloway, University of London

Michael Thomas, Professor of Cognitive Neuroscience at Birkbeck, University of London, and director of the Centre for Educational Neuroscience (CEN)

Discussion chaired by:

Becky Francis, Director, UCL Institute of Education


This fascinating debate set out to explore what can be expected from educational neuroscience for policy and practice, as well as how to recognise ‘neuro snake oil’ (see here for the CEN series on neurohits and neuromyths). Each panellist introduced their viewpoint and initial thoughts:

Michael Thomas began by saying that we learn because the brain is plastic, i.e. pathways and connections in the brain are made and changed based on our experiences. The problem is that learning is a complicated process with seven systems interacting in the brain, all varying in their development and in their sensitivity to socio-emotional factors. Understanding learning is also challenging as it is only one aspect of a very complex educational system. There are two main ways that neuroscience can be translated into something useful for teachers – understanding brain health so students are ready to learn, and understanding how learning takes place in different subject areas. He concluded by saying that although it is a challenge, it is one worth pursuing.

Becky Allen highlighted the impact of having an incomplete understanding of cognition, particularly the lack of confidence in cognitive theories, which are important for generalising across different groups of children and classrooms. Currently, ideas such as pretesting, redundancy, and desirable difficulties are being promoted by some influential figures in education, however these run the risk of being the new myths – they are based on a small number of studies with short-term outcomes. The key question is whether ‘learning how we learn’ will help us know how to teach. Each class group has around 30 individual learners, meaning that teachers spend the majority of their time with the group rather than considering learning at an individual level.

Cat Sebastian outlined the positive influence of neuroscience on understanding the adolescent age group; why teens behave like they do, and how connectivity in decision-making and emotional regulation continues to mature in this age group. This can help both teachers and students understand changes in risk-taking behaviours and the changing influence of peers. Neuroscience can also help explain why different interventions may be more successful with some groups than others, and can even motivate wider research in particular areas; e.g. there had been very little research on adolescence until neuroscience showed that development continues into and beyond this age group.

Steven Rose began by making the point that it is children who learn, not brains. Brains are embedded in children who are embedded within a school and society. He questioned whether answering questions about the brain can really help with learning and teaching. It may even distract from things that we do know and need to respond to, e.g. food banks in order to meet the current needs of students. Many of the neuromyths, including VAK (visual, auditory, kinaesthetic learning, or learning styles) and brain gym were considered ‘good science’ a couple of decades ago. He closed by predicting that some of the current developments in techniques and technologies may well turn out to be neuromyths of the future.

Becky Francis summed up the opening comments as three important considerations: 1) risks of early adoption of scientific ideas and fads; 2) the importance of interaction between different disciplines, and 3) considering the wider educational context.

The Q&A session drew out further themes and practical applications, but an important first question was about different timescales. Translation into practice needs to be evidence-based, and it is important to have intervening stages between neuroscience and practice – perhaps this is a new profession in itself? An analogy was drawn with medicine, which has taken 200 years to get to a position of being research-led but where doctors still treat the person. Scientists may take years to develop understanding; policy-makers have about 18 months; teachers need something for Monday morning! A way of managing this is to balance the time and disruption of changing practice, against the risk factors. For example, spaced learning has a one-off disruption and time cost, but it is not likely to be damaging, and therefore on balance may be something teachers could introduce.

The panel then responded to questions from the floor.

When asked about mental health, they highlighted the importance of shaping policy for allocating resources, but that neuroscience has a role in identifying risk factors, understanding of resilience, and how early life stresses can have long-reaching impacts on brain development.

A question on cognitive load theory (amount of information that can be held in working memory) highlighted the fact that some ideas may be introduced more easily because they are easily implemented, and help to make sense of the wide variation across a class. However, it may also lead to inappropriate expectations about how behaviour can be changed; e.g. training working memory does not lead to gains in other areas.

Finally, a question about funding research emphasised the importance of multidisciplinary research, where funding could be allocated on the basis of a problem to be solved, rather than being discipline-focused.

The session concluded with an opportunity for attendees to identify areas they would like neuroscience to answer, such as concerns parents may have about whether they have passed on difficulties or challenges to their children e.g. in maths.

This well-attended event emphasised once again the interest people have in understanding how knowledge about the brain may inform teaching practice and improve children’s learning. One important outcome from the discussion was the importance of multidisciplinary research; not only between education, psychology, cognitive science, and neuroscience, but also with social disciplines. The context of students’ learning is important and needs to be considered. Also, many are aware of neuromyths which has led to a certain amount of cautiousness about adopting new ideas before they are fully researched. Therefore, there is a balance to consider between time commitment and disruption vs risk. Overall, there was general agreement that it is useful and important to understand ‘why’ – why some interventions are more successful with some than others; why some children respond to certain teaching methods better than others; why there is such variation within and between each class. Through exploring answers to these questions, educational neuroscience will have something to offer to teachers and learners – but it is unlikely to be ready for Monday morning!

Click below to watch the video of the event: