Some insights regarding brain function tend to resonate with existing educational attitudes and concepts, and may be helpful in strengthening and consolidating existing practice. Examples of such insights arise from research on

• Brain plasticity. This emphasises the extent to which the structure and function of the brain can respond to environmental influences including education. Such studies tend to emphasise the general importance of educational influence on neurocognitive development, and will always find favour amongst those who feel passionate about the value and promise of education. (eg Immordino-Yang presented two complimentary case studies of boys who had undergone the surgical removal of an entire brain hemisphere (Immordino-Yang, 2007). Both were able to develop language and social skills far beyond expectations, by developing individual processing strategies that exploited the functionality associated with the remaining hemisphere).
• The role of phonological processing skills which includes an understanding of reading processes and reading difficulties. Studies have linked dyslexia to reduced functioning in areas of the brain associated with phonological processing, and have demonstrated that both the reading difficulties and thus reduced functioning are amenable to remediation using approaches that emphasise sound-spelling relationships (Shaywitz et al, 2004). Ongoing research continues to emphasise the importance of modern “phonics” approaches in the classroom.
• Creativity that shows that the inclusion of remotely associated concepts increases activity in brain regions linked to creative effort, supporting the use of such strategies as a means to foster creativity (Howard-Jones et al, 2005).
• Visualisation that shows visualising an object recruits most of the brain areas activated by actually seeing it (Kosslyn, 2005), supporting the use of visualisation as a learning tool.

However, although such neuroscientific studies may play a vital role in consolidating existing educational attitudes in some areas, perhaps the more salient influence of neuroscience in education will arise from more counter-intuitive findings. That is, it may be the educationally-relevant findings about brain function that are more surprising in their content and implications that stimulate the more dramatic changes in educational thinking and practice.

By definition, however, it is difficult to predict surprises. Yet, there are some areas of neuroscience research where results already appear to challenge the types of assumptions teachers work with, and may soon give rise to new directions. These areas will, therefore, now be given particular focus.

Early Numeracy

The acquisition of much formal mathematics relies on our ability to learn rules and procedures. This has been demonstrated by a neuroimaging study involving adults who were asked to calculate answers exactly. Researchers observed increased activity in areas of the brain involved in word association and language activity, the left frontal and angular gyri (Dehaene et al, 1999), as these adults pursued mathematical procedures by following formal mathematical steps that could be linguistically encoded. However, when the same individuals attempted to estimate answers, bilateral activity in the intraparietal sulci was linked to our more ancient and language-independent ability to approximate. Such an ability appears early in development. Even at six months, most of us can approximately differentiate between large numbers of items for ratios of between 1:2 and 2:3 (Starkey and Cooper, 1980) and it seems that we share this approximate number sense with other animals (Boysen and Capaldi, 1993). Such early mathematical ability is likely to have a critical role in ‘bootstrapping’ our capacity to formally grasp exact differences and procedures (credited to Spelke and Carey in Johnson, 2004; see also Carey, 2004), and should provide an improved basis for developing initial mathematical understanding of number and for remediating pupils’ difficulties in achieving this. For example, dyscalculia has been linked to a deficit in our ‘premathematical’ estimation abilities (Butterworth, 2008) and a study of low birth-weight adolescents with numerical difficulties revealed less gray matter in an area of the intra-parietal sulcus (Isaacs et al, 2001). Further research is needed to confirm the direction of cause and effect in such studies, but insights from brain imaging research are contributing to models of mathematical development useful in developing interventions. For example, in one intervention based on these concepts, it was demonstrated that children with dyscalculia showed considerable improvements in a broad range of calculation abilities when basic numerical and conceptual knowledge were focused upon at an early stage of mathematics education (Kaufmann et al, 2003). Dehaene has also used his own findings to develop educational software aimed at remediating dyscalculia. The software is based on the hypothesis that dyscalculia derives from a core deficit in number sense, or in relating number sense to numerical symbols (Wilson et al, 2006)

Another example of the potential influence of neuroscience on mathematics education comes from research into the role of fingers in early mathematical development. Since one might consider mathematics as a cognitive ability, the use of fingers to support calculation can be looked down upon as evidence of mathematical weakness. However, finger gnosis (being able to differentiate between different fingers in response to, say, one or more being touched) has been identified as a strong predictor of mathematical ability (Noel, 2005). Links between finger discrimination and mathematical ability have been studied in both children and adults. A functional Magnetic Resonance Imaging (fMRI) study has shown that, although behavioural outcomes can be the same, the activities produced when fingers are involved in approximating tasks vary with age (Kaufmann et al, 2008), suggesting their contribution varies with development. Eight year-old children produce an increase in activity in the intraprietal sulci when fingers are involved, but not adults. Kaufmann suggests fingers represent concrete embodied tokens involved in the estimation of number magnitude – an intimate involvement with our basic “number sense”. On this basis, children should not be discouraged from using fingers and teachers may be able to exploit their natural role more fully (Kaufmann, 2008). For example, in one intervention based on such ideas, new arrivals at three Belgian schools were identified as having poor finger gnosis and some of them received two-weekly 30 minute sessions of finger training for eight weeks (Gracia-Bafalluy and Noel, 2008). After training, these children were significantly better at quantification tasks than those who had not received the training.

Research on the relationship between our “animal number sense” and the early learning of mathematical concepts has only just begun. However, the concepts emerging are sufficiently well established and different to previous ideas of mathematical development to suggest the potential for improved educational practice in primary school classrooms in the next 1-2 decades.

Adolescence

It may be natural to consider that a teenager is essentially a young adult, with a fully formed brain but lacking the social experiences of his/her elders. However, scientific investigation has revealed a very different picture, with frontal and parietal regions still undergoing radical structural changes until the late teens, relative to other areas of the brain which appear more fully developed. Synaptic pruning (the cutting back of neural connections) and myelination (improving the efficiency of neural connections) also continue throughout adolescence (Sowell et al, 2003) in frontal regions. Such change in specific regions suggests the teenage brain may be less ready than an adult brain to carry out a range of specific processes, including directing attention, planning future tasks, inhibiting inappropriate behaviour, multitasking, and a variety of socially-orientated tasks. Although more research is needed, some psychological research backs this up, even showing a “pubertal dip” in some areas of performance, such as matching pictures of facial expressions to descriptors. In this task, 11-12 year olds perform worse than younger children (McGivern et al, 2002). Discontinuities have also been shown in abilities underlying social communication, such as taking on the viewpoint of another person, or so-called ‘perspective-taking’ (Blakemore and Choudhury, 2006; Choudhury et al, 2006). Some parts of what might be described as a “social brain” network are also activated differently in teenagers compared with adults when thinking about intentions (Catherine et al, 2008) and brain regions responsible for the control of impulses appear less well functionally connected in adolescents’ than in adults’ brains (Steven et al, 2007). Teenagers also appear to activate different areas of the brain from adults when learning algebraic equations, with this difference associated with a more robust process of long-term storage than that used by adults (Luna, 2004; Qin et al, 2004). Adolescents, then, are not simply older children or younger adults, and cognitive development cannot be expected to proceed in a continuous linear manner. Apart from explaining some of the difficulties teenagers experience, such changes also suggest how and why adolescence can be a potentially sensitive period for learning, within and beyond academic contexts. For example, teenagers often tend to perceive risks as smaller and more controllable than adults, and they are generally more vulnerable than adults or children to a range of activities which are inappropriately risky, such as gambling and drug taking. Appropriate decision making appears to require a balanced engagement between harm-avoidance and reward orientating processes that is regulated by processes within the prefrontal cortex, where teenage development may lag (Ernst et al, 2005). Imaging studies comparing adults and adolescents show reduced activity in these prefrontal areas when making risk-based decisions (Bjork et al, 2007; Eshel et al, 2007), and this reduced activity correlates with greater risk-taking performance (Eshel et al, 2007). Such studies provide new insights into how adolescent risk taking may be linked to neuro-maturational events and these insights may influence educational perspectives on teenage behaviour, helping to understand a potentially problematic, and sometimes even dangerous, period of children’s development (Baird et al, 2005).

It seems likely that these and future findings from neuroscience may generate new educational approaches in future years (eg strategies that take a more informed account of the temporary lagging of cognitive function in some areas). Findings on the adolescent brain are potentially illuminating and should stimulate educational changes, although the rate at which these will occur is difficult to predict. Paus believes that “the time is right for evidence-based, large-scale studies of interventions aimed at facilitating youth development. Neuroimaging-based approaches hold considerable promise, providing both the evidence as well as novel insights about the role of the environment in shaping the adolescent brain” (Paus, 2008). However, do such beliefs reflect an awareness of the considerable ethical, social and political issues that would be involved with such studies? Neuroimaging studies of interventions aimed at remediation of difficulties associated with dyslexia have already been carried out, but remediation of character (one of the five “c”s of positive youth development (Lerner, 2005)) would take neuroethical dilemmas to a new level.

Motivation

A burgeoning number of findings from neuroscience has supported some fresh educational thinking about motivation, including the type of intense engagement provided by computer gaming (Gee, 2003). Previous explanations of gaming motivation involve issues of fantasy, challenge and curiosity (Malone, 1981) but these appear inadequate in explaining the attraction of some traditional games such as “snakes and ladders” and “bingo”, or simple (but popular) computer games such as Tetris. This attraction may be due more to elements of chance-based uncertainty. The attraction of uncertainty is now gaining closer neuroscientific investigation, but it is a phenomenon well established by psychological experimentation (Atkinson, 1957) which has shown moderate risk taking (50% chance) heightens motivation.

Recent neuropsychological understanding of reward involves consideration of ‘wanting’ and ‘liking’ as two dissociable components, with the wanting of a reward being coded by levels of dopamine release in mid brain areas (Berridge and Robinson, 2003). The predictability of an outcome has been shown to influence this activity. In primates, it has been shown that maximum dopamine is released when the likelihood of receiving reward for success is about half way between totally unexpected and completely predictable, ie 50% likely (Fiorillo et al, 2003). Dopamine levels in this area of the human brain have been linked to our motivation to pursue a variety of pleasures, including sex, food, gambling (Elliot et al, 2000) and computer gaming (Koepp et al, 1988). The link between the predictability of an outcome and mid-brain dopamine activity is, therefore, helpful in explaining why humans are so attracted to activities involving elements of chance (Shizgal and Arvanitogiannis, 2003). Activity in this area has been studied non-invasively in humans during gaming using fMRI. These fMRI studies have shown that patterns of dopamine activity are predicted less by reward in ‘real’ absolute terms and seem more to do with winning the particular game being played. Activity can increase with reward size (Knutson et al, 2001) but, rather than being proportional to absolute monetary reward, activation peaks at the same level for the best available outcome in different games (Nieuwenhuis et al, 2005). The complex relationship between reward and motivation is thus strongly mediated by context.

When uncertainty is encountered in more real world instances, there are potentially more complex effects of context created by the social environment. These are illustrated by the way our natural attraction to uncertainty falls off when the task is perceived as educational. Students generally prefer low levels of academic uncertainty and choose problems well below moderate (<50%) challenge (Clifford, 1988; Harter, 1978) unless these are presented as games, when students will take greater risks (Clifford and Chou, 1991). This may suggest that individuals can be deterred from tackling academic tasks with higher levels of uncertainty due to the implications of failure for social status and esteem. In research involving classroom-based applications, these concepts provide a means to understand how learning games with elements of pure chance can increase uncertainty without impacting negatively on self-esteem, thereby raising motivation (Howard-Jones and Demetriou, in press – now published on-line). Neurocomputational modelling is now providing the tools to study the reward system in more detail (Elliott and Deakin, 2008).These techniques allow estimation of how dopaminergic activity in the reward system varies with the progress of a game. Such activity also mediates attention in the short term, and these models can predict when declarative learning (the type of greatest interest to educators) will occur during an educational game (Howard-Jones et al, 2009).

Neuroscience is providing concepts that are proving useful in understanding learning games and motivation in the classroom, particularly amongst males (Hoeft et al, 2008). However, introducing chance-based uncertainty into learning can conflict with the principle of reward consistency that is traditionally valued by education (OfSTED., 2001). Our increasing understanding of this region of the brain has, therefore, the potential to prompt a significant departure from present educational thinking.

Early screening for some developmental disorders

Event-Related Potentials (ERPs) refer to a set of distinct electrical signals emitted by the brain and detectable using a non-invasive technique involving the attachment of electrodes to the scalp. Some ERP waveforms of newborn infants have been identified that can differentiate between children who will later, at eight years old, be poor readers or be dyslexic (Molfese, 2000). Measurement of ERPs has been shown as an effective method of predicting dyslexia in new-borns with and without a family history of dyslexia (Guttorm et al DATE) and such techniques could form the basis of very early screening, so that children at risk of dyslexia are able to benefit as quickly as possible from suitable intervention(s). See also discussion by Friedrich of neural markers and specific language impairment (Friedrich, 2008). Such techniques and possibilities are not limited to literacy. Another type of ERP has been identified that is sensitive to children’s response to numerical distance (Szucs et al, 2007) that may be a helpful neural marker for magnitude processing in infancy. This signal may provide an early indicator of later educational risk in respect of mathematics.

The use of neural markers to provide very early detection of educational risk is an area identified by Goswami where a neuroscience approach may provide particular promise for education (Goswami, 2008).

Cognition and the Brain in the Curriculum: Curriculum Aims and Content

i) The influence of research on cognitive training (‘Brain training’)

There is increasing evidence to show that the cognitive training can reduce risk of Alzheimers (Wilson et al, 2002) and in normally functioning older adults. A 5-year study has shown that training can provide sustained improvements in a range of cognitive functions in this age group (Ball et al, 2002).

A study by Willis et al (2006) showed sustained improvements in targeted function over five years, following an intervention that consisted of only 10 sessions of about 60-75 minutes each. Positive effects were also been observed on daily functioning (phone, laundry, cooking etc) (Willis et al, 2006).

There is thus good evidence to show that brain function can be trained, in the sense that repeated practice on exercises that focus on a cognitive function, can produce improvement in that cognitive function. There is less evidence confirming impact on everyday functioning not specifically targeted by the training. As research continues, however, such evidence is emerging (Mahncke et al, 2006; Willis et al, 2006) and we are understanding more about the impact of training cognitive function on other, non-targeted areas. Most notably, it has been convincingly shown that fluid intelligence, which is seen as a good predictor of professional and academic achievement, can be improved by rehearsing working memory exercises, specifically an exercise based on the N-back test (Jaeggi et al, 2008). Unlike most other studies which have been undertaken with older participants including those at risk of dementia, the average age of the participants in this study was 25, demonstrating the relevance of this type of cognitive training to the younger population.

In terms of developing cognitive function amongst children, it is the targeting of WM, together with the closely allied concept of attention that has again produced the most interesting results. In a study involving children with ADHD, training of WM was found to successfully transfer to non-targeted areas of behaviour, producing improved complex reasoning skills and reduced parental ratings of ADHD symptoms (Klingberg et al, 2005). Training of visual and auditory attention has been found to benefit literacy achievement for children with dyslexia (Chenault et al, 2004), and a study using ERPs with children with specific language impairment (SLI) showed that neural mechanisms of selective auditory attention and the associated language difficulties can be remediated through auditory attention training (Stevens et al, 2008).

The research from science converges with other forces encouraging educators to move further away from content towards thinking skills and, more specifically, the training of cognitive function. Advances in technology are likely to continue improving our access to information, with some commentators believing this will increase the need for specialisation as it increasingly places “any human knowledge at the fingertips of any human” (Stewart, 2008). Such advances may, therefore, make it desirable for learners to be better at manipulating information than encoding and recalling it, with demand for specialisation making it more difficult to predict and include the type of content that an individual may benefit from in the initial stages of their education. These factors may combine with a burgeoning dialogue with neuroscience that also encourages greater emphasis upon cognitive function generally within education – since cognitive function is a central construct of cognitive neuroscience – and some way towards a redefinition of the aims of education as an attempt to “nurture” the brain and its processes. Increasing interest in the training of cognitive function as a means to enhance learning potential is also reflected in current public interest in “brain training” products, although it should be noted that no quality research exists that evaluates the claims made by the manufacturers of these products, or even the design principles upon which they are based. This public enthusiasm with the commercial products has already begun to extend itself to some schools.

ii) Teaching about the brain

There has also been a broader interest in the development of children’s cognitive function, in ways that include emotional aspects of behaviour. These include the development of “Executive Function” (EF) – an umbrella term referring to the underlying processes responsible for children’s ability to direct, maintain and focus their attention, manage impulses, self-regulate behaviour and emotion, plan ahead and demonstrate flexible approaches to problem solving. EF skills are predictive of academic achievement (Bull et al, 2008), and social and emotional development (Hughes, 1998). For this reason, attempts have been made to find ways of developing EF skills and some interventions in schools have reported positive results in terms of improved behaviour (Greenberg, 2006).

Unlike simple cognitive training, such programmes require learners to understand and reflect upon their behaviour in terms of a set of mental processes. They are thus delivering an explicit, if sometimes ill-defined, psychological content into the curriculum of many schools. These programmes are becoming associated with protecting the mental health of children as well as improving academic standards and behaviour. They have been explicitly linked to neuroscience by some experts (Greenberg, 2006), although (in the opinion of the author) such links would benefit from further scientific scrutiny and consensus. What does appear clear, however, is that the school-based curriculum education is becoming influenced by attempts to directly attend to the development of executive function, in order to promote emotional well-being, mental health and academic achievement. It has been known for some time that the level of education can influence mental health in later life, but there are now growing voices for educators to become actively involved in fostering the mental health of their learners. A 2005 policy paper produced by the Sainsbury foundation (together with the NHS confederation) predicts “By 2015, mental wellbeing should be a major concern for schools, from dedicated classroom time to the overall approach of the school towards its pupils and staff” (SCMH, 2005, p13.

It can also be predicted that advances in neuroscientific understanding may broaden the aims of education further still, thereby influencing the curriculum and the ways in which it is delivered. For example, rising levels of obesity amongst children has drawn new attention to the importance of exercise in schools, but this may gain further emphasis as neuroscientific understanding emerges about the processes by which exercise is linked to learning (Hillman et al, 2008). Understanding the processes by which even short bouts of exercise improve subsequent learning (Winter et al, 2007) makes it foreseeable that regular exercise breaks during the school day will become more popular as a means of raising academic standards and fostering mental and physical health. Although it is certainly not a good example of science, evidence for the likelihood of such developments arises from the popularity of Brain Gym. The principles of Brain Gym are unscientific and bizarre (Hyatt, 2007), but its popularity in the face of unfavourable media exposure must surely derive, in part, from its associations with academic achievement and neuroscience. Such associations, in the case of Brain Gym, do not withstand scrutiny, especially in terms of neuroscience, but it is likely that the authentic value of exercise in learning will become well understood by educators in future years. It can be predicted that, by 2025, neuroscience will have contributed to developing scientifically sound and educationally evaluated methods of incorporating frequent exercise breaks into the school day and these will have become established in most schools.

Part of the success of introducing such elements into the curriculum will depend on ensuring learner motivation, and this may depend on learners understanding something of neurocognitive function. Such understanding may have other benefits. It has been reported that providing learners with a basic knowledge of the brain can, by itself, provide significant help in improving self-image and academic achievement (Blackwell et al, 2007). In this study with adolescents, researchers informed learners about the structure and function of their brain, how learning changes the brain by producing new neuronal connections and about brain plasticity, and provided the clear message that the pupils themselves were in charge of this process. This promoted a positive change in classroom motivation. Grades for the control group, who had not received the intervention, continued downward while this trend was reversed for the intervention group.

Neuroscience-related issues arriving without invitation

The potential changes so far discussed are expected come about through initiatives involving educators, and their final form is likely to be mediated by educational understanding, sensitivities and opinions. Some influences involving neuroscience, however, may arrive without invitation.

One such issue is the use of cognitive enhancers. In the US, students are increasingly using prescription drugs in order to provide cognitive enhancement and thereby support their studies. Usage varies widely from one university to the next, with an average figure of 6.9% of students indulging in non-medical use of prescription stimulants (McCabe et al, 2005). In another study, however, that surveyed 1811 students at a large Southeastern US university, 34% reported the illegal use of ADHD stimulants (eg methylphenidate), mostly to improve their cognitive function during periods of fatigue and stress (DeSantis et al, 2008).

The production of new and stronger drugs for cognitive enhancement is likely to increase, driven partly by efforts to combat the effects of Alzheimer’s disease. One such drug is donepezil (marketed as Aricept) that increases levels of acetylcholine (ACh). ACh is thought to modulate the rate at which neural connections adjust themselves when learning, with increases in ACh thus able to bring about increases in learning rate. Donepezil reduces cholinesterase that mops up ACh, thereby increasing levels of ACh and improving cognitive function, including memory, amongst those suffering from Alzheimers (Roman and Rogers, 2004). The potential value of this drug for other users was demonstrated in a study that administered donepezil to healthy young adults for only 30 days, and revealed significant improvements in episodic memory performance (Gron et al, 2005).

Scientists have been speaking out in positive terms about the “new enhancement landscape” for healthy adults (Gazzaniga, 2005). In a recent article in Nature, one group of scientists suggested that the growing demand for cognitive enhancement should be responded to, and that the response should begin by “rejecting the idea that enhancement is a dirty word.”(Greely et al, 2008) However, a modest UK consultation showed more ambivalence (Horn, 2008), with concerns raised that included

• possible side and long term effects including personality change
• the devaluation of ‘normal’ achievement and the intrinsic value of effort and motivation in learning
• inequality if such drugs are expensive
• pressure to use such drugs and the exacerbation of an already over-competitive culture.

It seems likely that the use of cognitive enhancers amongst the general UK population will increase as public sensitivity diminishes and the drugs become more socially acceptable. This will stimulate significant ethical debate amongst educators. Some educational institutions may, with parental permission, choose to introduce drug testing. Since there are few clear precedents for the issues involved with these drugs (it is debatable whether a comparison with the use of drugs in sport is helpful) there may, for some time, exist a diverse range of attitudes and practices amongst learners and educational institutions in respect of cognitive enhancers. This may impede the development of any necessary legislation.

The influence of neuroscience on educational professional development

Psychology returns, accompanied by some neuroscience, to initial teacher training

Our burgeoning understanding of the human brain and mind, particularly in regard to the issues raised above, will increase the likelihood of some elements of psychology returning to initial teacher training. When teachers consider learning processes, they consider mental rather than neural processes, although in the future we can expect constructions about these mental processes to be increasingly informed by neurobiological understanding. Neuroscience, in itself, remains largely meaningless in educational terms, except insofar as it informs our psychological concepts about the learner’s mind. However, neuroscientific concepts are helpful in formulating and communicating educationally-relevant concepts about the mind (Howard-Jones, 2008).

The emergence of neuro-educational research and new types of educational professional

Although appearing under many different names (eg educational neuroscience, neuroeducational research, neuroscience and education) a field of research at the interface between neuroscience and education is becoming established. This provides some support for the prediction made by some that, in the future, hybrid professionals will emerge with expertise in both neuroscience and education (Pickering and Howard-Jones, 2007; Szucs and Goswami, 2007).

Developments that may occur in the future, but not by 2025

Genetic profiling in mainstream education

The area where genetic knowledge is first likely to impact, and appears likely to do so before 2025, is in the area of learning difficulties. Gene-based diagnoses of learning difficulties will be able to predict general learning difficulty as well as difficulties within specific areas such as maths (Plomin, 2008). Such very early predictions, combined with emerging educational understanding of effective interventions, will provide the soonest possible implementation of appropriate interventions. Genetic knowledge will provide opportunities for new levels of personalised learning and these should ameliorate or even prevent the manifestation of some learning difficulties.

Ultimately, genetic knowledge should allow educational programmes to be better tailored to suit all individuals according to their genetically-informed educational profiles. It has been suggested that, in the future, “Educogeneticists” will be able to provide informed recommendations to schools and families about how a child’s education may be planned in order to optimise academic outcomes (Grigorenko, 2007). Genetics may, therefore, be considered to have considerable educational potential beyond the early identification and amelioration of learning difficulties. This wider application, however, will only add further controversy to a plethora of ethical issues and questions about using genetic knowledge in education: what may result when genetic testing proceeds without full understanding of the educational intervention that may be needed? Who makes the decisions about testing and interventions, and by what processes? What precautions are needed to prevent this new educational opportunity feeding demand for genetic engineering and eugenics?

Since biotechnology companies are now marketing genetic tests directly to the public, it is possible that this issue may arrive without educational invitation by 2025, as with cognitive enhancers. One can imagine that, by 2025, all parents will have the opportunity to independently purchase a genetic profile of their child and ask what their school intends to do about it. However, whilst the educational application of smart pills requires only that a bottle is opened and a pill consumed, there remains much educational (or “edu-genetic”) research necessary in order to utilise genetic knowledge in mainstream education. This knowledge gap and the ethical issues mentioned above will provide some barrier to progress. Due to the potential benefits to individual learners, public attitudes are likely to become increasingly positive and demands for edugenetic approaches will grow, but it seems unlikely that schools will develop established approaches to genetically-based differentiation of mainstream teaching and learning by 2025.

Brain-Computer interfaces (BCIs) in mainstream education and ‘brain reading’

Brain-computer interfaces have developed the potential to provide valuable aid to some profoundly disabled individuals. For example, severely paralysed patients can control prosthetic limbs and computer cursors by thought alone. Non-invasive approaches often adapt the type of technology used by ERP and EEG measurements, using a patient’s electrical brain activity sensed by electrodes placed on the scalp. The EEG/ERP signal is analysed and interpreted automatically by a computer, which then produces an appropriate output. In this way, the user can generate some rudimentary signal to the outside world by producing the pre-defined type of thought that the computer is programmed to decode. For example, by imagining different body motions (eg left versus right hand), the user can generate different EEG signals that can be translated by the computer into responses to binary (yes/no) questions with high, but not perfect, accuracy (Neuper et al, 2006). As illustrated in this example, there are presently significant limitations upon the amount of information such BCI interfaces can communicate. These limitations arise from the noisiness of the signal and the variability of the signals produced by different individuals. Improvements in technology alone may not be sufficient to overcome these limitations, since significant advances may require greater neurobiological and psychological understanding of the signals themselves.

At present then, although the usefulness of non-invasive BCIs for the profoundly disabled may be possible in the next 1-2 decades, the likelihood of the wider population using them to communicate with everyday technology is something for the very far distant future when both our technology and, perhaps more importantly, our understanding of brain function is unimaginably superior than at present.

Similar issues limiting the advances in BCI’s over coming years will also apply to ‘brain reading’. Some experiments have shown that imaging technology can reveal socially sensitive and relevant information, such as racial group identity and unconscious racial attitudes. For example, white subjects with more negative evaluations of black faces showed increased amygdalic activity in response to unfamiliar black, compared with white, faces. There has also been some success in identifying the correlates of deception (Nunez et al, 2005) and such knowledge may be applied in counter-terrorism efforts in the future. Such examples, however, tend to compare a very small set of conditions, and results could only be used to differentiate between a correspondingly small set of possible alternatives regarding mental content (eg truth/lies). As with the notion of using a BCI to access one’s computer in an everyday sense, the possibility of reading the everyday contents of a learner’s mind will be science fiction until well beyond 2025. As Farah suggests (p1126) in relation to brain-reading, “even a major leap in the signal-to-noise ratio of functional brain imaging would leave us with gigabytes of more accurate physiological data whose psychological meaning would be obscure.” (Farah, 2002)

Improbable educational advances involving neuroscience

There are considerable differences between the views of educators and scientists about how neuroscience is relevant to education, although consensus is emerging in both areas that the relevance exists. Perhaps unsurprisingly, educators associate neuroscience chiefly with those brain-based concepts already commonly found in schools (Pickering and Howard-Jones, 2007). Unfortunately, these concepts are often not well supported by existing science, yet educational expectations of how neuroscience may influence education in the future are likely to be strongly influenced by these neuromyths. This was illustrated in 2000 when scientists Uta Frith and colleague Sarah-Jayne Blakemore were commissioned by the Teaching and Learning Research Programme (TLRP) to carry out a review of neuroscientific findings that may be of relevance to educators (Blakemore and Frith, 2000). This review attacked a number of unscientific myths and highlighted some scientific areas of potential interest to educators. In January 2001, to promote further discussion about a possible research agenda, the TLRP wrote to scientific and educational institutions, asking for comments on the report by Blakemore and Frith. Respondents were particularly asked to provide “identification of key research questions, … their priority … and estimate of their tractability (in terms of return on research effort)”. While scientists indicated areas such as learning disorders, memory and plasticity, educational respondents identified areas such as multiple intelligences and learning styles, which are problematic as robust and well-defined scientific concepts suitable for orientating neurocognitive research. The report on the consultation concluded that no collaborative research agenda had yet emerged (Des Forges, 2001).

Indeed, perhaps the most immediate benefit of the increased dialogue between neuroscience and education has been to highlight the large number of neuromyths that have developed within education. Some of the most prominent will now be listed below, none of which are likely to receive support from neuroscience in the coming years and should become less prominent by 2025, despite the hopes of many educators.

Multiple Intelligences (MI) Theory

Gardner’s MI theory proposed that, rather than a single all-purpose intelligence, it is more useful to describe an individual as possessing a small number of relatively independent intelligences (Gardner, 1983). Possible candidates for these intelligences include linguistic, musical, logical-mathematical, spatial, bodily-kinaesthetic, intrapersonal sense of self, interpersonal and Gardner has later proposed other possibilities such as naturalistic and existential intelligence (Gardner, 1999). MI theory is in direct opposition to the idea of a unitary general intelligence factor ‘g’, reflecting overall brain efficiency and the close interconnection of our mental skills. MI theory resonates with many educators, who see it as a robust argument against IQ-based education.

In a critical review of the evidence for MI theory, Waterhouse examined the empirical scientific evidence (Waterhouse, 2006). MI theory claims to be drawn from a wide range of disciplines including neuroscience. Indeed, Gardner has claimed “accumulating neurological evidence is amazingly supportive of the general thrust of MI theory”. However, Waterhouse points out that the general processing complexity of the brain makes it unlikely that anything resembling MI theory will ever emerge from neuroscience. Cognitive neuroscience is exploring the brain in terms of processes (vision, hearing, smell, etc) but not in terms of seeing intelligence, auditory intelligence or smelling intelligence. In the realm of neuroscience, it neither appears accurate or useful to reduce the vast range of complex individual differences at neural and cognitive levels to any limited number of capabilities.

Despite the absence of MI theory in the neuroscience literature, teachers heavily associate MI theory with neuroscience. (To confirm this, the author returned to the data collected from the 150 UK teachers who were asked the question “Please list any ideas that you have heard of in which the brain is linked to education” (Pickering and Howard-Jones, 2007). Of those teachers who responded to this question (121), most listed no more than 2-3 ideas. Of these, MI theory occurred 17 times (14%)).

Thus, in educational terms, MI theory appears like a liberator – providing teachers with the ’scientific’ license to celebrate diversity. In terms of the science, however, it seems an unhelpful simplification as no clearly delineated, limited set of capabilities arises from either the biological or psychological research.

Learning Styles

In educational terms, an individual’s learning style can be considered as a set of learner characteristics that influence their response to different teaching approaches. A survey in 2004 identified 71 different models of learning styles (Coffield et al, 2004) and our own survey showed almost a third of UK teachers had heard of learning styles, with most of those who used this approach reporting it as effective (Pickering and Howard-Jones, 2007). As with MI theory, which is also often interpreted by educators as a means to identify preferred modes of learning, the promotion of learning styles has benefited from a strong association with neuroscience. Many learning style models have a distinctly biological justification, with one of their major proponents, Rita Dunn, commenting that “at least three fifths of style is biologically imposed” (Dunn et al, 1990).

Very many educational projects have pursued improvement through tailoring programmes to meet individual learning styles but, as yet, there is no convincing evidence that any benefit arises. A review of such studies, concluded that matching instruction to meet an individual’s sensory strengths appears no more effective than designing content-appropriate forms of education and instruction (Coffield et al, 2004). Perhaps the best known inventory of learning styles within education is the one categorising individuals in terms of their preferred sense modality for receiving, processing and communicating information: visual, auditory or kinaesthetic (VAK). In a laboratory study of memory performance, participants’ own self assessment of their VAK learning style was shown to be out of line with more objective measures, and memory scores in different modalities appeared unrelated to any measure of dominant learning style (Kratzig and Arbuthnott, 2006). There was, instead, evidence that participants’ self-rating as kinaesthetic learners was related to visual performance, that they were self-rating their learning styles in ways possibly promoted by the inventory itself, and objective evidence from memory testing that suggested visual and kinaesthetic/tactile tasks were tapping the same underlying memory process. The authors concluded that educators’ attempts to focus on learning styles were “wasted effort”.

In approaches such as VAK, the implicit assumption appears to be that, since different modalities are processed independently in different parts of the brain, differences in the efficiency of these parts results in a clear modality-based method of classifying how learners are able to process information most efficiently. However, as pointed out by Geake, this flies in the face of what we know about the interconnectivity of the brain(Geake, 2008). Geake refers to a recent piece of experimental research in which five year olds showed themselves able to distinguish between groups of dots even when the numbers were too large for counting (Gilmore et al, 2007). They were then asked to repeat the task in auditory mode by counting clicks, and reproduced almost identical levels of accuracy. Geake points out that this is because input modalities in the brain are very interlinked. As yet, no evidence arising from neuroscience, or any other science, supports the educational usefulness of categorising learners in terms of their sensory modality or any other type of learning style. In the meantime, educators continue to be drawn to VAK as means to implement a type of differentiation between learners.

Learning styles based on these ideas are likely to diminish in their popularity as awareness grows of their ineffectiveness and lack of scientific basis. By 2025, one can be hopeful that differentiation of learners may be informed by a better understanding of the development of literacy, mathematical and other skills, and this understanding will undoubtedly be informed by insights from cognitive neuroscience. More speculatively, it is also possible in the more distant future (but not by 2025 – see above) of genetically-informed profiling of individual learners.

Left-Brain Right Brain

Another popular way of categorising learning style is in terms of “left-brain right-brain” theory(Springer and Deutsch, 1989). According to this theory, learners’ dispositions arise from the extent to which they are left or right brain dominant. It is true that some tasks can be associated with extra activity that is predominantly in one hemisphere or the other. For example, language is considered to be left lateralised. However, no part of the brain is ever normally inactive in the sense that no blood flow is occurring. Furthermore, performance in most everyday tasks, including learning tasks, requires both hemispheres to work together in a sophisticated parallel fashion. The division of people into left-brained and right-brained takes the misunderstanding one stage further. There is no reliable evidence that such categorisation is helpful for teaching and learning.

Educational Kinesiology (Brain Gym)

Educational kinesiology (or Edu-K, also often sold under the brand name of Brain Gym) was developed by Paul and Gail Dennison as a means to ‘balance’ the hemispheres of the brain so they work in an integrated fashion and thus improve learning (Dennison, 1981). Whatever the flaws in its theoretical basis (which are many and fatal), there is a lack of published research in high quality journals to make claims about the practical effectiveness of programmes such as Brain Gym to raise achievement. Of the studies published elsewhere, the lack of information about the exercises undertaken and/or the insufficient or inappropriate analysis of the results undermine their credibility (Hyatt, 2007). However, it may also be that programmes such as Brain Gym are contributing to learning, but for entirely different reasons than those used to promote them. As discussed above, there is an emerging body of multidisciplinary research supporting the beneficial effect of aerobic exercise on selective aspects of brain function of importance to education (Hillman et al, 2008). However, these advantages appear linked to the aerobic nature of the exercise, which is low in Brain Gym.

Implicit learning

Work with artificial grammars, in which participants are able to acquire grammatical rules by observing examples of artificial language, demonstrates our ability to learn implicitly, ie without being able to report explicitly what has been learnt. Such experiments have contributed to enthusiastic calls for more educational focus on implicit learning (Claxton, 1998). However, there are considerable barriers to the practical application of such ideas, making their usefulness to education questionable and causing some scientific authorities to label them a new source of neuromyth (Goswami, 2004). A non-specialist interpretation of the phenomenon of implicit learning might involve ideas about absorbing information and concepts from the environment without attending to them, but such ideas have no scientific basis. For example, in the artificial grammar scenario, formal rules may be acquired without the learner consciously formulating them, but the learner must pay considerable attention to the examples of artificial language in order to facilitate this. In a more real world context, we may also implicitly develop understanding about, for example, the motivations of people around us, without being able to articulate how we have achieved this. Again, however, this is only possible by paying attention to the behaviour of those people. “Implicit learning” does not equate to “learning without attention”, and it seems unlikely that such concepts will become usefully applied in education in the coming years.

A brain-based science of learning?

Although researchers at the interface of neuroscience and education have done much to counter the neuromyths prevailing in schools, they may also be guilty of inadvertently creating one. There has been much enthusiasm amongst policy makers for the creation of a “new” science of learning (OECD, 2002; OECD, 2007). This may be because neuroscience seems a more secure basis for learning theory, with its images of blood flow appearing more concrete than abstract psychological concepts. Indeed, it has been experimentally determined that including references to the brain (even irrelevant ones) increases the satisfaction of a reader (Weisberg et al, 2008). However, as with social and experiential perspectives on learning, the biological perspective is, on its own, limited in terms of what it can tell us. A science of teaching and learning which is chiefly based upon the brain is unlikely to develop in the foreseeable future, because neuroscientific perspectives struggle with many concepts (such as meaning and autonomy) that are central to educational aims and understanding (Howard-Jones, in press). On the other hand, this brief review has emphasised that greater inclusion of biological perspectives in educational thinking, alongside other perspectives, is increasingly desirable and probable.

Summary

It is anticipated that the following educational developments involving neuroscience may arrive by 2025:

• New educational approaches will become established for the teaching and learning of mathematics in the early years, as a result of insights from cognitive neuroscience.
• Adolescents will become recognised as a more distinct group of learners and educational approaches will be developed that are better tailored to meet their social, emotional and educational needs
• A new understanding of motivation will be developed and new approaches to engaging learners will become established (eg in areas involving the use of games) informed by insights into the brain’s reward system
• Early screening will be available for a range of learning disorders, using neural markers and genetic testing.
• Attendance to the training of some targeted cognitive functions, including working memory, will feature across year groups in the National Curriculum
• The reflective understanding and development of executive function will feature in the National Curriculum for young learners
• Understanding of mental health issues will become a stronger feature of the curriculum, as the aims of education become broader. This will include a basic understanding of brain function, with the associated academic benefits that such an understanding may bring.
• Exercise breaks will become a feature of the curriculum, as the link between exercise and academic achievement becomes clearer, and the UK struggles with increasing levels of obesity.
• The use of drugs to enhance cognitive function will become commonplace, remaining chiefly unchecked by legislation as the government remains unwilling to intervene in the absence of clear public consensus. The attitudes and practices amongst different groups of learners and educational institutions will diverge.
• Psychology, and some neuroscience, will become an established feature of teachers’ professional development and training.

• There will be no brain-based science of learning that is meaningful in educational terms, but a new field of neuro-educational research will become established, together with the development of professionals trained in both education and the relevant natural sciences (eg cognitive neuroscience, genetics). Biological perspectives will become an increasingly important component of educational understanding, practice and policy making.

References

Atkinson, J.W. (1957) Motivational Determinants of Risk Taking Behaviour. Psychological Review, 64, pp.359-72.

Baird, A, Fugelsang, J. and Bennett, C. (2005) “What Were You Thinking”? A Neural Signature Associated with Reasoning in Adolescence. Journal of Cognitive Neuroscience, S, pp.193-93.

Ball, K., Berch, D.B., Helmers, K.F., Jobe, J.B., Leveck, M.D., Marsiske, M., Morris, J.N., Rebok, G.W., Smith, D.M., Tennstedt, S.L., Unverzagt, F.W. and Willis, S.L. (2002) Effects of Cognitive Training Interventions with Older Adults:A Randomized Controlled Trial. Journal of American Medical Association, 288 (18), pp.2271-81.

Berridge, K.C. and Robinson, T.E. (2003) Parsing Reward. Trends in Neurosciences, 26 (9), pp.507-13.

Bjork, J.M., Smith, A.R., Danube, C.L. and Hommer, D.W. (2007) Developmental Differences in Posterior Mesofrontal Cortex Recruitment by Risky Rewards. Journal of Neuroscience, 27 (18), pp.4839-49.

Blackwell, L.S., Trzesniewski, K.H. and Dweck, C.S. (2007) Implicit Theories of Intelligence Predict Achievement across an Adolescent Transition: A Longitudinal Study and an Intervention. Child Development, 78 (1), pp.246-63.

Blakemore, S.J. and Choudhury, S. (2006) Development of the Adolescent Brain: Implications for Executive Function and Social Cognition. Journal of Child Psychology and Psychiatry, 47 (3), pp.296-312.

Boysen, S.T. and Capaldi, E.J. (1993) The Development of Numerical Competence: Animal and Human Models. Hillsdale, N.J., Erlbaum.

Bull, R., Espy, K.A., and Wiebe, S.A. (2008) Short-Term Memory, Working Memory, and Executive Functioning in Preschoolers: Longitudinal Predictors of Mathematical Achievement at Age 7 Years. Developmental Neuropsychology, 33 (3), pp.205-28.

Butterworth, B. (2008) Dyscalculia. Mental Capital and Wellbeing, State-of-Science Reviews. London, Government Office for Science.

Carey, S. (2004) Bootstrapping and the Origins of Concepts. Daedalus, pp.59-68.

Catherine, S., Burnett, S. and Blakemore, S. (2008) Neuroscience of Social Cognition in Teenagers: Implications for Inclusion in Society. Mental Capital and Wellbeing, State-of-Science Reviews. London, Government Office for Science.

Chenault, B., Thomson, J., Abbott, R.D. and Berninger, V.W. (2004) Effects of Prior Attention Training on Child Dyslexics’ Response to Composition Instruction. 24th Annual Meeting of the National-Academy-of-Neuropsychology. Seattle, WA., Lawrence Erlbaum Assoc Inc., pp.243-60).

Choudhury, S., Blakemore, S.J. and Charman, T. (2006) Social Cognitive Development During Adolescence. Social Cognitive and Affective Neuroscience, 1, pp.165-74.

Claxton, G. (1998) Knowing without Knowing Why. The Psychologist, 11 (5), pp.217-20.

Clifford, M.M. (1988) Failure Tolerance and Academic Risk-Taking in Ten- to Twelve-Year-Old Students. British Journal of Educational Psychology, 58, pp.15-27.

Clifford, M.M. and Chou, F.C. (1991) Effects of Payoff and Task Context on Academic Risk Taking. Journal of Educational Psychology, 83 (4), pp.499-507.

Coffield, F., Moseley, D., GHall, E. and Ecclestone, K. (2004) Learning Styles and Pedagogy in Post-16 Learning: A Systematic and Critical Review (Report No. 041543). London, Learning and Skills Research Centre.

Dehaene, S., Spelke, E., Pinel, P., Stanescu, R. and Tsivkin, S. (1999) Sources of Mathematical Thinking: Behavioral and Brain-Imaging Evidence, Science, 284, pp.970-74.

Dennison, P.E. (1981) Switching On: A Guide to Edu-Kinesthetics. Ventura, California Edu-Kinesthetics.

DeSantis, A.D., Webb, E.M. and Noar, S.M. (2008) Illicit Use of Prescription ADHD Medications on a College Campus: A Multimethodological Approach. Journal of American College Health, 56 (3), pp.315-23.

Dunn, R., Sklar, R.I., Beaudry, J. and Bruno, J. (1990) Effects of Mismatching Students’ Hemispheric Preferences on Mathematics Scores. Journal of Educational Research and Extension, 83 (5), pp.283-88.

Elliot, R., Friston, K.J, and D., Raymond J. (2000) Dissociable Neural Responses in Human Reward Systems. Journal of Neuroscience, 20 (16), pp.6159-65.

Elliott, R. and Deakin, B. (2008) “Neuroscience of Human Reward.” Mental Capital and Wellbeing, State-of-Science Reviews. London, Government Office for Science.

Ernst, M., Pine, D.S. and Hardin, M. (2005) Triadic Model of the Neurobiology of Motivated Behavior in Adolescence. Psychological Medicine, 36, pp.299-312.

Eshel, N., Nelson, E.E., Blair, R.J., Pine, D.S. and Ernst, M. (2007) Neural Substrates of Choice Selection in Adults and Adolescents: Development of the Ventrolateral Prefrontal and Anterior Cingulate Cortices. Neuropsychologia, 45 (6), pp.1270-79.

Farah, M.J. (2002) Emerging Ethical Issues in Neuroscience. Nature Neuroscience, 5 (11), pp.1123-29.

Fiorillo, C.D., Tobler, P.N. and Schultz, W. (2003) Discrete Coding of Reward Probability and Uncertainty by Dopamine Neurons. Science, 299, pp.1898-902.

Friedrich, M. (2008) “Early Neural Markers of Language Learning Difficulty in German.” Mental Capital and Wellbeing, State-of-Science Reviews. London, Government Office for Science.

Gardner, H. (1983) Frames of the Mind: The Theory of Multiple Intelligences. New York, Basic Books.

Gardner, H. (1999) Intelligence Reframed. New York, Basic Books.

Gazzaniga, M.S. (2005) Smarter on Drugs. Scientific American: Mind, 32-37.

Geake, J.G. (2008) Neuromythologies in Education. Educational Research, 50 (2).

Gee, J.P. (2003) What Video Games Have to Teach Us About Learning and Literacy. London, Palgrave Macmillan.

Gilmore, C.K., McCarthy, S.E. and Spike, E. (2007) Symbolic Arithmetic without Instruction, Nature, 447, pp.589-91.

Goswami, U. (2004) Neuroscience and Education. British Journal of Educational Psychology, 74, pp.1-14.

Goswami, U. (2008) Neuroscience in Education. Mental Capital and Wellbeing, State-of-Science Reviews. London, Government Office for Science.

Gracia-Bafalluy, M. and Noel, M.-P. (2008) Does Finger Training Increase Young Children’s Numerical Performance? Cortex, 44, pp.368-75.

Greely, H, Sahakian, B., Harris, J., Kessler, R.C., Gazzaniga, M.S., Campbell, P. and Farah, M.J. (2008) Towards Responsible Use of Cognitive-Enhancing Drugs by the Healthy. Nature, 456, pp.702-05.

Greenberg, M.T. (2006) Promoting Resilience in Children and Youth – Preventive Interventions and Their Interface with Neuroscience. In: Lester (ed) Conference on Resilience in Children, Arlington, VA., Blackwell Publishing, pp.139-50.

Grigorenko, E.L. (2007) How Can Genomics Inform Education? Mind, Brain and Education, 1 (1), pp.20-27.

Gron, G., Kirstein, M., Thielscher, A., Riepe, M.W. and Spitzer, M. (2005) Cholinergic Enhancement of Episodic Memory in Healthy Young Adults. Psychopharmacology, 182 (1), pp.170-79.

Guttorm, T.K., Leppanen, P.H.T., Poikeus, A.-M., Eklund, K.M., Lyytinen, P. and Lyytinen, H. (YEAR) Brain Event-Related Potentials (Erps) Measured at Birth Predict Later Language Development in Children with and without Familial Risk for Dyslexia. Cortex, 41, pp.291-303.

Harter, S. (1978) Pleasure Derived from Cognitive Challenge and Mastery. Child Development, 45, pp.661-69.

Hillman, C.H., Erickson, K.I. and Framer, A.F. (2008) Be Smart, Exercise Your Heart: Exercise Effects on Brain and Cognition. Nature Reviews Neuroscience, 9, pp.58-65.

Hoeft, F., Watson, C.L., Kesler, S.R., Bettinger, K.E. and Reiss, A.L. (2008) Gender Differences in the Mesocorticolimbic System During Computer Game-Play. Journal of Psychiatric Research, 42, pp.253-58.

Horn, G. (2008) Brain Science, Addiction and Drugs. London, Academy of Medical Sciences.

Howard-Jones, P.A. (in press) Philosophical Challenges for Researchers at the Interface between Neuroscience and Education. Journal of Philosophy of Education.

Howard-Jones, P.A. (2008) Fostering Creative Thinking: Co-Constructed Insights from Neuroscience and Education. Bristol, Escalate.

Howard-Jones, P.A., Blakemore, S.J., Samuel, E., Summers, I.R., and Claxton, G. (2005) Semantic Divergence and Creative Story Generation: An fMRI Investigation. Cognitive Brain Research, 25, pp.240-50.

Howard-Jones, P.A., Bogacz, R., Demetriou, S., Leonards, U. and Yoo, J. (2009) From Gaming to Learning: A Reward-Based Model of Decision-Making Predicts Declarative Memory Performance in a Learning Game. British Psychological Society Annual Conference, Brighton.

Howard-Jones, P.A. and Demetriou, S. (in press – now published on-line WHEN) Uncertainty and Engagement with Learning Games, Instructional Science.

Hughes, C. (1998) Executive Function in Preschoolers: Links with Theory of Mind and Verbal Ability. British Journal of Developmental Psychology, 16, pp.233-53.

Hyatt, K.J. (2007) Brain Gym: Building Stronger Brains or Wishful Thinking? Remedial and Special Education, 28 (2), pp.117-24.

Immordino-Yang, M.H. (2007) A Tale of Two Cases: Lessons for Education from the Study of Two Boys Living with Half Their Brains. Mind, Brain and Education, 1 (2), pp.66-83.

Isaacs, E.B., Edmonds, C.J., Lucas, A. and Gadian, D.G. (2001) Calculation Difficulties in Children of Very Low Birthweight – a Neural Correlate. Brain, 124, pp.1701-07.

Jaeggi, S.M., Buschkuehl, M., Jonides, J. and Perrig, W.J. (2008) Improving Fluid Intelligence with Training on Working Memory. Proceedings of the National Academy of Sciences (USA), 105 (19), pp.6829-33.

Johnson, M.H. (2004) Developmental Cognitive Neuroscience: An Introduction. 2nd edition.

Kaufmann, L. (2008) Dyscalculia: Neuroscience and Education. Educational Research, 50 (2).

Kaufmann, L., Handl, P. and Thony, B. (2003) Evaluation of a Numeracy Intervention Program Focusing on Basic Numerical Knowledge and Conceptual Knowledge:A Pilot Study. Journal of Learning Disabilities, 36 (6), pp.564-73.

Kaufmann, L., Vogel, S.E., Wood, G., Kremser, C., Schocke, M., Zimmerhackl, L.-B. and Koten, J.W. (2008) A Developmental fMRI Study of Nonsymbolic Numerical and Spatial Processing. Cortex, 44, pp.376-85.

Klingberg, T., Fernell, E., Olesen, P.J., Johnson, M., Gustafsson, P., Dahlstrom, K., Gillberg, C.G., Forssberg, H. and Westerberg, H. (2005) Computerized Training of Working Memory in Children with ADHD – a Randomized, Controlled Trial. Journal of the American Academy of Child and Adolescent Psychiatry, 44 (2), pp.177-86.

Knutson, B., Adams, C.M., Fong, G.W. and Hommer, D. (2001) Anticipation of Monetary Reward Selectively Recruits Nucleus Accumbens. Journal of Neuroscience, 21, RC159, pp.1-5.

Koepp, M.J., Gunn, R.N., Lawrence, A.D., Cunningham, V.J., Dagher, A., Jones, T., Brooks, D.J., Bench, C.J. and Grasby, P.M. (1988) Evidence for Striatal Dopamine Release During a Video Game. Nature, 393, pp.266-68.

Kosslyn, S.M. (2005) Mental Images and the Brain. Cognitive Neuropsychology, 22 (3-4), pp.333-47.

Kratzig, G.P. and Arbuthnott, K.D. (2006) Perceptual Learning Style and Learning Proficiency: A Test of the Hypothesis. Journal of Educational Psychology, 98 (1), pp.238-46.

Lerner, R.M. (2005) Promoting Positive Youth Development: Theoretical and Empirical Bases. Washington D.C., National Research Council/Institute of Medicine.

Luna, B. (2004) Algebra and the Adolescent Brain. Trends in Cognitive Sciences, 8, pp.437-439.

Mahncke, H.W., Connor, B.B., Appelman, J., Ahsanuddin, O.N., Hardy, J.L., Wood, R.A., Joyce, N.M., Boniske, T., Atkins, S.M. and Merzenich, M.M. (2006) Memory Enhancement in Healthy Older Adults Using a Brain Plasticity-Based Training Program: A Randomized, Controlled Study. Proceedings of the National Academy of Sciences of the United States of America, 103 (33), pp.12523-28.

Malone, T.W. (1981) Toward a Theory of Intrinsically Motivating Instruction. Cognitive Science, 4, pp.333-39.

McCabe, S.E., Knight, J.R., Teter, C.J. and Wechser, H. (2005) Non-Medical Use of Prescription Stimulants among US College Students: Prevalence and Correlates from a National Survey. Addiction, 100 (1), pp.96-106.

McGivern, R.F., Andersen, J., Byrd, D., Mutter, K.L. and Reilly, J. (2002) Cognitive Efficiency on a Match to Sample Task Decreases at the Onset of Puberty in Children. Brain and Cognition, 50, pp.73-89.

Molfese, D.L. (2000) Predicting Dyslexia at 8 Years of Age Using Neonatal Brain Responses. Brain and Language, 72, pp.238-45.

Neuper, C., Muller-Putz, G.R., Scherer, R. and Pfurtscheller, G. (2006) Motor Imagery and EEG-Based Control of Spelling Devices and Neuroprostheses. Event-Related Dynamics of Brain Oscillations. Amsterdam, Elsevier Science Bv, pp.393-409.

Nieuwenhuis, S., Heslenfeld, D.J., von Geusau, N.J.A., Mars, R.B., Holroyd, C.B. and Yeung, N. (2005) Activity in Human Reward-Sensitive Brain Areas Is Strongly Context Dependent. Neuroimage, 25 (4), pp.1302-09.

Noel, M.-P. (2005) Finger Gnosia: A Predictor of Numerical Abilities in Children? Child Neuropsychology, 11, pp.413-20.

Nunez, J.M., Casey, B.J., Egner, T., Hare, T. and Hirsch, J. (2005) Intentional False Responding Shares Neural Substrates with Response Conflict and Cognitive Control. Neuroimage, 25, pp.267-77.

OECD (2007) Understanding the Brain: Birth of a New Learning Science. Paris, OECD.

OECD (2002) Understanding the Brain:Towards a New Learning Science. Paris, OECD Publications.

OfSTED (2001) Improving Attendance and Behaviour in Secondary Schools. London, OfSTED.

Paus, T. (2008) Mapping Brain Maturation and Development of Social Cognition During Adolescence. Mental Capital and Wellbeing, State-of-Science Reviews. London, Government Office for Science.

Pickering, S.J., and Howard-Jones, P. (2007) Educators’ Views on the Role of Neuroscience in Education: Findings from a Study of Uk and International Perspectives. Mind, Brain and Education, 1 (3), pp.109-13.

Plomin, R. (2008) Genetics and the Future Diagnosis of Learning Disabilities. Mental Capital and Wellbeing, State-of-Science Reviews, London, Government Office for Science.

Qin, Y., Carter, C.S., Silk, E.M., Stenger, V.A., Fissell, K., Goode, A. and Andersen, J.R. (2004) The Change of the Brain Activation Patterns as Children Learn Algebra Equation Solving. Proceedings of the National Academy of Sciences (USA), 101, pp.5686-5691.

Roman, G.C. and Rogers, S.J. (2004) Donepezil: A Clinical Review of Current and Emerging Indications. Expert Opinion on Pharmacotherapy, 5 (1), pp.161-80.

SCMH (2005) The Future of Mental Health: A Vision for 2015. London, The Sainsbury Centre for Mental Health.

Shaywitz, B.A., Shaywitz, S.E., Blachman, B.A., Pugh, K.R., Fullbright, R.K., Skudlarski, P., Mencl, W.E., Constable, R.T., Holahan, J.M., Marchione, K.E., Fletcher, J.M., Lyon, G.R. and Gore, J.C. (2004) Development of Left Occipitotemporal Systems for Skilled Reading in Children after a Phonologically-Based Intervention. Biological Psychiatry, 55 (9), pp.926-33.

Shizgal, P. and Arvanitogiannis, A. (2003) Gambling on Dopamine. Science, 299, pp.1856-58.

Springer, S.P. and Deutsch, G. (1989) Left Brain, Right Brain. New York, Freeman.

Starkey, P. and Cooper, R.G. (1980) Perception of Numbers by Human Infants, Science, 4473, pp.1033-35.

Steven, M.C., Kiehl, K.A., Pearlson, G.D. and Calhoun, V.D. (2007) Functional Neural Networks Underlying Response Inhibition in Adolescents and Adults. Behavioural Brain Research, 181 (1), pp.12-22.

Stevens, C., Fanning, J., Coch, D., Sanders, L. and Neuille, H. (2008) Neural Mechanisms of Selective Auditory Attention Are Enhanced by Computerized Training: Electrophysiological Evidence from Language-Impaired and Typically Developing Children. Brain Research, 1205, pp.55-69.

Stewart, W.J. (2008) Technology Futures. Mental Capital and Wellbeing, State-of-Science Reviews. London, Government Office for Science.

Szucs, D. and Goswami, U. (2007) Educational Neuroscience: Defining a New Discipline for the Study of Mental Representations. Mind, Brain and Education, 1 (3), pp.114-27.

Szucs, D., Soltesz, F., Jarmi, E. and Csepe, V. (2007) The Speed of Magnitude Processing and Executive Functions in Controlled and Automatic Number Comparison in Children: An Electro-Encephalography Study. Behavioral and Brain Functions, 3, pp.20.

Waterhouse, L. (2006) Multiple Intelligences, the Mozart Effect, and Emotional Intelligence: A Critical Review. Educational Psychologist, 41 (4), pp.207-25.

Weisberg, D.S., Keil, F.C., Goodstein, J., Rawson, E. and Gray, J. (2008) The Seductive Lure of Neuroscience Explanations. Journal of Cognitive Neuroscience, 20 (3), pp.470-77.

Willis, S.L., Tennstedt, S.L., Marsiske, M., Ball, K., Elias, J., Mann Koepke, K., Morris, J.N., Rebok, G.W., Unverzagt, F.W., Stoddard, A.M. and Wright, E. (2006) Long-Term Effects of Cognitive Training on Everyday Functional Outcomes in Older Adults. Journal of American Medical Association, 296 (23), pp.2805-14.

Wilson, A.J., Dehaene, S., Pinel, P., Revkin, S.K., Cohen, L. and Cohen, D. (2006) Principles Underlying the Design Of “The Number Race”, an Adaptive Computer Game for Remediation of Dyscalculia. Behavioral and Brain Functions, 2 (19).

Wilson, R.S., Mendes de Leon, C.F., Barnes, L.L., Schneider, J.A., Bienias, J.L., Evans, D.A. and Bennett, D.A. (2002) Participation in Cognitively Stimulating Activities and Risk of Alzheimer Disease. Journal of the American Medical Association, 287 (6), pp.742-48.

Winter, B., Breitenstein, C., Mooren, F.C., Voelker, K., Fobker, M., Lechtermann, A., Krueger, K., Fromme, A., Korsukewitz, C., Floel, A. and Knecht, S. (2007) High Impact Running Improves Learning. Neurobiology of Learning and Memory, 87, pp.597-609.

Commonly used abbreviations :

fMRI – functional Magnetic Resonance Imaging (fMRI)

OfSTED – Office for Standards in Education, Children’s Services and Skills

ERP – Event Related Potential

EEG – Electroencephalogram

ADHD – Attention Deficit Hyperactivity Disorder

WM – Working Memory

EF – Executive Function

SLI – Specific Language Impairment

ACh – Acetylcholine

BCI – Brain Computer Interface

MI – Multiple Intelligences theory

VAK – Visual Auditory Kinaesthetic

SCMH – The Sainsbury Centre for Mental Health

Note that reward is being used here in the psychological sense, i.e. as a process, or set of processes, by which behaviour is reinforced.

In the type of N-back task used in this research, participants are asked to observe a sequence of digits or letters, and asked to recall the item that was N items back

Memory processes will, however, still remain of key importance in education although semantic memory (for knowing “how to”) may increase its significance relative to declarative memory (explicit recall of facts and events)

An extreme view of this redefinition has been provided by Koizumi: “… education should be designed to guide and inspire the construction of the basic architecture for information processing in the brain by preparing and controlling the input stimuli given to the learners.” Koizumi, H. (2004) The Concept of ‘Developing the Brain’: A New Science for Learning and Education, Brain and Development, 26, 434-41. p435.

For example: http://www.ltscotland.org.uk/ictineducation/gamesbasedlearning/sharingpractice/braintraining/introduction.asp

WM is usually considered one component of EF

Programmes through the curriculum can promote mental health … important characteristics of such programmes include those that enable children to correctly identify and regulate one’s feelings…” DfEE. (2001) “Promoting Children’s Mental Health within Early Years and School Settings.” (Nottingham, Department for Education and Employment), p10

Invasive BCIs involve the implantation of electrodes, produce cleaner signals and perform a little better, but clearly these are not likely to become acceptable amongst those who suffer no serious disability and so these devices are not considered here.

This document has been commissioned as part of the UK Department for Children, Schools and Families’ Beyond Current Horizons project, led by Futurelab. The views expressed do not represent the policy of any Government or organisation.

The 20^th century has been called the “century of the gene”. The designation is unusually neat. The idea of the gene – as a kind of elementary particle of inheritance – was defined roughly at the start of the century, and the structure of the genetic material DNA figured out almost exactly half way through.

You could say the second half of the century was then taken up with exploring the action of genes at the smallest level – a molecular biology in the real sense of the word. The molecules involved are an intricately poised set of micro-machines – or to be more up to date, nanomachines – information carriers, and links between them. They have all co-evolved in ways we are still elucidating, but their elementary workings were first puzzled out in simple bacteria. The eukaryotic (that is, nucleated) cells of more complex organisms mainly use the same mechanisms, but with many additional refinements.

In the last couple of decades of the century, it became possible to think about achieving a complete knowledge not just of how genes work in general, but of all the genes of an organism. And, astoundingly, that has now been done – beginning with the tiniest viruses and working up to the total complement of genes and DNA of the organism we are most interested in, the human genome. (Genes and DNA are not quite the same, as only a small proportion of our DNA sequence encodes working genes. The functions of all the rest are still being worked out.)

The century of the genome?

The completion of the human genome sequence was a landmark. Three billion bases (the paired chemical units whose order preserves genetic information in the DNA double helix) are now catalogued in the databases. Among them are stretches of DNA which encode around 25,000 genes. Keep that number in mind.

However, this notable success has also proved disappointing in some respects. There were great expectations that generating the whole sequence would somehow lay out the secrets of life, and at least cure cancer and shed light on many other conditions. These were partly built up by the biologists’ campaign rhetoric which was needed to get the genome project funded. They were also partly due to a slightly exaggerated idea of genes’ importance which grew up during the post-DNA decades when they were such a focus for brilliantly successful research. And there was a tendency to argue that, since the last fifty years of biology have been so astonishingly productive, the next fifty will be equally spectacular. Maybe so, but the new problem of making sense of all the information in the genome, and how it is used, has occasioned a new realism, and given rise to new programmes of work.

Simply recording the strings of text which make up the “book of life” does not reveal all the keys to how it is read. To change metaphors, the gene catalogue is a parts list, not an assembly guide. But even that is too static. Genomes, and even to some extent genes, are dynamic entities. The text is extensively annotated, with chemical tags joined directly to the DNA and with more ephemeral signal molecules which embrace and release particular genetic sequences and affect their activity. Most, perhaps all, human genes which code for proteins, it turns out, produce more than one product, according to circumstance. Many genes, and many portions of what used to be dismissively termed “junk” DNA, regulate expression of other genes. Considering the genome, in short, reveals new layers of complexity.

The outcome is a recognition that we are entering a new era of “post-genomic” biology, but what that means is more clearly defined in terms of what it is not than what it will actually be like. It will depart from the largely reductionist, and in their time very fruitful, assumptions which underpinned the early days of molecular genetics and the planning of the genome projects. Genes seen as unchanging, crystalline information stores, which can be unplugged and swapped around like Lego bricks, are a less dominant image now. Defined DNA sequences are still a source of fundamental information – and the key to inheritance. But the use of that information is seen in context, and feeds into complex, shifting circuits and networks. Our ability to extract that information is now impressive. Making sense of it is harder.

Genome to cell

There seems general agreement that post-genomic biology will be systems biology. Again, it is unclear just what the term will mean. It describes an aim: to understand an organism as a total system, or a system of systems. The smallest system is thus a single cell. Sometimes, in bacteria or yeast, the cell is an organism in its own right. New ways are emerging which take the measure of the entire cellular system. New coinages go with them. After the genome comes the transcriptome, the inventory of all the bits of DNA which are being read at any moment and used to make the message carrying nucleic acid RNA. Then there is the proteome, the catalogue of all the different proteins being made, and – more comprehensively still – the metabolome, the complete list of all the chemicals the cell is using. The most comprehensive of all would be the physiome, which includes all the above.

The next level is to sketch the circuit diagrams which link all these. Again, the circuits, or networks, may be genetic – sets of genes, which influence each other – or metabolic – chemical pathways linked by stepwise synthesis or degradation, controlled by enzyme proteins which are themselves subject to regulation. In between are cellular signalling chemicals, usually small molecules, made under genetic control but also responding to external stimuli. One approach to systems biology is to build computer models which bring together everything which is known about these kinds of connections, and which reproduce key features and responses of actual cells. From this point of view, yeast is the best-modelled organism so far.

Along with this effort there is also much work on so-called synthetic biology. Like systems biology, this is as much an aspiration as a reality. In one sense it is an extension of existing genetic engineering, now a well-developed set of techniques in most organisms. Although genes are not just simple plug-in modules, they can often be treated as such in simple cases – and with skilled use of suitable enzymes they can be unplugged and, with suitable modifications, plugged into new locations. They may even work in completely different contexts. Thus, in a recent feat which has been used as an emblem of one kind of synthetic biology, the complete set of genes needed to make a precursor of the plant-derived substance artemesenin, an anti-malarial drug, has been stitched into a strain of yeast, which duly manufactures the alien chemical.

As well as moving around existing genes, other synthetic biology efforts involve synthesising new ones, designing parts of possible biological systems which would work to produce a desired result, and trying to define the “minimal” genome – perhaps for an organism which could then support an additional suite of artificial genes. At the moment, this is fairly low-level, if fascinating, stuff, and a long way from building organisms to order. It does presage enormous possibilities – not least a culture of open-source biotech and biohacking. In the medium-term, while one may expect to see some genetically-engineered pets making headlines, the main action will be in bacteria. The industrial and technical potential, for pharmaceuticals, fuels, and, alarming to some, weapons, is impressive. It is unlikely, though, that any of this will be directly applicable to humans in the next few decades, so it is not discussed further here.

Cell to organism

The workhorse of systems biology and, on occasion, synthetic biology – yeast – is at least one of the more complex of the two main types of cell. But complex organisms, of course, are enormous assemblies of cells, which can give rise to many more interactions. A grown human has at least ten trillion cells, of 200 different types. In various combinations they build all the different tissues and organs. More ambitious systems modellers are trying to tackle individual organs – the heart, say, or the liver. This involves emergent properties of collections of cells, like regulation of heartbeat, which is achieved through a subtle combination of electrical and chemical controls. Systems biology started out in the 1960s with mathematical modelling of cardiac cells. Work to integrate understanding of cardiac muscle rhythm, from genetic influences to the electrophysiology of the whole organ, is now far advanced, but not yet complete. And modelling an entire organism, whose tissues and organs influence one another – via hormone action, for example – is another level of difficulty altogether.

And finally … the brain

Modelling an entire organism would include the nervous system. But some nervous systems are more elaborate than others. The most complex, and most versatile is the one which centres on a human brain. That versatility arises, somehow, from the way the brain’s 100 billion neurons are selected during development – during which many neurons die – and, even more, from how the connections between them are selected. No-one knows how many neural connections there are in a human brain, but each neuron has between 1,000 and 10,000 links (synapses) to other neurons. Do the maths, and that means an adult human brain may have 1,000,000,000,000,000 intercellular connections. Among other things, this means that ideas about genes determining much about the details of individual brains are implausible. A few tens of thousands of genes can shape the overall structure of neurons, their patterns of development, and the ways they find their way to the right location in the growing brain and decide which other neurons to build connections with. But the scope for hard wiring is a lot smaller than would be needed for much in the way of direct links between genes and complex behaviours.

How such behaviour actually emerges is unlikely to be elucidated in detail any time soon. Computer models can now simulate networks of a few thousand neurons. But study of real neural circuits in the lab still most often focuses on little bits of the invertebrate neural system which have a few dozen neurons. Understanding how the whole brain develops in any kind of detail is thus a long way off. Although neuroscience has made many intriguing advances in recent decades, and many startling new techniques are adding to the store of useful observations about how brains work, we understand rather little about most higher brain functions.

Biology and education

Education, the process by which a human being becomes a fully-functioning member of their culture, is obviously underpinned by our biology. Will better knowledge of biology shed light on the process? That would presumably mean understanding learning, a feature of brains. So whatever other factors in our biology may act on the path to learning – whether genetic, hormonal, metabolic, even dietary – they have their effect in the brain. Whatever the ultimate cause, if there is a single cause, the effects of any of these factors will involve all the levels from the molecular to the fully integrated nervous system.

This suggests caution about linking biology to education. But it need not mean that, because neither the workings of the brain nor all the links between, say, genes and brains are understood, that near future research will not be relevant to education. It does mean that it is more likely to yield hints, and possibilities for indirect intervention, than well worked out paths to improvement. It is not difficult to affect the brain. The tricky thing is to affect it in the way, and only the way you want. We already have a collection of somewhat blunt instruments. Some, like ECT, are not too distant from trying to bring a radio back on station by thumping it. Others, like a double espresso, are perhaps more akin to trying to improve the performance of your hard drive by dousing it in WD40. And of course there are a range of surgical techniques for excision or ablation of troublesome bits of the most complex organ. With that starting point, how hard can it be to do better?

Future prospects

Diagnosis

Having moved up to brains, let us retrace our steps and go back to genes. The human genome project was about technology as well as science – a distinction which is becoming harder to draw in any case. Sequence one genome, sequence them all. And not just all species, but all individuals of a species, as often as you want.

That is already possible. The easy prediction is that it will become practical as well. As with Moore’s “law” which charts the exponential increase in the number of transistors in a single integrated circuit over the last five decades, DNA sequencing has been getting steadily faster and cheaper for decades. In that light, the genome project was one portion of a fairly steady trajectory. The same is true for the complementary operation, artificial DNA synthesis.

Figure 1: Cost per Base of DNA Sequencing and Synthesis

The sequencing trend is a powerful indicator that one aspect of near future biology will be diagnostic. The thousand dollar genome is now on the horizon. Another tenfold improvement would bring a complete genome sequence within the realm of routinely affordable medical tests. There are already microarray techniques on offer – using many RNA fragments of known sequence to detect the presence of particular stretches of DNA – which can test for hundreds of thousands of markers. These are usually single nucleotide polymorphisms (SNPs, or “snips”) – locations where the genome differs in different individuals by substitution of a nucleotide base for one of the other three. They are identified and mapped by a process which is indifferent to whether the base in question is part of a functional gene, but they act as markers for a gene or genes which are nearby and are inherited together with the base location in question. Existing genome surveys are typically a mixture of checking for a large number of SNPs and a smaller number of well-characterised genetic variants. But these are technicalities. The point is that they are a definite step on the path to “personal genomics”.

Getting useful information out of a personal genome will also depend on providing interpretative software which is usable by the owner of the genome. But the general idea is already being exploited as a sales pitch. The old image of genes as powerful determinants of outcomes is a good fit with the marketing hype of people who deal in commercial clairvoyance. The (justifiably) high-tech sheen of genetic testing is initially convincing, and goes along with a rhetoric of empowerment through knowledge. The current offer is typically to test your DNA sample (a cheek scraping or just a little saliva suffices) for a suite of gene variants held to affect probability of a range of medical conditions such as heart disease or Alzheimer’s. In the first case, knowing you have a high risk of heart disease might help you follow all the good advice about diet and exercise, or begin taking a daily dose of cholesterol-lowering drugs, before the damage is done. In the second, well, at least you or your family would have an incentive to deal with the bureaucracy required to instate an enduring power of attorney before cognitive deterioriation sets in.

As this suggests, one consequence of personal genomics might be to increase the number of what are termed the “worried well”. Some of them might be children. Some might be parents whose children were identified as at risk of a particular condition. Suppose there actually was a gene variant which gave a clear indication of a high risk of schizophrenia, for instance. Knowing a child had it would create stresses which are just the kind a person at such risk ought to avoid.

Personal genomics will throw up concerns like this – if not necessarily that particular one. It may also have other effects on young people’s sense of identity. This will relate to their history and cultural affiliation as well as the possibility of particular abilities or disabilities. One child may rule themselves out of athletics because they do not have the best combination of genes for fast action muscle fibres. Another will be forbidden from doing athletics at all because they are at risk of sudden death from cardiomyopathy (children in Italy are already tested for this – non-genetically – before starting school sports).

But both children could also be reading their personal genome in ways which reinforce, or loosen tribal or other loyalties. The genome can be interpreted as a record of genealogy and ancestry as well as a harbinger of medical futures. And if the habit of comparing genomes down generations catches on, some will have questions to ponder about their actual paternity. Maternity, in general, is less open to doubt, but adoption and some kinds of assisted reproduction obscure maternal origins in ways which genome analysis can overcome. This is already possible through established techniques of DNA fingerprinting. The forecast is simply that this kind of information will be more generally in use.

The overall effects of this ready availability of total genetic information could go one of two ways. And it is difficult to judge which is more likely. One possibility is that, having heard for decades that genes are powerful, the full genome readout will be seen as data of great significance. That itself might induce activism (I’m going to beat the odds here) or fatalism (there’s nothing you can do so why try?).

On the other hand, it may become more apparent that genes, and certainly individual genes, only produce their effects in concert with many other factors. In a future which is likely to see much more information, from mobile phone records to medical tests and financial history, archived and cross-referenced in databases – a future, perhaps of ID cards, surveillance and biometrics – genomic information could come to be seen as just some more information, rather than crucial information.

Will there be more specific education implications when readouts of large subsets of genes of medical or neurological interest are widely and cheaply available? Similar advances can be anticipated in subtler analyses like which genes are actually active, perhaps in particular brain regions. Genetic analysis will advance in tandem with refinement of other techniques for registering brain structure and activity – such as ECG and functional magnetic resonance imaging (fMRI). Further ahead, this information will be fed into computer-simulations of brain regions, exemplified by the “Blue Brain” project in which an IBM supercomputer is running a simulation of a biologically realistic cortical column. So far, this models an assembly of 10,000 rat neurons, with their many millions of individual connections. Finding ways of incorporating molecular and genetic data to influence how the model neurons are wired up is one of the future ambitions of the designers. If that can be done, then it might be possible to use such a model to understand the effects of changes in genetic or chemical make-up of the simulated neurons, but this is a long-term prospect.

Where might diagnostic work lead in the meantime? Around half of human genes are active in the brain. Variant forms of those genes will be associated with many different effects – and sometimes different genes will go with what looks like the same effect. Conditions like autism, or dyslexia, which are neurodevelopmental in origin, will be shown to arise in complex ways. Some genetic variants will make them more likely. Unravelling all of them is also likely to show that these, like many other conditions, are labels, which cover a more complex set of possibilities.

How much difference this will make is harder to say. Early and accurate diagnosis could assist timely remediation – one to one help with reading, say. A better understanding of brain development might allow some aspects of early education to be tailored more closely to the timing of crucial stages of neural wiring. This will be aided by the increasing richness of comparative genomics. Studying which genes have evolved most rapidly since humans and chimps began to diverge from their last common ancestor, for example, may yield further clues about changes in the brain which were crucial for the emergence of language and culture.

However, at the moment it appears unlikely that genetic analysis alone will lead to findings which have a big impact on other, more general aspects of educational practice. Studies aimed at identifying genes which influence traits which are seated in the brain – from general intelligence to anxiety or depression – have one fairly consistent result. (Some hold that studies of most traits have the same feature.) They find, not a few genes of large effect, but many genes with a small influence. While these undoubtedly combine to produce a substantial genetic shaping of such traits, and can provide important clues about mechanisms and avenues to explore for treatment, they will make individual prognosis hard to read. Again, any individual gene effect is heavily modulated by other influences, genetic and non-genetic. This is also the reason why the discussion of “designer babies” produced by selection during IVF and similar procedures is misleading. Gross defects can be avoided, but selecting for other traits is likely to prove elusive.

The small number of apparent exceptions to these complexities unearthed so far turn out to be less specific in their effect than first thought. A well-known example is the FOXP2 gene, which appears to exist in altered forms which disrupt aspects of spoken language ability. However, more detailed studies – including comparisons with other species – suggest that the gene product helps regulate a suite of other genes involved with fine muscular co-ordination.

The general tendency to find a large number of genes with small, additive (or occasionally synergistic) effects also reduces the chances of radical simplification of approaches to general issues relating to health and well-being which are likely to be of concern to the education system. Two important examples are obesity and depression. The latter is widely believed to be increasing among the young, a trend sometimes seen as part of the “affluenza” thesis. The evidence for an increase in adolescent depression in recent decades turns out to be poor- and the historical incidence in previous eras is, of course, much harder to gauge. However, depression – and even more so, suicide, is a real issue for those caring for the young. At present, it appears that mixed approaches, taking into account what young people themselves have to say about mental health, may be the way forward, rather than application of any new biological insights.

The same is probably true of obesity, which certainly does appear to be on the rise among children and young people in affluent countries, and will compromise their later health prospects. As the recent Foresight report recognised, this is a problem which will only be affected, if at all, by policies which draw on many actors, in government and elsewhere, rather than succumbing to a magic bullet.

Enhancement

The advance of diagnostics via genomic readouts and functional imaging will likely be gradual, and if so could be assimilated without too much trouble by existing systems for assessment – such as those for identifying special needs. But what of the prospect for more dramatic effects on education more generally? There is much speculation about souped-up technological futures in which education is transformed. Much of it rests on the predicted convergence between biology, nanotechnology, information technology and cognitive science (bio-nano-info-cogno). Some of these visions can be ruled out for the foreseeable future. The notion that a defined body of knowledge – a new language, say – could be acquired by plugging a chip into some kind of neural interface founders on the fact that we have essentially no idea what the internally stored neural coding might be, beyond saying that memory somehow depends on patterns of firing of cells connected in the right way.

Assume then that learning will occur, if it occurs, in brains configured as they are now. As with diagnosis, complete understanding of how they work may be far off, but is not a prerequisite for thinking about how their general performance might be improved. We do know a lot about neurons, synapses, and neurotransmitters. We will know more. How might this knowledge be applied?

Start with synapses. These are a pretty essential part of any theory of learning and memory. They are inter-cellular junctions which modulate signals electrochemically. At the moment, we are building up knowledge of the individual parts. The molecular biology of synaptic transmission is tolerably well understood, although, as a recent review puts it “study of the mechanisms underlying plasticity [of synaptic connection] and hence memory and learning, is proving thornier than expected.”

However, there could be routes to enhanced performance which do not need a fully worked out systems biology of synapses. Genomics will lead to a full inventory of neurotransmitters molecules, their multiple receptors and the enzymes which control their transport and turnover. This offers great scope for refining blunt instruments like the current generation of psychoactive drugs. The selective serotonin reuptake inhibitors – SSRIs – widely prescribed for depression, for example, are an advance on earlier medication because they only influence one transmitter. But they still increase levels of a neurotransmitter substance for which there are at least a dozen different receptors. Tuning drug design to select particular receptors or cell sub-types would enable more specific effects.

There is a range of other molecules which are being characterised and might have desirable effects in the brain. They include neurotrophic factors, which aid growth and survival of neurons. Further ahead, gene expression can, in principle, be affected by “antisense” RNA molecules which bind to particular gene sequences and block their use. However there is a major block to their application in the brain as they are large molecules and cannot pass the so-called “blood brain barrier”. Indeed, many small molecules are similarly excluded from the brain. Direct administration of new drugs to the brain, and ideally to particular brain regions presents obvious difficulties and achieving it by reasonably non-invasive means will depend on a good deal more research intended, in the first place, for therapeutic application.

Assuming that hurdle can be jumped, there are some forecasters who envisage a future which offers an array of brain-active drugs which, in the right combinations, would enhance mood, attention, sensory acuity, and memory, if not yet actual understanding. Used in the right combinations, such drugs could enhance learning by making more efficient use of study time and aiding retention. They might be suitable for use by future adults, whose portfolio careers require continual development of new skills. Another more general enhancement is relevant here. Increasing lifespan appears a fairly likely possibility arising from the improvement of genetic knowledge and cell biology. It is, in many ways, a logical consequence of the medical conditions which will be a focus of research in the West.This need not be radical life-extension. The limit of the existing human phenotype is around 120, so simply enabling some appreciable proportion of people to reach this age without falling into decrepitude would be a big change. Some people enjoying this extra life-expectancy would no doubt want continuing education to help make use of their time. This takes us beyond our timespan, though as such a change, by definition, would take a long time to work through the population.

Meantime, demand for the relatively crude cognitive enhancers which already exist suggests that such drugs will be widely sought if they become available. There are a few reservations, though. Some of the existing drugs work, to an extent – enhancing memory, for example. But more powerful effects to come could easily be a mixed blessing. Forgetting, which can also be potentially enhanced by drug use, is as important as remembering. Effective educational use would require specific, not general memory enhancement, and would need to be short term (the more effective retention, not the actual memories). Similarly, general improvement of synaptic transmission or nerve impulses might speed up brain processing. But one serious theory of the origins of autism suggests that it arises from enhanced brain functions, including sensory acuity, which leads to the person blocking out some external signals as a way of coping with unbearably strong stimulation. Attempts at enhancement of a complex system which is poorly understood can easily fall foul of the annoying fact that you can have too much of a good thing.

There are also likely to be trade-offs between enhancements. A well-known example is a strain of mice engineered to be more receptive to a particular neurotransmitter which, among many other roles, is involved in formation of memory. Giving adult mice more of a variant of the receptor normally prevalent in babies did appear to improve their capacity for mouse-level learning. However it also made them more sensitive to pain.

As in this example, many of the changes envisaged as achievable through drugs are potentially reachable through genetic modification. As emphasised above, ways of doing this are already known, in principle. But getting reliable results, indeed any results at all, in human patients has proved extremely difficult. The chance of any non-medical gene alteration which affects brain function being proven, and agreed to be ethically acceptable for trial in humans, by 2025, appears vanishingly small.

References

Academy of Medical Sciences and Royal Academy of Engineering (2007) Systems Biology: A Vision for Engineering and Medicine. Available from http://www.raeng.org.uk/policy/engagement/pdf/Systems_Biology_Report.pdf

Amaral, P. et al (2008) The Eukaryotic Genome as an RNA Machine. Science, 319, p.1787.

Costello, J. et al (2006) Is there an epidemic of child or adolescent depression? Journal of Child Psychology and Psychiatry, 47, pp.1263 – 1271.

Endy, D. (2005) Foundations for Engineering Biology. Nature Reviews, 438, pp.449-453.

Feero, W.G. et al (2008) The Genome Gets Personal-Almost. Journal of the American Medical Association, March 19.

Flint, J. and Shifman, S. (2008) Animal models of psychiatric disease. Current Opinion in Genetics and Development, July

Foresight (2007) Tackling Obesities: Future Choices. Final project report, DIUS. www.foresight.gov.uk

Hunter, P. et al (2002) The IUPS human physiome project. European Journal of Physiology, 445, pp.1-9.

Keasling, J. et al (2006) Production of the antimalarial drug precursor artemisinic acid in engineered yeast. Nature, 440, pp.940-943

Maher, B. (2008) The case of the missing heritability. Nature, 456, pp.18-21

Miller, G. (2006) A better view of brain disorders, Science, 313, pp.1376-1379

Nurse, P. (2008) Life, logic and information. Nature, 454, pp.424-6

Paul Nurse

Paykel, E.S. (2000) Not an age of depression after all? Incidence rates may be stable over time. Psychological Medicine, 34, pp.489-490

Pearson, H. (2006) What is a gene? Nature, 441, pp.398-401

Pennisi, E. (2007) Breakthrough of the Year. Human Genetic Variation. Science, 318, pp.1842-3.

Stix, G. (2008) Jacking into the Brain–Is the Brain the Ultimate Computer Interface? Scientific American, November.

Thorleifsson, G. (2008) Genome-wide association yields new sequence variants at seven loci that associate with measures of obesity. Nature Genetics, Published online 14 December 2008 | doi:10.1038/ng.27

Vijg, J. and Campisi, J. (2008) Puzzles, promises and a cure for ageing. Nature, 454, pp.1065-71

Westerhoff, H.V. and Palsson, B. (2004) The evolution of molecular biology into systems biology. Nature Biotechnology, 22, pp.1249-52.

Woese, C. (2004) A New Biology for a New Century. Microbiology and Molecular Biology Reviews. June, pp.173-186.

Zimmer, C. (2008) Searching for Intelligence in Our Genes. Scientific American, October

For recent (but, inevitably, already out of date) commentaries, see Pearson, 2004; Amaral, 2008.

For one view of the way forward, from a leading cell biologist and research strategist, see Nurse, 2008.

See Westerhoff, 2004; Endy, 2005, and, more comprehensively, Academy of Medical Sciences, 2007.

See Hunter et al, 2002, for the original outline of the physiome project.

As described in Keasling et al, 2006.

See Feero, 2008.

See Miller, 2006.

Details can be found at http://bluebrain.epfl.ch/

For the story for psychiatric disease see Flint, 2008. For obesity, see Thorlieffson et al, 2008. More generally, see Maher, 2008.

See Paykel, 2000 and Costello et al, 2006.

For one such recent initiative in the UK, see http://www.right-here.org.uk

Foresight, 2007.

As argued in Stix, 2008.

On ageing, and prospects for a “cure”, see Vijg, 2008.

This document has been commissioned as part of the UK Department for Children, Schools and Families’ Beyond Current Horizons project, led by Futurelab. The views expressed do not represent the policy of any Government or organisation.

Educators in schools and colleges bear great responsibility and are subject to an enormous weight of expectation, yet must operate in environments where both influence and resources are limited. Consequently, education seems particularly vulnerable to fads and trends (Slavin, 1999), to the promise that ‘the next big thing’ might resolve its many difficulties. Neuroscience is intrinsically fascinating and obviously relevant to education because its subject matter – the brain – is vital to our ability to learn. At the same time, it would be naive to imagine that neuroscience alone can solve many of the problems encountered by educators.

The 1990’s ‘decade of the brain’ saw massive investment in neuroscience across the English-speaking world. Colourful images generated by brain scanners were taken up eagerly by the media, and studies linking brain areas and systems to capabilities or problems are now rarely out of the news. In a cultural moment where neuroscience, genomics, and pharmacology are jointly enrolled within ‘the politics of life itself’ (N. Rose, 2001) these developments have justifiably caught the public imagination. They have contributed to a climate where the expectations associated with neuroscience are sometimes unrealistically high, where neuroscience has perhaps been positioned as education’s most-recent ‘next big thing’.

This report will review current trends in neuroscience as they relate to education. Relevant research methods will be briefly described, some mis-uses of neuroscience will be discussed, and a sketch of current progress in neuroscience relevant to education will be presented. Various limitations upon the application of neuroscience to education will then be described, and some possible future implications of neuroscience for education will be suggested. First, a very brief guide to the structure and function of the brain (drawn primarily from Beaumont, Kenealy, and Rogers, 1996; Damasio, 1999; S. Rose, 1997) will set out something of what is taken to be known and largely uncontroversial, whilst also providing a common language for the discussions that follow.

2 The brain

2.1 Neurons and synapses

Neurons are the active nerve cells of the brain. Each neuron has a cell nucleus, and a large number of branching structures known as dendrites, which are available for potential connections with other neurons. Each neuron also has a long extension called an axon, down which electrical messages flow from the cell nucleus. The end of the axon branches out into connectors called pre-synaptic terminals.

The junction between neurons is called a synapse. At a synapse, the pre-synaptic terminals of one neuron are in close contact with the dendrites of another. Although communication within a neuron is electrical, communication across a synapse is chemical. When a neuron ‘fires’ it discharges an electrical signal down its axon, which releases a neurotransmitter into the synapse; this chemical then gets absorbed by the next neuron. Its absorption changes that neuron’s chemical properties, influencing the probability that it will also fire.

2.2 Systems

The average adult human brain is thought to contain around 100 billion neurons, most of which are formed before birth. Whilst the number of connections between these neurons varies hugely, they are frequently connected to many thousands of others. The staggering number of connections thus created allows the brain to function as an extraordinarily complex ’system of systems’. These systems are distributed throughout the brain, contain multiple feedback loops, and are organised into neural networks.

Consequently, all but the very simplest activities involve activity in more than one area. The great majority of functions are carried out by multiple, more specific, elementary processes. These processes are typically carried out in parallel, rather than sequentially. This means that there are frequently degrees of redundancy: not all of the processes the brain conducts may be strictly necessary for a function to be successfully completed.

There is good evidence that some abilities are closely associated with neural activity localised to particular areas, which are presumed to be key nodes in the systems of which they are a part. But the brain also exhibits individual variation and marked plasticity. Lesions to a specified area do not always have the effect that would be predicted, and brain regions usually specialised for one kind of function can get recruited for others: for example, in the brains of visually-impaired people, areas that usually process vision can be used for reading Braille (Roder and Neville, 2003).

2.3 Gross structure and functions

The outer surface of the adult human brain is called the cortex. It has two hemispheres, left and right, joined by a thick bundle of connecting tissues, the corpus callosum. The cortex is subdivided into a series of lobes. Although virtually every activity recruits neurons in multiple areas, each lobe is primarily associated with broadly different kinds of function:

Frontal lobes: planning, reasoning, inhibition of behaviour

Temporal lobes: hearing, memory, object recognition

Parietal lobes: integration of information, sensation, spatial processing

Occipital lobes: visual processing

Because the size of the brain is constrained by the skull some areas of cortex are pushed back upon themselves in folds or sulci, giving the cortex a wrinkled appearance. Within the cortex lie a series of deeper, evolutionarily-older structures, including the hippocampus and the amygdala. Below the cortex lies the cerebellum or ‘little brain’, which is implicated in movement, balance and habit. Beneath this, running down into the spinal cord, the brainstem regulates sleep, wakefulness, and the basic homeodynamic functions necessary for life.

2.4 Growth and Development

Although the brain already has most of its neurons at birth, many of the synaptic connections between them have yet to be formed. Brain volume quadruples between birth and maturity, and most of this increase is due to the formation of new synapses. Three processes are involved.

Synaptogenesis is the general term for the growth of axons and dendrites and the formation of new synapses. There are genetically pre-programmed periods of intense synaptogenesis in different areas of the brain at different times. Synaptogenesis frequently results in more synaptic connections being formed than will ever be needed. Periods of intense synaptogenesis are therefore followed by periods of pruning, when unused connections are eliminated. Once neurons have become organised into relatively stable networks, they frequently get coated in a layer of fatty tissue called myelin. This process, called myelinisation, increases the speed of electrical conduction within the neuron.

The brain remains able to form new synapses throughout life. Learning skills, or responding to an injury or stroke, often produces measurable synaptogenesis in relevant brain areas. Nevertheless, large parts of the brain’s overall structure are in place at birth or shortly afterwards, and subsequent myelinisation may further restrict the ability of neurons to form new connections. Consequently, there are also limits upon the brain’s ability to re-organise itself.

3 Research in neuroscience

Knowledge of the brain and its functions has always been bound up with the technologies used to both investigate and imagine it. In the past, dissection and staining techniques yielded relatively static images of the brain, and its operation was conceptualised using mechanical, telegraphic or hydraulic metaphors (Daugman, 2001). By contrast, contemporary knowledge is frequently bound up with computational metaphors and generated using new imaging methods. However, other research methods remain relevant and continue to contribute important elements of contemporary understanding.

3.1 Lesion studies

Studies of human brain lesions – due to trauma, or following surgery – have long provided an important source of knowledge. Damasio’s influential ’somatic marker’ hypothesis was prompted by such studies, whilst the unusual case of ‘H.M.’ continues to inform contemporary memory studies. But human brain lesions are variable and cannot be replicated, so other methods are often needed to interpret their effects.

3.2 Animal experiments

In these experiments, neuroscientists can systematically manipulate brain functions, for example by excising specified areas of tissue. They can also impose extreme, systematic environmental variations, such as complete loss of sight in one eye, or presentation of only one kind of stimulus to a visual field. Using such methods, neuroscientists have been able to establish some very reliable findings.

3.3 Imaging

Many of the most widely taken-up findings in contemporary neuroscience rely upon brain imaging techniques that allow the activity of different brain areas to be reliably measured in real time. They create dynamic representations of neural activation which show how brain activity is associated with the performance of different tasks.

Magnetic resonance imaging is sometimes said to provide the ‘gold standard’ of brain imaging. Structural magnetic resonance imaging (MRI) scanners can chart the relative size of brain structures, whilst functional magnetic resonance imaging (fMRI) scanners identify differential patterns of brain activity.

Electroencephalography (EEG) uses a net of electrodes attached to the head to measure electrical activity across the scalp; this corresponds to neural activity in the brain. EEG does not produce a pictorial image, but generates graphs with peaks and troughs that represent variation in levels of neural activity. These are described in terms of both their direction and the time-lag between a stimulus and their occurrence. For example, N400 would describe a trough, or negative peak, occurring 400milliseconds after a stimulus.

Functional near-infra-red spectroscopy (fNIRS) is only just beginning to be widely used. It measures the diffusion of near infra-red light projected through the skull; the intensity of the light being diffused is modulated by the level of neural activity, because the light-scattering properties of neurons change when they are active. fNIRS is portable and does not need the person to keep entirely still; some systems can even be used wirelessly, allowing almost free movement.

For various reasons, other imaging technologies are likely to be less useful in relation to education. For example, positron emission tomography (PET) uses radioactive tracers, and so is not suitable for use with children. Magnetoencephalography (MEG) scanners are most effective at detecting activity in bundles of neurons lying in the sulci parallel to the scalp, and seem to be most useful for studying simple sensory and motor processes. They are poor at detecting signals from deeper inside the brain, and less useful for studying complex cognitive activities.

Using these methods, neuroscientists have made striking advances in our knowledge of how the brain works. However, with respect to education their work has also sometimes been misinterpreted, so this survey of their findings will begin by addressing these misunderstandings.

4 Neuromyths

The climate of unrealistic expectations surrounding neuroscience created fertile ground for the establishment and circulation of ‘neuromyths’. This term was coined in a 2002 OECD report, and subsequently taken up by neuroscientists concerned about the misuse of their discipline. Neuromyths are culturally-prevalent misunderstandings of neuroscience that have been used to justify educational interventions. Many of these interventions have been developed and marketed by commercial interests. There is little evidence for their efficacy, and their basis in neuroscience is frequently tenuous. Neuromyths include:

• the need for ‘balance’ between functions localised to left and right cortical hemispheres
• the idea that specific physical exercises can have effects upon particular brain functions
• that there are ‘brain buttons’, areas of the body where applying physical pressure stimulates brain activity
• that there are global preferences of learning style (typically characterised as visual, auditory or kinetic) rooted in neural differences
• that there are known critical periods for acquiring certain abilities
• that increased synaptogenesis can be fostered by placing children in more complex or ‘enriched’ environments
• that there are ‘male’ and ‘female’ brains
• that educational programmes can use implicit learning to teach advanced cognitive skills without effort or attention

There are serious flaws with each of these claims (Goswami, 2004a). For example, although the brain shows clear evidence of hemispheric specialisation, outside of special education ‘balance’ need not be a concern. This is because, for the overwhelming majority of people, the corpus callosum provides an abundant flow of bi-directional communication. Similarly, the evidence for critical periods in the acquisition of some abilities mostly comes from animal brains, is mostly confined to sensory and motor areas, and is challenged by evidence of subsequent, ongoing plasticity. For example, even after an experimental manipulation completely deprived one of a kitten’s eyes of sight for a year, subsequent use of this eye resulted in significant synaptogenesis (Chow and Stewart, 1972). Likewise, there is no evidence that early environments ‘enriched’ in specific ways generate noticeable benefits for human brain growth and development. Again, the evidence comes from animals, and actually shows only that impoverished environments (laboratory cages, rather than larger spaces corresponding more closely to natural environments) restrict development (Bruer, 1997).

Although they may take some time yet to fade from the popular imagination, these neuromyths have been decisively dismissed by neuroscientists. In their wake, it is now perhaps easier to realistically assess the contribution that neuroscientists might make to education. Their work is usually divided into the sub-disciplines of cognitive, social and affective neuroscience (although cognitive science – cognitive neuroscience supplemented by computer modelling – is also sometimes distinguished). To date, it is cognitive neuroscience that has attracted most interest in education.

5 Research in cognitive neuroscience

Most of the educationally-relevant research in cognitive neuroscience has been into the developmental sequence of the human brain, and into the acquisition of basic skills such as reading and number. However, there has been some interest in other skills, and in applying cognitive neuroscience to special education.

5.1 Brain development and ’sensitive periods’

As we have seen, the idea of critical periods in human brain development is a neuromyth. Nevertheless, studies of developing humans do show that there are numerous periods of synaptogenesis and pruning in the brain: one in early childhood, another in the early teenage years; and further marked development, particularly in the frontal lobes, in early adulthood (Gogtay et al, 2004).

Because the timing of these processes is relatively invariant, there are identifiable developmental sequences for basic abilities such as visual processing. These sequences mean that more complex, late-acquired abilities – for example, depth perception – may be more sensitive to disruption or environmental deprivation (Goswami, 2004a). However, the evidence suggests that even for these abilities some recovery is possible.

It has therefore been suggested that humans may not have critical periods so much as sensitive periods: times when the developing human brain is ‘primed’ for certain kinds of experience and especially able to respond to it effectively. However, the practical educational significance of any such sensitive periods remains unknown.

5.2 Reading

An important strand of research in cognitive neuroscience is focused upon reading and reading problems. Imaging studies have shown that, in adults, whilst reading is mostly localised to left-hemisphere systems, there is some variation in patterns of activation as a consequence of different languages. They have also shown that in English-speaking children without reading difficulties there seems to be a developmental sequence of brain activity as reading skills are acquired. In the early stages, reading activates areas of both left and right temporal lobes, but right hemisphere activity gradually declines and significant amounts of activity occur across three main left-hemisphere sites. One of these sites, known as the ‘Visual Word Form Area’ (VWFA) sits at the junction of the temporal and occipital lobes. In mature readers, it seems to be involved both in visual recognition of words and also in phonology (Paulesu et al, 2001).

However, in children identified as being developmentally dyslexic (failing to learn to read, despite being of average intelligence) this sequence appears not to be followed. In these children significant activation of the right hemisphere continues, and there is also less activation in the VWFA and other left hemisphere sites associated with skilled reading (Shaywitz et al, 2002). These studies have also suggested that after targeted remediation (for example, teaching in phonological skills) this unusual pattern of activity can become more normal (Temple et al, 2003).

5.3 Number

Imaging studies have identified various brain areas, in both hemispheres, that seem to be associated with the successful recognition and manipulation of numbers. One area, the horizontal segment of the bilateral intraparietal sulcus (HIPS) seems to be especially critical for judging quantity, size and differences between numbers (Dehaene, Molko, Cohen and Wilson, 2004). Studies have also shown that there are links between areas of the brain important for visuo-spatial processing and those important for recognising the magnitude of numbers (Hubbard, Piazza, Pinel and Dehaene, 2005).

Some research suggesting the validity of this link has been carried out with girls diagnosed with Turner Syndrome, who typically have both visuo-spatial and number-processing difficulties (Ross, Zinn and McCauley, 2000). This study found unusual patterns of growth or orientation of neurons in the right intraparietal sulcus. Similarly, an imaging study of low-birth weight children who had problems with mathematics found that they had reduced neural density in the left intraparietal sulcus (Isaacs, Edmonds, Lucas and Gadian, 2001).

5.4 Other applications of cognitive neuroscience

Neuroscientists have studied how neural systems enable a range of other abilities relevant to education, perhaps most notably bilingualism, geometry and music. For example, geometry and music are thought to be linked to mathematics, because they involve abilities to identify abstract forms and use visuo-spatial processing. It has been proposed that there is also a link here to ‘theory of mind’: the ability to fully understand that others are also thinking, feeling beings, failures of which are said to produce autism (Baron-Cohen, Bolton, Wheelwright and Scahill, 1998). There has also been significant work in cognitive neuroscience exploring the ways that sleepiness influences the ability to learn, remember, and perform tasks of various kinds.

5.5 Special education

Cognitive neuroscience has a potentially close link with special education. On the one hand, it may be especially useful in the development and evaluation of targeted interventions and treatment. On the other, the varying difficulties of children in special education may provide natural ‘experiments’ that help cognitive neuroscientists identify the brain systems associated with particular kinds of ability. Goswami (2004b) proposes that imaging techniques might be used to distinguish between different cognitive theories of dyslexia. There are theories implicating phonological deficit, visual recognition and dysfunctions of movement associated with the cerebellum. Since each theory implies a very different pattern of brain activation, imaging studies of children identified as dyslexic might be able to test their validity. She also proposes that imaging could be used to distinguish between delayed development and deviant development, and so find out whether children who fail to meet developmental milestones are simply developing more slowly, or whether they are developing differently.

6 Research in social and affective neuroscience

Social neuroscience examines how neural systems are involved in socio-cultural processes. Exponents describe it as an attempt to understand the brain systems that enable social behaviour by combining insights and approaches from biological and social research (Cacioppo and Berntson, 1992; Harmon-Jones and Winkielman, 2007). Social neuroscientists do this by examining how the brain mediates social cognition, interpersonal exchanges, group interactions, and relationships (Decety and Keenan, 2006).

One of the most remarked upon recent findings in social neuroscience is the discovery of mirror neurons (Gallese and Goldman, 1998). These are neurons which fire, both when an action is performed and when we see the same action being performed by another. Mirror neurons might be thought of as supplying part of the neural basis for experiences of empathy, since they seem to offer the possibility of quite literally feeling something of what another person feels. They might contribute to the learning of embodied skills by helping us anticipate how the performance of them might feel, and may also be implicated in interpersonal sensitivity and so-called emotional intelligence. It has even been suggested that mirror neurons help to supply the ‘theory of mind’ posited by Baron Cohen to be lacking in autism, and some research suggests that there may be deficits in the mirror neuron systems of children diagnosed with autism (Dapretto et al, 2005).

Affective neuroscience investigates the brain systems involved in processing emotion, or affect. Much of the foundational work in this area was conducted using animals (eg Panksepp, 1998) and provided the basis for an understanding of how the mammalian brain enables emotion and feeling. More recent work with humans (Damasio, 1999; Le Doux, 2000) has further developed this understanding, and simultaneously challenged the idea that ‘cognitive’ processes typically operate without input from emotional systems. Increasingly, affective neuroscience has provided evidence to show that emotion guides and assists thinking and decision-making, rather than simply impeding or biasing them

Currently, the best-known research in affective neuroscience is probably that reported by Damasio (1994, 1999, 2003). Damasio has conducted numerous studies (some experimental, some based upon work brain-injured people) demonstrating how the frontal lobes contain vital nodes of the ’somatic marker’ system, which is important in enabling social conventions and ethical rules to be acquired and used. Learning and remembering are central to education, and research in affective neuroscience has uncovered associations between memory and emotion, to which the hippocampus and amygdala are frequently thought to be important. Other aspects of affective neuroscience are sometimes seen as relevant to education because of their possible relevance to affective and mood disorders, and behavioural problems such as those associated with diagnoses of ADHD.

For the most part, however, educators have shown relatively little interest in the findings of social and affective neuroscience, despite the obvious relevance to their practice of social relations, social influence, emotions and feelings. In a rare paper discussing the potentials for education of these sub-disciplines, Imordino-Yang and Damasio (2007) survey evidence linking the frontal lobes to emotionality and to decision-making in social settings. They argue that because emotion guides cognition, it is especially important to the transfer of skills from the classroom. Whilst schools can teach basic skills, the choice of which skills to actually use takes place in wider social settings and is guided by emotional imperatives. They note that emotion also plays a significant role in moral reasoning and social relations, as well as contributing centrally to art, literature, self expression and creativity in general.

As this brief overview suggests, neuroscience does indeed hold much promise for education. However, there are also problems that make the ready application of neuroscience to education more difficult than it might at first seem. Some of these problems will now be described, under three separate headings: conceptual limitations; limitations of particular methods; and general limitations.

7 Conceptual limitations

7.1. The ‘mereological fallacy’

Bennett and Hacker (2003) demonstrate that much contemporary neuroscience is characterised by what they call the ‘mereological fallacy’. This is their term for a marked tendency in contemporary neuroscience to treat the brain as though it were simply equivalent to the person or the mind. Once this erroneous assumption is made, it can lead to errors of reasoning and interpretation.

Examples of the mereological fallacy appear in one paper which makes the restrictive claim that education “involves the shaping of individual brains via targeted experience in the classroom” (Szucs and Goswami, 2007), and in another which defines the purpose of educational neuroscience as “nurturing the brain” (Ito, 2004). Although not strictly untrue, these claims are highly selective. In taking such a narrow focus they seem to downgrade or exclude many aspects of education that practitioners would see as vital. In offering such a partial view of the goals and the nature of education, they are potentially misleading with respect to the educational potentials of neuroscience.

The mereological fallacy can also mislead neuroscientists themselves with respect to the meaning of their own findings. For example, Bennett and Hacker (2003) describe how it leads neuroscientists to use the metaphor of ‘maps’ in the brain (patterns of brain activation systematically related to features of a stimulus) and then mistakenly claim that the brain ‘reads’ these ‘maps’ just as a person reads an atlas – even though, both logically and practically, this simply cannot be the case. So the mereological fallacy has the potential to foster deep conceptual errors in contemporary neuroscience, errors which may impact upon its educational application.

7.2 Reductionism

Reductionism is the favouring of explanations at the smallest, most basic level. In the present context it involves the assumption that brain processes and features can themselves explain our abilities and experiences. For example, if atypical patterns of brain activation are found amongst individuals who have reading difficulties, a reductionist interpretation might be that these atypical patterns of activity are the cause of their difficulties. An alternative, non-reductionist interpretation is that atypical abilities might well be associated with atypical brain activity, but that this tells us little about the origins of the difficulties.

Forms of reductionism are often seen as essential to good science, but the concern here is that reductionism might get over-extended. Stanovich (1998) welcomes the application of neuroscience to education but simultaneously notes that our ability to identify precisely how the brain constrains learning is still quite primitive, even for a skill so extensively studied as reading. He warns that without adequate psychological and behavioural accounts of reading difficulties, applications of neuroscience to this problem will always be inappropriately reductive.

7.3 Reification

Reification is the ‘making real’ of a phenomenon in a certain way. An example in education is the reification of some children’s disruptive and disorganised behaviour as an instance of the psychiatric diagnosis Attention Deficit-Hyperactivity Disorder (ADHD). This is not to deny that some children behave inappropriately and seem to have trouble concentrating, nor to deny that such children can be disruptive and difficult for both parents and teachers. The concern arises when these problems simply get attributed to a disease – ADHD – which, in turn, gets associated with a putative neural flaw (eg Rubia, 2002)

ADHD is a controversial diagnosis, and its validity has been questioned. Timimi (Timimi and Taylor, 2004) observes that the recent ‘epidemic’ of ADHD is difficult to explain if the condition has a neural or genetic basis. He notes that prevalence rates for ADHD vary hugely, and that imaging studies have used unacceptably small samples, failed to control for the effects of medication, and failed to produce consistent evidence of a brain abnormality. He proposes instead that we should look at the ways in which biological and social immaturity become meaningful in a culture where extended family support is frequently absent, schools are pressured and lack moral authority, and parents and families are themselves frequently ‘hyper-active’, over-worked, and unsure of the appropriate ways to discipline their children.

Clearly, the contribution that we imagine neuroscience can make to the resolution of these kinds of problems will differ greatly according to whether or not we conceptualise ADHD as ‘real’ in this way

7.4 Unjustified normativity

In order to be sure that a pattern of brain activation related to a particular ability is dysfunctional, we need to know what functional brain activation looks like. However, the brain’s parallel processing and associated redundancy mean there is often more than one way for it to complete an activity. Steven Rose (1997) argues that many brain functions have what he calls a ‘norm of reaction’, meaning that they exhibit some degree of adaptive flexibility in the face of challenges. So long as challenges remain within typically-occurring limits, the brain can adapt and cope; outside of this range performance tails off sharply. The limits within which performance is largely unaffected provide the norm of reaction.

This concern can be illustrated by considering the relationship between IQ and the brain. Although identical twins reared together have highly-correlated IQ scores, imaging studies show that the relative sizes of different areas of their brains are sometimes significantly different (Steinmetz, Herzog, Schlaug, Huang and LAnke, 1995). At the neural level there is no single route to the same IQ score, and so attempts to find ‘optimal’ or ‘ideal’ brain features enabling performance on these tests are likely to fail.

This has two implications. Firstly, that an unusual pattern of neural activity does not necessarily lead to a performance deficit, and so might not be useful diagnostically in the absence of behavioural or performance indicators. Secondly, that some caution is needed when interpreting findings, even when a suitable control group has been used. Research might be applied with more confidence if the relevant norms of reaction were known, but even for widely studied abilities we are still very far from having this kind of population-level data.

These conceptual problems are interlinked and mutually-reinforcing, such that one may frequently lead to, or be associated with, the other. At the same time they feed into, and are sometimes either amplified or concealed by, methodological limitations.

8 Limitations of particular methods

8.1 Animal experiments

Although the core structure of all mammalian brains is quite similar, there are significant inter-species differences with respect to the relative size and development of different areas. Studies that demonstrate sensitive periods for the acquisition of basic abilities in animals find variation even between experimental species. Similarly, the timing and duration of periods of synaptogenesis varies significantly. Most obviously, with respect to education, animals simply do not have many of the complex cognitive abilities with which educators are frequently concerned.

8.2 MRI and fMRI

These scanners have reasonably good spatial resolution and are able to identify areas of activity down to a scale of around 1mm. This is still extremely coarse in relation to the multiple branching dendrites that connect neurons, but it is adequate to resolve patterns of activity to broadly specifiable brain regions.

However, the temporal resolution of MRI scanning is poor. The scanner works by assessing levels of blood oxygen: the more oxygen being taken up by the neurons in an area, the more neural activity is assumed to be taking place. By this means the scanner can only identify activity occurring over fairly lengthy periods of time – as a minimum, around one second. Many brain processes, which typically take only milliseconds to initiate and complete, are therefore missed. Moreover, the temporal sequence of activation between different regions is often impossible to discern.

8.3 EEG

Whilst its temporal resolution is excellent the spatial resolution of EEG is very poor, and it can sometimes only locate signals to within one or other of the brain’s cortical hemispheres.

8.4 fNIRS

The spatial resolution of fNIRS is significantly less than fMRI, being no more accurate than one square centimetre, although its temporal resolution is better (down to 0.01 seconds). However, fNIRS has two other important limitations. First, it can only image relatively small sections of the brain at a time. Second, it can only image the cortex immediately below the skull, and cannot resolve activity occurring deeper inside the brain.

9 General limitations

Whilst each of the research methods used in neuroscience has its own specific limitations, there are also more general limitations that flow from the need to generate scientifically acceptable, valid and reliable data, under controlled conditions and using recognised procedures. Whilst these conditions and procedures are necessary to obtain robust findings, they inevitably carry their own costs.

9.1 The subtraction method

Results of imaging studies are typically obtained by comparing the average patterns of brain activation between groups performing a target task and groups performing a control task. Comparisons are usually made by subtracting the pattern of activation in the control task from the pattern in the target task. Whatever significant activation remains is then assumed to be associated with features of the target task. But this means that if both tasks activate the same brain region, the results of the comparison will suggest that this region is not significantly activated.

Neuroscientists considering this problem have identified a set of brain areas called the default network, which seem to be activated in all imaging studies. The default network is thought to be composed of interacting subsystems located primarily in the temporal and pre-frontal lobes, and to enable activities such as planning for the future, recalling the past, evaluating the actions of others, and assessing the likely outcome of decisions (Buckner, Andrews-Hanna and Schacter, 2008). Whilst these activities are clearly relevant to education, learning, and the transfer and application of knowledge and skills, their investigation using imaging techniques will be difficult.

9.2 Groups and averages

Imaging studies typically compare group averages in the performance of different tasks. As a consequence, when findings are reported individual variations in performance are often rendered mostly invisible. Conversely, patterns of ‘average’ activation can emerge that were not found in any of the actual individuals studied (Cacioppo et al, 2003). Studies show there are often differences in the patterns of brain activation of individuals performing the same task, as well as marked individual differences between individuals in the relative size of different brain structures. These findings suggest that caution is needed when applying findings from studies of groups to individual cases.

9.3 Practical problems

With the exception of EEG and fNIRS, imaging procedures require participants to keep entirely still, because head movement interferes with data collection. Although some head movement is possible with EEG, these studies typically take place in dimly-lit, soundproofed rooms, and movement is still not entirely free.

MRI and fMRI scanners completely surround participants’ heads, occlude their field of vision, and make a loud humming noise during operation. Participants are unable to move, gesture, or interact and communicate naturally with others. Stimuli are presented via headphones or on a small LCD screen, and responses communicated by pressing buttons with fingers.

A constant concern in imaging studies is that random environmental stimuli and participant movements or reactions might interfere with the relatively transient biological signals being measured. Often, this problem is managed by study designs that require participants to undergo many trials of the same repetitive task. This can mean that experiments become quite lengthy, often lasting for two or more hours.

One consequence of all these problems is that social interaction is almost impossible, cannot be conducted spontaneously and by naturalistic means, and is often excluded from consideration. Another is that many abilities and tasks are studied using ‘analog’ versions of those displayed in everyday life. With respect to education, a particularly important consequence is that the alienating, repetitive, time-consuming nature of many studies is likely to make them especially difficult for children.

10 Current progress assessed

Both the problems and the potentials of neuroscience with respect to education have been extensively discussed in recent years, and scholars have taken up a range of positions.

Some think that the application of neuroscience to education is a fundamentally flawed enterprise because the distance between learner and brain, classroom and neuron, is simply too great. In one much cited commentary, Bruer (1997) describes the link between education and neuroscience as ‘a bridge too far’. He observes that there is already a link between education and cognitive psychology, and – in cognitive neuroscience – a link between cognitive psychology and the brain. However, he argues, attempts to forge a direct link between education and neuroscience are forever doomed to failure because cognitive psychology will always be needed to mediate between them.

Bruer’s arguments recognise many of the limitations described above, which clearly do create problems for any straightforward application of neuroscience to education. Many of these limitations are also recognised even by those who, compared with Bruer, are relatively optimistic (eg Byrnes and Fox, 1998; Szucs and Goswami, 2007). However, it is less widely accepted that it must always be cognitive psychology that mediates between educational practice and neuroscientific research; some, for example, propose using perspectives from evolutionary psychology and developmental systems theory (Brown and Bjorklund, 1998). It is also fair to say that, in these discussions, the methodological limitations receive far more attention than the kinds of conceptual limitations identified in this report.

Other scholars, whilst acknowledging many of these methodological limitations, argue that progress is already being made in applying neuroscience to education, albeit more slowly than previous unrealistic expectations might have suggested. Since cognitive neuroscience is a hybrid that already includes elements of cognitive psychology, these scholars see the possibility of direct links between education and neuroscience. They argue that advances have been made – for example, in the study of bi-lingualism, language acquisition and conceptual change – that can be applied more or less directly both to the identification of difficulties and the evaluation of treatment programmes (Pettito and Dunbar, in press). They argue that neuroscience can inform education with respect to the early diagnosis of special needs, the evaluation of interventions, and an increased understanding of individual differences in learning (Goswami, 2004a).

Consequently, there have been recent calls for a distinct strand of ‘educational neuroscience’. Pettito and Dunbar (in press) describe educational neuroscience as ‘an exciting and timely new discipline’ whilst Szucs and Goswami (2007) call for an educational neuroscience involving the study of ‘mental representations’ distributed across multiple brain areas and encoded in neural networks. Educational neuroscience, they propose, will identify biological constraints upon the patterns of neural activation that encode mental representations, and use related neural markers in the early identification of problems. For example, they suggest that the N100 response to auditory stimuli in babies is a reliable marker of auditory processing that might be used to identify potential language difficulties, even “when there are no cognitive or behavioural variables at all”. They similarly propose that neuro-imaging might be used “in populations for whom informative behavioural data are difficult to collect”, for example children with attention difficulties.

Despite this optimism, controversy may continue. For example, Harre’s (2002) rigorous guide to cognitive science explicitly warns against the use of neural markers in the absence of other variables. He notes that three levels of description (persons, organisms and molecules) are always needed to provide meaningful accounts of human activity, since causality operates differently at each level. He argues that because these levels are nested together in a context defined first of all by the embodied social relations of living persons, it is only valid to look for neural correlates (organism or molecule level) if we can first be confident that the ability concerned is actually being enacted (person level). Whilst Harre’s arguments are not directed specifically toward either educational neuroscience or this particular example, they are relevant here. Similarly, the view that imaging techniques could easily be used to identify the neural bases of attentional difficulties in children perhaps ignores many of the limitations described above. It is a form of reductionism; it sidesteps the problems with ADHD that Timimi and others have described; and it downplays the practical difficulties of persuading typically restless children, who presumably might not co-operate readily with a classroom test, to nevertheless submit to the typically extensive and sustained constraints of neuro-imaging.

In this regard, it is perhaps notable that despite the many links between neuroscience and special education, the application of neuroscience to intellectual impairment (learning disability) has so far been negligible. The perennial lack of funding for intellectual impairment research is no doubt a contributory factor. However, it is undoubtedly also because of the ethical and practical difficulties of using imaging technologies with intellectually-impaired people, who may neither tolerate, nor understand the need for, the constraints on movement, vision and interaction that are typically required. Children with intellectual impairments are perhaps the archetypal population for whom meaningful cognitive and behavioural data are hard to collect (Hall, 1984). If neuro-imaging is not as readily applicable to this population as Szucs and Goswami propose, then some of their optimism with respect to its potentials is maybe premature.

There is nevertheless widespread recognition that neuroscience has made conspicuous progress in recent years, and that brain imaging technologies which allow real-time, non-invasive imaging have opened up many new avenues of investigation. It is also recognised that cognitive, social and affective neuroscience have uncovered aspects of the ways in which brain systems enable activities – such as reading – that are central to education. These findings may be useful in the design of educational interventions, in the formulation and testing of theories of skill acquisition, learning and teaching, and in the identification and treatment of various kinds of difficulties. Moreover, even amongst those who disagree about the significance of these findings, areas of consensus have begun to emerge. Most scholars do accept that there are clear possibilities for neuroscience to inform educational practice. At the same time, most recognise that many findings are still far too general and provisional to be applied directly in educational settings. Consequently, it is also widely accepted that progress will be slow, and that the potentials of neuroscience for education have for the most part not yet been realised.

Similarly, in the wake of the discussions over neuromyths, there is some recognition that applying neuroscience to education is not straightforward. Amongst other things, this suggests an emergent consensus that many established educational practices will continue to be as vital as they were before. Relatedly, there is growing recognition that evidence from neuroscience will always need to be weighed against evidence derived from other sources for its import to be properly evaluated. Even Byrnes and Fox (1998) who are very enthusiastic about applying neuroscience to education, recommend that neuroscientific evidence should only be considered important by educators when it both gets confirmed by multiple methods (eg both fMRI and EEG), and when it accords with other evidence already gathered using cognitive or behavioural measures.

So it seems that neuroscience will continue to be only one of many influences upon education, and that its impact will remain contingent upon it’s binding with events, practices and knowledge arising elsewhere. With this in mind, some possible future implications of neuroscience for education will now be sketched: first, with regard to developments that seem likely in most probable futures, and second with respect to possible developments that are more contingent and uncertain.

11 Probable futures

11.1 Emergence of a sub-discipline

Publications dealing with the educational implications of neuroscience have steadily increased in recent years: a new journal (Mind, Brain and Education) has been launched, research groups (eg Cambridge University’s Centre for Neuroscience in Education) have been established, and there have already been explicit calls for a new sub-discipline of educational neuroscience. Whilst no consensus has yet emerged, there seems to be a convergence of academic, educational, institutional and commercial interests here which may make some such development all but inevitable (timescale: less than five years)

11.2 More effective targeting of evaluation and remediation

Neuroscientists have already made substantial progress in identifying the basic abilities underpinning many skills. This progress seems likely to continue, yielding yet more fine-grained knowledge of the nature of some educational difficulties. This may lead to more precise evaluation of these difficulties, and the application of more targeted programmes of remediation. These developments are likely to be most rapid with respect to reading and language learning, and to involve tailoring existing interventions rather than developing new ones (timescale: five to ten years)

11.3 The mutation of cognitive psychology, and its effects

Partly due to its associations with neuroscience, cognitive psychology is already mutating, and this process seems likely to gather pace. There is increasing interest in the approaches known as embodied, enactive, embodied or situated cognition, and a growing recognition that human cognition is never disembodied information processing. As cognitive psychology mutates, its changes will impact back upon education. As a result, it seems likely that findings in social and affective neuroscience, to date largely ignored, will appear more relevant. This, in turn, will lead to renewed educational emphasis on the social and emotional dimensions of teaching and learning (timescale: five to fifteen years)

11.4 Emergence of new neuromyths

Debates about the potentials and dangers of applying neuroscience to education seem set to continue. Similarly, there will be ongoing political and institutional pressures to improve educational outcomes, and commercial interests will continue to operate in the education sector. In a climate where the conceptual limitations of neuroscience are infrequently recognised, this creates the potential for new neuromyths (timescale: five to ten years)

12 Possible futures

12.1 Proliferation of fNIRS

The portability and flexibility of fNIRS could make it ideal for use in education, making possible in-situ neural testing in the classroom. Because it is limited to imaging small sections of cortex immediately below the skull, it may be most helpful in identifying atypical neural activity associated with quite specific cognitive difficulties. However, its use is contingent on two developments. First, more fine-grained knowledge of the neural bases of those difficulties (as in 11.2 above); second, the technology becoming affordable and widely available (timescale: ten to fifteen years)

12.2 Population-level norms of reaction

Many universities and hospitals now have brain scanners, creating the potential to image the brains of very large numbers of people as they conduct the same tasks. In this way, population level norms of reaction for abilities such as reading might be established which would greatly assist interpretation of imaging data. However, this would require substantial funding (timescale: five to fifty years)

12.3 The impact of DSM V

At periodic intervals, the diagnostic systems psychiatrists use to classify problems (such as ADHD) are subject to revision. DSM V, the new version of the most influential system, is expected in May 2012, although its content is presently unknown. Depending on how the psychiatric categories most immediately relevant to education (primarily conduct and mood disorders) are modified in DSM V, there will be consequences: for education generally, and particularly for the relationship of neuroscience to education (timescale: five years and more)

12.4 Improvements in imaging

Scientists continue to develop imaging technologies, seeking improved power, spatial and temporal resolution, portability and flexibility. Any such developments will impact upon neuroscience as it relates to education, perhaps especially so if they enable more investigation of the default network and greater understanding of temporal sequences of brain activation (timescale: any)

12.5 Other technological developments

Developments in neuroscience are already bound up with those elsewhere, so its influence upon education will to some extent depend on advances in other technologies, specifically, genomics, psychopharmacology, and nano-technology. Genomics identifies patterns of genetic transmission and might help illuminate the neural bases of some difficulties, leading to possible new interventions. Developments in psychopharmacology already impact upon education through the occasional use of cognitive enhancers, which seems likely to continue amongst some groups. Other applications depend on developing methods of drug delivery for brain chemicals such as peptides, which are known to be psychoactive but cannot be effectively administered. If this problem could be solved, Panksepp (2004) identified the possibility of ’socio-emotional education’ using peptides with vulnerable children. Silva (2006) notes that nano-technology might be used to develop new molecules suitable for such purposes, as well as for purposes such as targeted neural regeneration (timescale: any)

Bibliography

Baron-Cohen, S., Bolton, P., Wheelwright, S. and Scahill, V. (1998) Autism occurs more often in the families of physicists, engineers and mathematicians. Autism, 2, pp.296-301.

Beaumont, J.G., Kenealy, P.M. and Rogers, M.J.C. (1996) The Blackwell Dictionary of Neuropsychology. Oxford, Blackwell.

Bennett, M.R. and Hacker, P.M.S. (2003) Philosophical Foundations of Neuroscience. Oxford, Blackwells.

Brown, R. and Bjorklund, D. (1998) The biologizing of cognition, development and education: approach with cautious enthusiasm. Educational Psychology Review, 10 (3), pp.355-373.

Bruer, J. (1997) Education and the brain: a bridge too far. Educational Researcher, 26 (8), pp.4-16.

Buckner, R., Andrews-Hanna, J. and Schacter, D. (2008) The brain’s default network: anatomy, function and relevance to disease. Annals of the New York Academy of Sciences, 1124, pp.1-38.

Byrnes, J. and Fox, N. (1998) The educational relevance of research in cognitive neuroscience. Educational Psychology Review, 10 (3), pp.297-342.

Cacioppo, J. and Berntson, G. (1992) Social psychological contributions to the decade of the brain: doctrine of multilevel analysis. American Psychologist, 47 (8), pp.1019-1028.

Cacioppo, J., Berntson, G., Lorig, T., Norris, C., Rickett, E. and Nusbaum, H. (2003) Just because you’re imaging the brain doesn’t mean you can stop using your head: a primer and set of first principles. Journal of Personality and Social Psychology, 85 (4), pp.650-661.

Chow, K. and Stewart, D. (1972) Reversal of structural and functional effects of long-term visual deprivation in cats. Experimental Neurology, 34, pp.409-433.

Damasio, A.R. (1994) Descartes Error: emotion, reason and the human brain. London, Picador.

Damasio, A.R. (1999) The Feeling of What Happens: body, emotion and the making of consciousness. London, William Heinemann.

Damasio, A.R. (2003) Looking for Spinoza: joy, sorrow and the feeling brain. Orlando, Harvest.

Dapretto, M., Davies, M., Pfeifer, J., Scott, A., Sigman, M., Bookheimer, S. et al. (2005) Understanding emotions in others: mirror neuron dysfunction in children with autism spectrum disorders [Electronic Version]. Nature Neuroscience. Available from http://www.fil.ion.ucl.ac.uk/SocialClub/nn1611_Mirror_Neurons_in_autism.pdf. Accessed 29/11/08

Daugman, J. (2001) Brain metaphor and brain theory. In: Bechtel, W., Mandik, P., Mundale, J. and Stufflebeam, R. (eds) Philosophy and the Neurosciences: a reader, pp.23-36. Oxford, Blackwells.

Decety, J. and Keenan, J. (2006) Social Neuroscience: A New Journal [Electronic Version]. Social Neuroscience. Available from http://www.cognitiveneurosciencearena.com/whatissocialneuroscience.asp. Accessed 29/11/08

Dehaene, S., Molko, N., Cohen, L. and Wilson, A. (2004) Arithmetic and the brain. Current Opinion in Neurobiology, 14, pp.218-224.

Gallese, V. and Goldman, A. (1998) Mirror neurons and the simulation theory of mind-reading. Trends in Cognitive Sciences, 2 (12), pp.493-501.

Gogtay, N., Giedd, J., Lusk, L., Hayashi, K., Greenstein, D., Vaituzis, A. et al. (2004) Dynamic mapping of human cortical development during childhood through early adulthood. Proceedings of the National Academy of Science (USA), 101 (21), pp.8174-8179.

Goswami, U. (2004a). Neuroscience and Education. British Journal of Educational Psychology, 74, pp.1-14.

Goswami, U. (2004b) Neuroscience, education and special education. British Journal of Special Education, 31 (4), pp.175-183.

Hall, D.M.B. (1984) The Child with a Handicap. Oxford, Blackwell Scientific Publications.

Harmon-Jones, E. and Winkielman, P. (2007) Social Neuroscience. New York, Guilford Press.

Harre, R. (2002) Cognitive Science: a philosophical introduction. London, Sage Publications.

Hubbard, E., Piazza, M., Pinel, P. and Dehaene, S. (2005) Interactions between number and space in parietal cortex. Nature Reviews Neuroscience, 6, pp.435-448.

Imordino-Yang, M. and Damasio, A.R. (2007) We feel therefore we learn: the relevance of affective and social neuroscience to education. Mind, Brain and Education, 1 (1), pp.3-10.

Isaacs, E., Edmonds, J., Lucas, A. and Gadian, D. (2001) Calculation difficulties in children of very low birth weight: a neural correlate. Brain, 124, pp.1701-1707.

Ito, M. (2004) ‘Nurturing the brain’ as an emerging research field involving child neurology. Brain Development, 26 (7), pp.429-433.

Le Doux, J. (2000) Emotion Circuits in the Brain. Annual Review of Neuroscience, 23, pp.155-184.

Panksepp, J. (1998) Affective Neuroscience. Oxford, Oxford University Press.

Panksepp, J. (2004) Emotional Consciousness: animalian sources of affective values and the core self. Paper presented at the Science, Self and Meaning conference (annual conference of the Consciousness and Experiential Psychology Section of the British Psychological Society), Oxford.

Paulesu, E., Demonet, J.-F., Fazio, F., McCrory, E., Chanoine, V., Brunswick, N. et al. (2001) Dyslexia: cultural diversity and biological unity. Science, 291, pp.2165-2167.

Pettito, L. and Dunbar, K. (in press) New findings from educational neuroscience on bilingual brains, scientific brains, and the educated mind. In: Fischer, K. and Katzir, T. (eds), Building Usable Knowledge in Mind, Brain and Education. Cambridge, Cambridge University Press.

Roder, B. and Neville, H. (2003) Developmental functional plasticity. In: Grafman, J. and Robertson, I. (eds), Handbook of Neuropsychology, 2^nd edition. Oxford, Elsevier Science.

Rose, N. (2001) The Politics of Life Itself. Theory, Culture and Society, 18 (6), pp.1-30.

Rose, S. (1997) Lifelines: life beyond the gene. Oxford, Oxford University Press.

Ross, J., Zinn, A. and McCauley, E. (2000). Neurodevelopmental and psychosocial aspects of Turner Syndrome. Mental Retardation and Developmental Disability Research Review, 6, pp.135-141.

Rubia, K. (2002) The dynamic approach to neurodevelopmental psychiatric disorders: use of fMRI combined with neuropsychology to elucidate the dynamics of psychiatric disorders, exemplified in ADHD and schizophrenia. Behavioural Brain Research, 130 (1-2), pp.47-56.

Shaywitz, B., Shaywitz, S., Pugh, K., Mencl, W., Fulbright, R., Skudlarski, P. et al. (2002) Disruption of posterior brain systems for reading in children with developmental dyslexia. Biological Psychiatry, 52 (2), pp.101-110.

Silva, G. (2006) Neuroscience nanotechnology: progress, opportunities and challenges. Nature Reviews Neuroscience, 7, pp.65-74.

Slavin, R. (1999) Faddism in education, and its alternatives. In: Cizek, G. (ed) Handbook of Educational Policy. Orlando, Academic Press.

Stanovich, K. (1998) Cognitive neuroscience and educational psychology: what season is it? Educational Psychology Review, 10 (4), pp.419-426.

Steinmetz, H., Herzog, A., Schlaug, G., Huang, Y. and LAnke, R. (1995) Brain (a)ssymetry in monozygotic twins. Cerebral Cortex, 5, pp.296-300.

Szucs, D. and Goswami, U. (2007) Educational neuroscience: defining a new discipline for the study of mental representations. Mind, Brain and Education, 1 (3), pp.114-127.

Temple, E., Deutsch, G., Poldrack, R., Miller, S., Tallal, P., Merzenich, M. et al. (2003) Neural deficits in children with dyslexia ameliorated by behavioral remediation: Evidence from functional MRI. Proceedings of the National Academy of Science (USA), 100 (5), pp.2860-2865.

Timimi, S. and Taylor, E. (2004) ADHD is best understood as a cultural construct. British Journal of Psychiatry, 184, pp.8-9.

This document has been commissioned as part of the UK Department for Children, Schools and Families’ Beyond Current Horizons project, led by Futurelab. The views expressed do not represent the policy of any Government or organisation.

Adolescence in humans is the period of psychological and social transition between childhood and adulthood. The beginning of adolescence, around the onset of puberty, is characterised by dramatic changes in hormone levels and, as a result, in physical appearance. This period of life is also characterised by psychological changes in terms of identity, self-consciousness and mood. After puberty, children become more aware of the opinions and emotions of both themselves and other people around them. The typical teenager is more moody and uncommunicative than a younger child (at least towards adults), and may take unnecessary risks. In addition, there are educational changes: some children go through an educational performance dip in Year 8 (age 12-13). Hormonal fluctuations alone might not account for these changes; recent neuroscience research shows that there are also dramatic transformations in the brain during adolescence.

Much has been known about early brain development since experiments on animals carried out in the 1950s and 1960s. One major developmental process affects the ‘wiring’ of brain cells (neurons) – the intricate network of connections (synapses) between neurons. Early in development, the brain begins to form new synapses, so that the synaptic density – the number of synapses per unit volume of brain tissue – greatly exceeds adult levels. This process of synaptic proliferation, ’synaptogenesis’, lasts up to several months, depending on the species of animal (Rakic et al, 1986; Bourgeois et al, 1994).

The increase in the number of synapses is followed by a period of synaptic elimination (or pruning) in which excess connections wither away. This process is pre-programmed to a large extent – it will happen in all environments. However, the environment can also influence synaptic pruning (Hubel and Wiesel, 1964), in that frequently used connections are strengthened and infrequently used connections eliminated. Indeed, this research has been used to argue that ages zero to three represent a ‘critical period’ for brain development. However, this argument neglects the fact that the animals in which this early development research was carried out, such as cats and monkeys, do not go through the same extended developmental period as humans, and are sexually mature at a much younger age.

This research suggested that brain development is particularly sensitive to environmental influences very early in life. It was not until the 1970s that research on post-mortem human brains revealed that some areas of the human brain, in particular the frontal cortex, continue to develop well beyond childhood. The frontal cortex is the area responsible for cognitive abilities such as the ability to make plans, to remember to do things in the future, to multi-task, and it inhibits inappropriate behaviour (executive functions). The frontal cortex also plays an important role in self-awareness and understanding other people. Peter Huttenlocher, at the University of Chicago, collected post-mortem brains from humans of all ages and found that the frontal cortex was remarkably different in the brains of pre-pubescent children and post-pubescent adolescents (Huttenlocher et al, 1979, 1983, 1997). While in sensory brain areas such as the visual cortex, synaptogenesis and synaptic pruning occur relatively early and synaptic density has reached adult levels by mid-childhood, synaptic reorganisation in the frontal cortex continues until well into adolescence. Huttenlocher found that the number of synapses in the frontal lobe is high around puberty, after which their number decreases (due to synaptic pruning) throughout adolescence.

Another developmental mechanism that occurs for several decades in the frontal cortex is myelination. As neurons develop, they build up a layer of myelin on their axon (the long fibre transmitting signals from each brain cell). Myelin is a fatty substance that insulates the axons and vastly increases the speed of transmission of electrical impulses from neuron to neuron. Whereas sensory and motor brain regions become fully myelinated in the first few years of life, axons in some cortical regions, particularly the frontal and parietal lobes, continue to be myelinated well into adolescence in the human brain (Yakovlev and Lecours, 1967). This finding suggests that the transmission speed of neurons in these areas may increase after puberty.

Recent MRI studies of the developing brain

Until recently, the structure of the human brain could be studied only after death. In recent years, non-invasive brain imaging techniques, particularly Magnetic Resonance Imaging (MRI), have enabled scientists to study development of the living human brain. In the past decade, a number of MRI studies have provided further evidence of the ongoing maturation of the cortex into adolescence and even into adulthood. These studies show that the amount of white matter in various cortical regions, including the frontal cortex and temporo-parietal cortex, increases between childhood and adulthood (Giedd et al, 1999; Paus et al, 1999, 2001; Durston et al, 2001). Myelin appears white in MRI scans. Therefore, the increase in white matter seen to occur throughout adolescence may represent an increase in axonal myelination.

At the same time, there is a change in the volume of grey matter (made up of cell bodies, dendrites and synapses) in various cortical regions during adolescence (Giedd et al, 1999; Sowell et al, 1999; Gogtay et al, 2004). Several large MRI studies, which have acquired brain scans from hundreds of people of different ages, have consistently shown that grey matter volume in the frontal cortex increases gradually during childhood and peaks at around the onset of puberty (around 11 in girls and 12 in boys) (Giedd et al, 1999). This is followed by a gradual decrease in the volume of grey matter during adolescence and early adulthood. It has been suggested that this pattern of grey matter development may, in part, be due to an increase in the number of synapses during childhood, followed by synaptic pruning during adolescence (eg Giedd et al, 1999).

Cognitive development during adolescence

The brain regions that undergo particularly protracted development during adolescence, that is prefrontal and temporo-parietal cortices, are involved in a variety of cognitive abilities, including executive functions and social cognition. In the past few years, empirical research has looked at the cognitive changes that occur during adolescence. As yet we do not know how these relate directly to the structural brain changes described above, but research in this area is progressing rapidly.

Development of self-concept during adolescence

Anecdotal evidence and self-reported data indicate that children become progressively self-conscious and concerned with other people’s opinions as they go through puberty and adolescence (Adams and Berzonsky, 2003). The period of adolescence seems to involve both the establishment of a sense of self as well as a process of orienting towards others. The emergence of the social self is marked by a period of heightened self-consciousness, during which adolescents become preoccupied with other people’s concerns about their own actions, thoughts and appearance. Social psychological studies have investigated changes in social thinking during adolescence and emphasise that this phase is characterised by a focus on “what other people think”.

Social psychologists have posited several theories regarding this apparent increase in social and emotional sensitivity. Elkind’s (1967) twin constructs, the Imaginary Audience (IA) and Personal Fable (PF) have been particularly influential. According to the IA theory, adolescents believe that everyone is as concerned about their behaviour as they are, and construct an abstract (and imaginary) audience observing their every move. The PF, on the other hand, is the tendency for adolescents to believe they are unique, invulnerable, and destined for greatness. Between them, the IA and PF cover many behaviours regarded as particular to adolescence: heightened self-consciousness, increased concern with the opinions of others, and susceptibility to peer pressure (IA); and reckless behaviours such as drug use and unprotected sex (PF). Although it is generally accepted that many (though not all) adolescents go through a phase of constructing IAs and PFs, there is little consensus as to why they do so, or the underlying neural basis (Vartanian, 2000).

Elkind’s (1967) own explanation was that the IA and PF arise from cognitive development between childhood and adolescence, and that this phase is characterised by ‘cognitive egocentrism’ ie a difficulty in differentiating one’s own thoughts from those of others. The idea of a developing ’self-concept’ is at the centre of many theories of adolescent social adjustment. Although the precise definition varies between studies, it is generally conceptualised as the way in which individuals view and treat themselves, and is seen as a product of interpersonal interactions (Benjamin, 1993; Ybrandt, 2008). A recent study found that having a negative self-concept (high scores on self-hate, self-neglect and self-blame) was associated with both internalising behaviours such as depression and anxiety, and externalising behaviours such as delinquency and aggression (Ybrandt, 2008). Therefore, it is important that adolescents are supported emotionally as they develop their self-concept. While such findings have immediate real-world applicability, however, they tell us little about the underlying brain mechanisms. Furthermore, it is difficult to disentangle cause and effect. Does having a negative self-concept cause depression and anxiety, or vice versa? Or are both caused by a third factor?

The development of perspective-taking during adolescence

The brain regions that undergo the most significant development during adolescence include those areas involved in self awareness and in the ability to understand other people’s perspectives. Given that the social environment dramatically changes during adolescence, and that the brain undergoes a restructuring process, it might be expected that social cognitive abilities such as self awareness and perspective-taking develop during this period.

We recently investigated the development of perspective taking during adolescence (Choudhury et al, 2006). Pre-adolescent children (age 9 years), adolescents (age 13 years) and adults (age 24 years), were tested using a perspective-taking task. In the First Person Perspective (1PP) condition, the participant was asked to imagine how s/he would feel in various scenarios. An example of such a scenario was “You just had an argument with your best friend. How do you feel?” In the Third Person Perspective (3PP) condition, the participant was asked how someone else would feel in the same set of scenarios. The participant was asked to choose one of two possible emotions in answer to each question, as quickly as possible. The results demonstrated that the difference in reaction time (RT) between 1PP and 3PP decreased significantly with age. The difference in RT in both groups of younger participants was larger and spread almost equally in both directions, whereas among adults there was little difference in timing for 3PP and 1PP. A similar RT to 3PP and 1PP, as shown by the adult group, is likely to indicate the highest proficiency in perspective taking. In contrast, the most pronounced difference in RT between 1PP and 3PP, seen in the pre-adolescent group, would indicate relatively inefficient processing. It might be speculated therefore that, prior to adolescence, the difference in RT reflects an immature cognitive mechanism for perspective-taking. Whether this response pattern among the adolescents is a result of a relative difficulty in differentiating between the first- and third person, or because children of this age group are less inclined, or find it more difficult, to enter into another person’s ‘mental shoes,’ requires further investigation. The differences between age groups may also be influenced by differences in social experience. Compared with children and adolescents, adults are generally more skilled at instinctively inferring the perspectives of other people. Perhaps adults show no difference between RTs for 1PP and 3PP as a result of their mature neural circuitry supporting social cognition, as well as their greater social experience.

Functional development of the social brain during adolescence

Recent functional neuro-imaging studies have investigated social brain development during adolescence and there is some indication that, for social cognitive tasks, activity in the frontal cortex decreases between adolescence and adulthood. A recent fMRI study investigated the development of this ability by asking participants to think about what action they would take given a particular intention (Blakemore et al, 2007). Adolescents (aged 12-18) and adults (aged 22-37) were scanned while answering questions about intentional causality (e.g. “You want to see what’s on at the cinema; do you look in a newspaper?”), or physical causality (eg “A huge tree suddenly comes crashing down in a forest; does it make a loud noise?”). Consistent with the self/social knowledge study described above, adolescents activated part of the dorsal MPFC more than did adults when thinking about their own intentions compared to during physical causality judgments. In contrast, in the same comparison (intentional – physical), adults activated part of the right superior temporal sulcus more than did adolescents.

A different fMRI study investigated the development of high level communication using an irony comprehension task and found that children (aged between 9 and 14) engaged frontal regions (medial PFC and left inferior frontal gyrus) more than did adults (Wang et al, 2006). A similar result was found in a recent fMRI study that investigated changes during adolescence of the neural processing of social emotion in the first- or third-person perspective (Burnett et al, 2008). Adult (age 22 to 32) and adolescent (age 10 to 18) participants read scenarios that described either social emotions (guilt or embarrassment) or basic emotions (fear or disgust), and were asked to imagine these scenarios happening either to themselves (self condition) or to someone else (their mother – other condition). First, activity in the dorsal MPFC during social relative to basic emotion was higher in the adolescent group than in the adult group. Second, the left temporo-parietal junction (TPJ) showed differential activity to protagonist and emotion, depending on age group. Specifically, this region differentiated better between self and other in adults than it did in adolescents, while in adolescents the left TPJ was more responsive to the difference between social and basic emotion irrespective of perspective.

These results suggest that the neural strategy for social and self understanding changes between adolescence and adulthood. Although the same neural network is active, the relative roles of the different areas change, with activity moving from anterior (medial prefrontal) regions to posterior (temporal) regions with age. One possible explanation for the decrease in prefrontal activity between adolescence and adulthood in these studies is that the PFC is still being organised during adolescence, and is therefore less efficient: more activity is required to achieve the same task (see Blakemore, 2008, for review). Further studies are needed to explore this possibility.

Susceptibility to peer influence

Adolescents are particularly susceptible to peer influence (Steinberg and Silverberg, 1986). The consequences of peer influence have been well researched, both in the laboratory and in a socio-cultural context. For example, it has been found that, while adults who commit crimes do so alone, most adolescent crimes are committed with peers (Zimring, 1998). This suggests that peer influence may contribute to teenage engagement in inherently risky activity (although it may also reflect the fact that teenagers spend more time with peers than do adults). A recent laboratory study by Gardner and Steinberg (2005) looked at incidences of risky driving in a car simulation video game when adolescents and adults played either alone or with two friends present. It was found that the presence of peers led to an increase in risky driving, for example, failing to stop at a yellow traffic light, specifically in adolescents. Levels of risk taking did not differ for adult participants depending on whether they were alone or with peers, and adolescents showed the same level of risk taking as adults when they were alone. However, in the presence of peers, the number of risks taken was greatly increased (see Figure 1).

These findings suggest that decision-making and planning are more fallible in adolescents when in the presence of peers. The immaturity of the PFC could explain these consequences, in that the developing system is less able to cope under added emotional pressure. This is a plausible but speculative idea which needs further work. At present, cognitive neuroscience has not addressed whether the immaturity of the PFC could also explain the susceptibility to peer pressure in the first place.

Gender differences

Cognitive changes during adolescence may not be equally applicable to males and females, or they may follow different time courses. Anatomically, sex differences have been reported in grey and white matter volume during adolescence (Yurgelun-Todd et al, 2002), as well as in the time course of neural development (Giedd et al, 1996, 1997), with grey matter volume peaking at around age 11 in girls and age 12 in boys. Links between anatomical differences and behaviour are also apparent. A recent study (Silveri et al, 2006) looked at impulse control (the ability to control and regulate behaviour), and found that different regions of white matter were associated with task performance in male and female teenagers. The findings suggest subtle differences in the brain networks recruited for cognitive control by males and females during development. Such gender differences at the neural level may help to explain behavioural differences in executive processes such as decision-making and risk assessment. For example, several studies have found that, relative to females, male adolescents give more weight to the potential benefit of a risk than its potential costs (Gardner and Steinberg, 2005), and are also more prone to risk-taking in the presence of peers (Parsons et al, 2000).

Gender differences in social behaviour have also been well documented during the adolescent years. For example, young adolescent females are much more likely to use social aggression such as ostracism during interpersonal interaction (Cairns et al, 1989), while males tend to use physical aggression. Additionally, teenage girls who have a negative self-concept (high levels of self-hate, self-neglect and self-blame) are more likely to engage in internalising behaviours, ie depression, anxiety and withdrawn behaviour, while boys tend to engage in more outwardly aggressive externalising behaviours (Moffitt et al, 2001; Roussos et al, 2001).

The next step is to link these social observations with what we are learning about the brain. Brain development patterns may shed light on gender-related differences in the onset of various mental illnesses (Silveri et al, 2006; Kovacs et al, 2003; Evenson et al, 1993). This is especially likely given that the tendency for certain disorders to be more prevalent in one gender than the other first emerges during adolescence (Angold et al, 2004), for example, higher incidences of depression/anxiety in females. However, psychosocial factors such as gender expectations and differential exposure to social stressors (Kendler et al, 2001) may be important in explaining gender differences in both clinical and typical populations. An important goal, therefore, for cognitive neuroscience is to integrate research on the brain with findings from the social and behavioural sciences.

Implications for society and future developments

The teenage years are a time of marginalisation for many, and it could be argued that highlighting differences in behaviour and neuro-anatomy merely serves to increase the sense in which members of this age group are not yet fully active members of society. However, despite meeting the material needs of this age group better than at any previous point in human history, many developed countries, including the UK, are seeing rising rates of mental illness, disaffection and criminal behaviour (UNICEF, 2007). Improving knowledge of adolescent development at the neural, as well as psychosocial, level will only increase the chances of helping those who need it.

In practical terms, research on the development of the neural structures underlying the appraisal of risk and reward may have implications for the criminal justice system (Blakemore and Choudhury, 2006; Greene and Cohen, 2004). Additionally, recent efforts have been made to integrate findings about brain development into educational policy, both for special educational needs, such as autism and dyslexia, and for typical development (Goswami, 2006). One potentially positive implication of the neural development occurring during adolescence is that the teenage brain is well adapted to learning. Although a ‘Year 8 dip’ in academic performance has been reported, this might correspond, at least in part, to the reorganisation of the brain so that it can learn more efficiently. Appropriate education is crucial during the adolescent years. The data suggest that it is not too late for those still struggling with educational attainment.

An important next step is to extend these efforts to the pastoral side of education, in order to inform anti-bullying and extra-curricular policies. One purely speculative possibility is that, just as the environment influences synaptic pruning in the first few years of life, so might it have an impact on the pruning that occurs in the frontal cortex during adolescence. There are no tools as yet to look at pruning in the living human brain. However, if the environment influences synaptic pruning during adolescence, this has implications for what kind of experiences adolescents should encounter, both academically and socially. Secondary school is often socially stressful (Erath et al, 2007), just at the time when the social brain is undergoing profound development. Provide a social environment at school that is more in line with neural maturation might be useful. It might be fruitful to include in the curriculum some teaching on the changes occurring in the brain during adolescence. Adolescents might be very interested to learn about the changes that are going on in their brains.

Medical policy could also benefit from research on the adolescent brain. For example, treatment of substance abuse disorders may require modification for this population in light of the differences in risk and reward circuitry. Similarly, mood disorders such as depression and anxiety may differ between adolescence and adulthood due to differences affecting regulation and emotional sensitivity. Implications are also raised for young people who take recreational drugs such as cannabis, as the effects they have on the developing teenage brain are likely to be different to those on the adult brain, and may have longer term consequences. Indeed, regular cannabis use is associated with a significant increase in the risk of a later schizophrenia diagnosis, and this risk is even higher if the onset of use occurs during adolescence (Arseneault et al, 2004). This may well be because the brain is more vulnerable to the effects of the drug during development (Pope et al, 2003). Education of young people based on new findings about the brain may act as a more effective deterrent against heavy and regular use than current techniques.

Conclusions

Research on neuro-cognitive development during adolescence is still a relatively new field. However, in the past few years there have been some important developments. Research is currently exploring how the brain changes and how these changes might help to explain certain aspects of typically teenage behaviour, such as risk taking and emerging competence in interpersonal interactions. In turn, these findings might contribute to improving the quality of education and pastoral care for this age group, and may also have implications for the way young people are seen in the eyes of the law and are treated by the medical profession. In all, the research discussed throughout this review serves to highlight that adolescents are a distinct sector of society with specific needs.

Acknowledgements

Our research is funded by the Royal Society, the Wellcome Trust and the BBSRC. SJB is a Royal Society University Research Fellow. SB is funded by the Wellcome Trust four year PhD programme in neuroscience at UCL. CS is funded by a BBSRC PhD studentship.

References

Adams, G.R. and Berzonsky, M.D. (2003) The Blackwell Handbook of Adolescence. Oxford, Blackwell.

Angold, A., Costello, E.J. and Worthman, C.M. (2004) Puberty and depression: the roles of age, pubertal status and pubertal timing. Psychol Med, 28 (1), pp.51-61.

Arseneault, L., Cannon, M., Witton, J. and Murray, R.M. (2004) Causal association between cannabis and psychosis: examination of the evidence. Br J Psychiatry, 184, pp.110-117

Benjamin, L.S. (1993) Interpersonal diagnosis and treatment of personality disorders. New York, Guildford Press.

Blakemore, S.J. and Choudhury, S. (2006) Development of the adolescent brain: implications for executive function and social cognition. Journal of Child Psychology and Psychiatry, 47 (3-4), pp.296-312.

Blakemore, S.J., den Ouden, H., Choudhury, S. and Frith, C. (2007) Adolescent development of the neural circuitry for thinking about intentions. Social Cognitive and Affective Neuroscience, 2 (2), pp.130-139.

Blakemore, S.-J. (2008) The social brain in adolescence. Nature Reviews Neuroscience, 9 (4), pp.267-277.

Bourgeois, J.P., Goldman-Rakic, P.S. and Rakic, P. (1994) Synaptogenesis in the prefrontal cortex of rhesus monkeys. Cerebral Cortex, 4 (1), pp.78-96.

Burnett, S., Bird, G., Moll, J., Frith, C. and Blakemore, S.-J. Development during adolescence of the neural processing of social emotion. Journal of Cognitive Neuroscience, in press, 2008.

Cairns, R.B., Cairns, B.D., Neckerman, H.J. and Ferguson, L.L. (1989) Growth and aggression: 1. Childhood to early adolescence. Developmental Psychology, 25, pp.320-330.

Choudhury, S., Blakemore, S.J. and Charman, T. (2006) Social cognitive development during adolescence. Social Cognitive and Affective Neuroscience, 1 (3), pp.165-174.

Durston, S., Hulshoff, P., Casey, B.J., Giedd, J.N., Buitelaar, J.K. and van Engeland, H. (2001) Anatomical MRI of the Developing Human Brain: What Have We Learned? Journal of the American Academy of Child and Adolescent Psychiatry, 40 (9), pp.1012-1020

Elkind, D. (1967) Egocentrism in adolescence. Child Development, 38, pp.1025-34.

Erath, S.A., Flanagan, K.S. and Bierman, K.L. (2007) Social anxiety and peer relations in early adolescence: behavioural and cognitive factors. Journal of Abnormal Child Psychology, 35, pp.405-416.

Evenson, R.C., Meier, S.T. and Hagan, B.J. (1993) Sex differences in the age of onset of affective disorders. Comparative Psychiatry, 34 (3), pp.187-191.

Gardner, M. and Steinberg, L. (2005) Peer influence on risk taking, risk preference, and risky decision making in adolescence and adulthood: an experimental study. Developmental Psychology, 41 (4), pp.625-635.

Giedd, J.N., Blumenthal, J., Jeffries, N.O., Castellanos, F.X., Liu, H., Zijdenbos, A., Paus, T., Evans, A.C. and Rapoport, J.L. (1999) Brain development during childhood and adolescence: a longitudinal MRI study. Nature Neuroscience, 2 (10), pp.861-863.

Giedd, J.N., Castellanos, F.X., Rajapakse, J.C., Vaituzis, A.C. and Rapoport, J.L. (1997) Sexual dimorphism of the developing human brain. Progress in Neuropsychopharmacological and Biological Psychiatry, 21 (8), pp.1185-1201.

Giedd, J.N., Snell, J.W., Lange, N., Rajapakse, J.C., Casey, B.J., Kozuch, P.L., Vaituzis, A.C., Vauss, Y.C., Hamburger. S.D., Kaysen, D. and Rapoport, J.L. (1996) Quantitative magnetic resonance imaging of human brain development: ages 4-18. Cerebral Cortex, 6 (4), pp.551-560.

Gogtay, N., Giedd, J.N., Lusk, L., Hayashi, K.M., Greenstein, D., Vaituzis, A.C., Nugent, T.F.3rd, Herman, D.H., Clasen, L.S., Toga, A.W., Rapoport, J.L. and Thompson, P.M. (2004) Dynamic mapping of human cortical development during childhood through early adulthood. Proceedings of the National Academy of Science, USA, 101 (21), pp.8174-8179.

Goswami, U. (2006) Neuroscience and education: from research to practice? Nature Reviews Neuroscience, 7 (5), pp.406-411.

Greene, J. and Cohen, J. (2004) For the law, neuroscience changes nothing and everything. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 359, pp.1775-1785.

Hubel, D.H. and Wiesel, T.N.(1964) Effects of monocular deprivation in kittens. Naunyn Schmiedebergs Arch Exp Pathol Pharmakol., 248, pp.492-497.

Huttenlocher, P.R. and Dabholkar, A.S. (1997). Regional differences in synaptogenesis in human cerebral cortex. Journal of Comparative Neurology, 387 (2), pp.167-178.

Huttenlocher, P.R. (1979). Synaptic density in human frontal cortex – developmental changes and effects of ageing. Brain Research, 163, pp.195-205.

Huttenlocher, P.R., De Courten, C., Garey, L.J., and Van Der Loos, H. (1983). Synaptic development in human cerebral cortex. International Journal of Neurology, 16-17, pp.144- 154.

Kendler, K.S., Thornton, L.M. and Prescott, C.A. (2001) Gender differences in the rates of exposure to stressful life events and sensitivity to their depressogenic effects. American Journal of Psychiatry, 158 (4), pp.587-593.

Kovacs, M., Obrosky, D.S. and Sherrill, J. (2003) Developmental changes in the phenomenology of depression in girls compared to boys from childhood onward. Journal of Affect Disorders, 4 (1), pp.33-48.

Moffitt, T.E., Caspi, A., Rutter, M. and Silva, P.A. (2001) Sex Differences in Antisocial Behaviour: Conduct Disorder, Delinquency, and Violence in the Dunedin Longitudinal Study. New York, Cambridge University Press.

Parsons, J.T., Halkitis, P.N., Bimbi, D. and Borkowski, T. (2000) Perceptions of the benefits and costs associated with condom use and unprotected sex among late adolescent college students. Journal of Adolescence, 23 (4), pp.377-391.

Paus, T., Collins, D.L., Evans, A.C., Leonard, G., Pike, B. and Zijdenbos, A. (2001) Maturation of white matter in the human brain: a review of magnetic resonance studies. Brain Research Bulletin, 54 (3), pp.255-266.

Paus, T., Zijdenbos, A., Worsley, K., Collins, D.L., Blumenthal, J., Giedd, J.N., Rapoport, J.L. and Evans, A.C. (1999) Structural maturation of neural pathways in children and adolescents: in vivo study. Science, 283, pp.1908-1911.

Pope, H.G.Jr, Gruber, A.J., Hudson, J.I., Cohane, G., Huestis, M.A. and Yurgelun-Todd, D. (2003) Early-onset cannabis use and cognitive deficits: what is the nature of the association? Drug and Alcohol Dependence, 69 (3), pp.303-310.

Rakic, P., Bourgeois, J.P., Eckenhoff, M.F., Zecevic, N. and Goldman-Rakic, P.S. (1986) Concurrent overproduction of synapses in diverse regions of the primate cerebral cortex. Science, 232, pp.232-235.

Roussos, A., Francis, K., Zoubou, V., Kiprianos, S., Prokopiou, A. and Richardson, C. (2001) The standardization of Achenbach’s Youth Self-Report in Greece in a national sample of high school students. European Child and Adolescent Psychiatry, 10 (1), pp.47-53.

Silveri, M.M., Rohan, M.L., Pimentel, P.J., Gruber, S.A., Rosso, I.M. and Yurgelun-Todd, D.A. (2006) Sex differences in the relationship between white matter microstructure and impulsivity in adolescents. Magnetic Resonance Imaging, 24 (7), pp.833-841.

Sowell, E.R., Thompson, P.M., Holmes, C.J., Jernigan, T.L. and Toga, A.W. (1999) In vivo evidence for post-adolescent brain maturation in frontal and striatal regions. Nature Neuroscience, 2 (10), pp.859-861.

Steinberg, L. and Monahan, K.C. (2007) Age differences in resistance to peer influence. Developmental Psychology, 43 (6), pp.1531-1543.

Steinberg, L. and Silverberg, S.B. (1986) The vicissitudes of autonomy in early adolescence. Child Development, 57 (4), pp.841-851.

Stevens, M.C., Kiehl, K.A., Pearlson, G.D. and Calhoun, V.D. (2007) Functional neural networks underlying response inhibition in adolescents and adults. Behavioural Brain Research, 181 (1), pp.12-22.

UNICEF (2007) An overview of child well-being in rich countries. Unicef Innocenti Research Centre.

Vartanian, L.R. (2000) Revisiting the imaginary audience and personal fable constructs of adolescent egocentrism: A conceptual review. Adolescence, 35, pp.639-661.

Wang, A.T., Lee, S.S., Sigman, M. and Dapretto, M. (2006). Neural basis of irony comprehension in children with autism: the role of prosody and context. Brain, 129 (4), pp.932-943.

Yakovlev, P.I. and Lecours, A.R. (1967) The myelogenetic cycles of regional maturation of the brain. In: Minkowski, A. ed. Regional Development of the Brain in Early Life. Oxford, Blackwell Scientific, pp.3-70

Ybrandt, H. (2008) The relation between self-concept and social functioning in adolescence. Journal of Adolescence, 31 (1), pp.1-16

Yurgelun-Todd, D.A., Killgore, W.D. and Young, A.D. (2002) Sex differences in cerebral tissue volume and cognitive performance during adolescence. Psychol Rep, 91 (3,1), pp.743-757.

Zimring, F.E. (1998) American Youth Violence. Oxford, Oxford University Press.

This document has been commissioned as part of the UK Department for Children, Schools and Families’ Beyond Current Horizons project, led by Futurelab. The views expressed do not represent the policy of any Government or organisation.