Educational neuroscience

Human language is a unique faculty of the mind[32] and the ability to understand and produce oral and written language is fundamental to academic achievement and attainments.[33] Children who experience difficulties with oral language raise significant challenges for educational policy and practice;[34] National Strategies, Every Child a Talker (2008). The difficulties are likely to persist during the primary school years[35] where, in addition to core deficits with oral language, children experience problems with literacy,[36] numeracy[37] and behaviour and peer relations.[38] Early identification and intervention to address these difficulties, as well as identification of the ways in which learning environments can support atypical language development are essential.[34] Untreated speech and language needs result in significant costs both to the individual and to the national economy (ICAN, 2006).

Over the last decade, there has been a significant increase in neuroscience research examining young children's processing of language at the phonetic, word, and sentence levels.[39] There are clear indications that neural substrates for all levels of language can be identified at early points in development. At the same time, intervention studies have demonstrated the ways in which the brain retains its plasticity for language processing. Intense remediation with an auditory language processing program has been accompanied by functional changes in left temporo-parietal cortex and inferior frontal gyrus.[29] However, the extent to which these results generalize to spoken and written language is debated.[40]

The relationships between meeting the educational needs of children with language difficulties and the findings of neuroscience studies are not yet established. One concrete avenue for progress is to use neuroscientific methods to address questions that are significant to practice in learning environments. For example, the extent to which language skills are attributable to a single common trait, and the consistency of such a trait over development, are matters of debate.[41] However, direct assessments of brain activity can inform these debates.[42] A detailed understanding of the sub-components of the language system, and the ways these change over time may inevitably yield implications for educational practice.

Mathematical skills are important not only for the national economy but also for an individual's life chances: low numeracy increases the probability of arrest, depression, physical illnesses, unemployment.[43] One of the main causes of low numeracy is a congenital condition called dyscalculia. As the Foresight report on Mental Capital and Wellbeing puts it, "Developmental dyscalculia – because of its low profile but high impacts, its priority should be raised. Dyscalculia relates to numeracy and affects between 4–7% of children. It has a much lower profile than dyslexia but can also have substantial impacts: it can reduce lifetime earnings by £114,000 and reduce the probability of achieving five or more GCSEs (A*-C) by 7–20 percentage points. Home and school interventions have again been identified by the Project. Also, technological interventions are extremely promising, offering individualised instruction and help, although these need more development." (Executive Summary, Section 5.3) Understanding typical and atypical mathematical development is a crucial underpinning for the design of both the mainstream mathematics curriculum and for helping those who fail to keep up.[44] Over the past ten years, a brain system for simple number processing has been identified[45][46] and a handful of studies of children's brains that shed light on its development.[9]

An increasing convergence of evidence suggests that dyscalculia may be due to a deficit in an inherited core system for representing the number of objects in a set, and how operations on sets affect number[47] and in the neural systems that support these abilities.[9] This core deficit affects the learner's ability to enumerate sets and to order sets by magnitude, which in turn make it very difficult to understand arithmetic, and very hard to provide a meaningful structure for arithmetical facts. Twin[48] and family[49] studies suggest that dyscalculia is highly heritable, and genetic anomalies, such as Turner's Syndrome, indicate an important role for genes in the X chromosome.[50]

This suggestion that dyscalculia is caused by a deficits in a core deficit in number sense is analogous to the theory that dyslexia is due to a core deficit in phonological processing. Despite these similarities in terms of the scientific progress, public awareness of dyscalculia is much lower than it is for dyslexia. The UK's Chief Scientific Advisor, John Beddington, notes that, "developmental dyscalculia is currently the poor relation of dyslexia, with a much lower public profile. But the consequences of dyscalculia are at least as severe as those for dyslexia."[51]

The application of neuroscience to understanding mathematical processing has already resulted in understanding beyond the early cognitive theories. Cognitive neuroscience research has revealed the existence of an innate ‘number sense’ system, present in animals and infants as well as adults, that is responsible for basic knowledge about numbers and their relations. This system is located in the parietal lobe of the brain in each hemisphere.[45][52] This parietal system is active in children and adults during basic numerical tasks,[53][54] but over the course of development it appears to become more specialised. Furthermore, children with mathematical learning disabilities (dyscalculia) show weaker activation in this region than typically developing children during basic number tasks.[9] These results show how neuroimaging can provide important information about the links between basic cognitive functions and higher level learning, such as those between comparing two numbers and learning arithmetic.

In addition to this basic number sense, numerical information can be stored verbally in the language system, a system that neuroscience research is beginning to reveal as qualitatively different at the brain level to the number sense system.[55] This system also stores information about other well learned verbal sequences, such as days of the week, months of the year and even poetry, and for numerical processing it supports counting and the learning of multiplication tables. While many arithmetic problems are so over learned that they are stored as verbal facts, other more complex problems require some form of visuo-spatial mental imagery.[56] Showing that these subsets of arithmetic skills are supported by different brain mechanisms offers the opportunity for a deeper understanding of the learning processes required to acquire arithmetic proficiency.

Neuroimaging studies of mathematical learning disabilities are still rare but dyscalculia is an area of increasing interest for neuroscience researchers. Since different neural mechanisms contribute to different elements of mathematical performance, it may be that children with dyscalculia show variable patterns of abnormality at the brain level. For example, many children with dyscalculia also have dyslexia, and those that do may show different activation of the verbal networks that support maths, while those who have dyscalculia only, may show impairments of the parietal number sense system. Indeed, the few studies carried out on children with dyscalculia only point to a brain level impairment of the number sense system.[9][57]

Such evidence is beginning to contribute to a theoretical debate between researchers who believe that dyscalculia is caused by a brain level deficit of the number sense and those who believe that the disorder stems from a problem in using numerical symbols to access the number sense information. With the continued development of theoretical models of dyscalculia that generate explicit testable hypotheses, progress should be rapid in developing research which investigates the link between mathematical learning disorders and their neural correlates.[22]

In the last 10 years, there has been an explosion of interest in the role of emotional abilities and characteristics in contributing to success in all aspects of life. The concept of Emotional Intelligence (EI)[58] has gained wide recognition and is featured in the Foresight report on Mental Capital and Wellbeing. Some have made influential claims that EI is more important than conventional cognitive intelligence, and that it can more easily be enhanced.[59] Systematic research has yet to provide much support for these claims, although EI has been found to be associated with academic success[4][60] and there is some evidence that it may be of particular importance for groups at-risk of academic failure and social exclusion. In spite of the weak evidence base, there has been a focus on promoting the social and emotional competence, mental health and psychological wellbeing of children and young people,[61] particularly in schools as the result of the investment in universal services, prevention and early intervention (e.g., the Social and Emotional Aspects of Learning (SEAL) project in the UK [DfES, 2005, 2007]).

The neural basis of emotional recognition in typically developing children[62] has been investigated, although there is little neuroimaging work on atypically developing children who process emotions differently.[4] Males are commonly over-represented in these atypically developing populations and a female advantage is commonly reported both on EI measures and on most areas of emotion processing. In processing facial expressions the female advantage appears best explained by an integrated account considering both brain maturation and social interaction.[63]

Prefrontal brain damage in children affects social behavior, causing insensitivity to social acceptance, approval or rejection.[64] These brain areas process social emotions such as embarrassment, compassion and envy. Moreover, such damage impairs cognitive as well as social decision making in real world contexts[57][65] supporting the Vygotskian view that social and cultural factors are important in cognitive learning and decision making. This view emphasizes the importance of bringing together neuroscientific and social constructionist perspectives, in this case in examining the influence of emotion on transferable learning.[66]

However, there are currently many gaps in the attempt to bring together developmental science and neuroscience to produce a more complete understanding of the development of awareness and empathy.[67] Educational research relies on pupil's accurate self-report of emotion, which may not be possible for some pupils, e.g., those with alexithymia—a difficulty in identifying and describing feelings, which is found in 10% of typical adults. Emotional awareness can be measured using neuroimaging methods[68] that show that differing levels of emotional awareness are associated with differential activity in amygdala, anterior insular cortex, and the medial prefrontal cortex. Studies of brain development in childhood and adolescence show that these areas undergo large-scale structural changes.[69] Hence, the degree to which school-age children and young adults are aware of their emotions may vary across this time period, which may have an important impact on classroom behaviour and the extent to which certain teaching styles and curriculum approaches might be effective.

Neuroimaging work is also beginning to help in the understanding of social conduct disorders in children. For example, callous-unemotional traits in children are a particularly difficult problem for teachers to deal with, and represent a particularly serious form of conduct disturbance. Jones et al. (2009)[70] showed that children with callous-unemotional traits revealed less brain activation in the right amygdala in response to fearful faces, suggesting that the neural correlates of that type of emotional disturbance are present early in development.

Researchers from the Centre for Educational Neuroscience in London have been instrumental in developing a research body that investigates how social cognition develops in the brain. In particular, Sarah-Jayne Blakemore, co-author of "The Learning Brain", has published influential research on brain development related to social cognition during adolescence. Her research, suggests that activity in brain regions associated with emotional processing undergo significant functional changes during adolescence.[71]

Attention refers to the brain mechanisms that allow us to focus on particular aspects of the sensory environment to the relative exclusion of others. Attention modulates sensory processing in "top-down" fashion. Maintaining selective attention toward a particular item or person for a prolonged period is clearly a critical underpinning skill for the classroom. Attention is the key cognitive skill impaired in ADHD resulting in difficulty in completing tasks or attending to details.[72] Aspects of attention may also be atypical in children showing anti-social behaviour and conduct disorders. From the perspective of basic neuroscience, recent evidence suggests that attention skills may be one of the human brain functions that respond best to early intervention and training (e.g.[73]).

Further, from a neuroconstructivist perspective attention is a vital mechanism through which the child can actively select particular aspects of their environment for further learning. Executive functions include the abilities to inhibit unwanted information or responses, to plan ahead for a sequence of mental steps or actions, and to retain task-relevant and changing information for brief periods (working memory).[74] Like attention, executive function abilities provide a critical platform for the acquisition of domain-specific knowledge and skills in an educational context. Further, recent studies show that preschool training of executive skills may prevent early school failure.[75][76] Children with ADHD, anti-social behaviour, conduct disorders and autism can all show atypical patterns of executive function. Basic neuroscience studies have identified the primary brain structures and circuits involved in executive functions, including the prefrontal cortex, in adults. However, much research remains to be done to understand the development of this circuitry, and the genetic and neural bases of individual differences in executive function.[77] Foresight Mental Capital and Wellbeing Project specifically identifies and highlights the importance of attention and executive function skills in the future challenges for difficulties in learning (sections 2.2.4 and 2.4 in "Learning Difficulties: Future Challenges").

Despite Willingham's assertion[22] that the potential for neuroscience to contribute to educational practice and theory is already beyond doubt, he highlights three challenges that must be overcome to marry the two disciplines effectively.

The Goals Problem: Willingham suggests that education is a so-called "artificial science" that seeks to construct an ‘artifact’, in this case a set of pedagogic strategies and materials. Neuroscience, on the other hand is a so-called "natural science", concerned with the discovery of natural principles that describe neural structure and function. This difference means that some goals set by education are simply impossible to answer using neuroscience research, for example, the building of character or aesthetic sense in children.

The Vertical Problem: Levels of analysis: Willingham suggests that the highest level of analysis employed by neuroscientists is the mapping of brain structure and activity onto cognitive function, or even the interaction of cognitive functions (i.e. the impact of emotion on learning). Within neuroscience research these functions are studied in isolation for the sake of simplicity, and the nervous system as a whole, functioning in its entirety with all its huge composition of functional interactions, is not considered. For educators, on the other hand, the lowest level of analysis would be the mind of a single child, with levels increasing to incorporate the classroom, neighborhood, country etc.

Thus, importing research about a single cognitive factor in isolation, into a field in which context is essentially important creates an inherent difficulty. For example, while rote learning may be shown to improve learning in the research laboratory, the teacher cannot implement that strategy without considering the impact on the child's motivation. In return, it is difficult for neuroscientists to characterize such interactions in a research setting.

The Horizontal Problem: Translating research findings: While education theory and data are almost exclusively behavioral, findings from neuroscience research can take on many forms (e.g. electrical, chemical, spatial, temporal etc.). The most common form of data taken from neuroscience to education is the spatial mapping of brain activation to cognitive function. Willingham (2009) highlights the difficulty in applying such spatial information to educational theory. If a certain brain region is known to support a cognitive function relevant for education, what can actually be done with that information? Willingham suggests that this ‘horizontal problem’ can be solved only when a rich body of behavioral data and theories already exist,[80] and points out that such methods have already been successful in identifying subtypes of dyslexia (e.g.[81][82]).

Willingham suggests that what is essential for a successful union of neuroscience and education is that both fields have realistic expectations of one another. For example, educators should not expect that neuroscience will provide prescriptive answers for educational practice, answers for educational goals that are incompatible with neuroscientific methods (e.g. aesthetic training), or levels of analysis beyond the individual level. Finally Willingham suggests that neuroscience will only be useful to educators when targeted at a specific problem at a fine grained level of analysis, such as how people read, but that these data will only be useful in the context of well developed behavioral theories.

Other researchers, such as Katzir & Pareblagoev[30] have pointed out that neuroimaging methodology as it stands may not be suitable for the examination of higher level cognitive functions, because it relies primarily on the ‘subtraction method’. By this method, brain activity during a simple control task is subtracted from that of a ‘higher order’ cognitive task, thus leaving the activation that is related specifically to the function of interest. Katzir & Pareblagoev suggest that while this method may be very good for examining low level processing, such as perception, vision and touch, it is very hard to design an effective control task for higher order processing, such as comprehension in reading and inference making. Thus, some researchers[83][84] argue that functional imaging technologies may not be best suited for the measurement of higher order processing. Katzir & Pareblagoev, suggest that this may not be a deficit of the technology itself, but rather of the design of experiments and the ability to interpret the results. The authors advocate using experimental measures in the scanner for which the behavioural data is already well understood, and for which there exists a strong theoretical framework.

Another recent review of the educational neuroscience debate by Varma, McCandliss and Schwartz[85] focuses on eight primary challenges, divided into scientific challenges and practical challenges, facing the field, and attempts to transform those challenges into opportunities.

Methods: Neuroscience methods create artificial environments and thus cannot provide useful information about classroom contexts. Furthermore, the concern is that if neuroscience begins to influence educational practice too heavily, there may be a de-emphasis of contextual variables, and solutions to educational problems may become primarily biological rather than instructional. However, Varma et al. argue that novel experimental paradigms create the opportunity to investigate context, such as brain activation following different learning procedures[86] and that neuroimaging can also allow for the examination of strategic/mechanistic developmental changes that cannot be tapped by reaction time and behavioural measures alone. Furthermore, Varma et al. cite recent research that shows that the effects of cultural variables can be investigated using brain imaging (e.g.[87]), and the results used to draw implications for classroom practice.

Data: Knowing the brain region that supports an elementary cognitive function tells us nothing about how to design instruction for that function. However, Varma et al. suggest that neuroscience provide the opportunity for a novel analyses of cognition, breaking down behaviour into elements invisible at the behavioural level. For example, the question of whether different arithmetic operations show different speed and accuracy profiles is the result of different efficiency levels within one cognitive system versus the use of different cognitive systems.

Reductionist Theories: Applying neuroscience terminology and theory to educational practice is a reduction and is of no practical use to educators. Nothing is gained be redescribing a behavioural deficit in neuroscientific terms. Varma et al. point out that reductionism is a mode by which sciences are unified, and that the co-opting of neuroscience terminology does not necessitate the elimination of education terminology, it simply provides the opportunity for interdisciplinary communication and understanding.

Philosophy: Education and neuroscience are fundamentally incompatible, because attempting to describe behavioural phenomena in the classroom by describing physical mechanisms of the individual brain is logically wrong. However, neuroscience may help to resolve internal conflicts within education resulting from differing theoretical constructs and terminologies used within subfields of education by providing a measure of uniformity with regards to results reporting.

Costs: Neuroscience methods are highly expensive, and the expected outcomes do not justify the costs. However, Varma et al. point out that educationally relevant neuroscience may attract addition funding to education research rather than usurping resources. The essential claim of educational neuroscience is that the two fields are interdependent and that a portion of the funding allocated collectively to the two fields should be directed towards shared questions.

Timing: Neuroscience, while expanding rapidly, is still in relative infancy with regards to the non-invasive study of healthy brains, and thus education researchers should wait until more data is collected and distilled into succinct theories. Contrary to this, Varma et al. argue that some success is already evident. For example, studies examining the success of dyslexia remediation programmes[88] have been able to reveal the impact of these programmes on the brain networks supporting reading. This in turn leads to the generation of new research questions.

Control: If education allows neuroscience in the door, theories will increasingly be cast in terms of neural mechanisms and debates will rely increasingly on neuroimaging data. Neuroscience will cannibalise resources and education research will lose its independence. Varma et al. argue that the assumption of an asymmetric relationship between the two fields is unnecessary. Education has the potential to influence neuroscience, directing future research into complex forms of cognition and education researchers can help Educational Neuroscience to avoid naïve experiments and repetition of earlier mistakes.

Neuromyths: Thus far most of the neuroscience findings applied to education have turned out to be neuromyths, irresponsible extrapolations of basic research to education questions. Furthermore, such neuromyths have escaped beyond academia and are being marketed directly to teachers, administrators and the public. Varma et al. respond that the existence of neuromyths reveals a popular fascination with brain function. Appropriate translation of educational neuroscience results and well established collaborative research can decrease the likelihood of neuromyths.

Researchers such as Katzir & Pareblagoev[30] and Cacioppo & Berntson (1992)[89] argue that as well as neuroscience informing education, the educational research approach can contribute to the development of new experimental paradigms in neuroscience research. Katzir and Pareblagoev (2006) suggest the example of dyslexia research as a model of how this bidirectional collaboration might be achieved. In this case, theories of reading processes have guided both the design and interpretation of neuroscience research, but the existing theories were developed primarily from behavioural work. The authors suggest that the establishment of theories, which delineate required skills and subskills for educationally relevant tasks, is an essential requirement for educational neuroscience research to be productive. Furthermore, such theories need to suggest empirically testable connections between educationally relevant behaviours and brain function.

Kurt Fischer, director of Harvard University's Mind, Brain and Education graduate program states "One of the reasons there is so much junk out there is that there are so few people who know enough about education and neuroscience to put the thing together".[90] Educators have been reliant upon others’ expertise for the interpretations from Neuroscience hence have not been able to discern whether the claims made are valid or invalid representations of the research. Without a direct access to the primary research educators may be at risk of misusing results from neuroscience research.[91] The need for so called ‘middlemen’ in the translation of research to practice has led to a situation where the application of cognitive neuroscience research findings is running ahead of the research itself.

In order to negate the need for middlemen, some researchers have suggested the need to developed a group of neuro-educators, a specially trained class of professionals whose role would be to guide the introduction of cognitive neuroscience into educational practice in a sensible and ethical manner. Neuro-educators would play a pivotal role in assessing the quality of evidence purporting to be relevant to education, assessing who is best placed to employ newly developed knowledge, and with what safeguards, and how to deal with unexpected consequences of implemented research findings.[92]

Byrnes & Fox (1998)[93] have suggested that developmental psychologists, educational psychologists and teachers generally fall into one of four orientations with respect to neuroscientific research "(1) those who readily accept (and sometimes over interpret) the results of neuroscientific studies; (2) those who completely reject the neuroscientific approach and consider the results of neuroscientific studies meaningless; (3) those who are unfamiliar with and indifferent toward, neuroscientific research; and (4) those who cautiously accept neuroscientific findings as being a proactive part of the total pattern of findings that have emerged from different corners of the cognitive and neural sciences". Greenwood (2009)^[85] suggests that as the body of knowledge available to educators increases, and the ability to be expert in all areas diminishes, the most productive standpoint would the fourth outlined by,^[87] that of cautious acceptance of neuroscientific findings and proactive collaboration.

Bennett & Rolheiser-Bennett (2001)[94] point out that "teachers must be aware of and act on the science within the art of teaching". They suggest that educators must become aware of other methods and incorporate them into their practice. Furthermore, Bennett and Rolheiser-Bennett suggest that specific bodies of knowledge will play an important role in informing educators when making important decisions with regards to the "design of learning environments". The bodies of knowledge discussed include multiple intelligences, emotional intelligences, learning styles, the human brain, children at risk and gender. As the authors explain these and other areas are just "lenses designed to extend teachers’ understanding of how students learn, and from that understanding, to make decisions about how and when to select, integrate, and enact items in the ... list".^[88]

Mason[14] supports calls for a two-way constructive collaboration between neuroscience and education, whereby, rather than neuroscience research simply being applied to education, findings from neuroscience research would be used to constrain educational theorizing. In return, education would influence the types of research questions and experimental paradigms used in neuroscience research. Mason also gives the example that while pedagogical practice in the classroom may give rise to educational questions regarding the emotional bases of performance on school tasks, neuroscience has the potential to reveal the brain basis of higher-order thinking processes and thus may help to understand the role that emotion plays in learning and open new areas of study of emotional thought in the classroom.

The idea that the two hemispheres of the brain may learn differently has virtually no grounding in neuroscience research.[4] The idea has arisen from the knowledge that some cognitive skills appear differentially localised to a specific hemisphere (e.g., language functions are typically supported by left hemisphere brain regions in healthy right handed people). However, massive amount of fibre connections link the two hemispheres of the brain in neurologically healthy individuals. Every cognitive skill that has been investigated using neuroimaging to date employs a network of brain regions spread across both cerebral hemispheres, including language and reading, and thus no evidence exists for any type of learning that is specific to one side of the brain.

A critical period is a timeframe during the early life of an animal during which the development of some property or skill is rapid and is most susceptible to alteration. During a critical period, a skill or characteristic is most readily acquired. During this time, the plasticity is most dependent on experiences or environmental influences. Two examples of a critical period are the development of binocular vision and linguistic skills in children. The critical periods neuromyth is an overextension of certain neuroscience research findings (see above) primarily from research into the visual system, rather than cognition and learning. Although sensory deprivation during certain time periods can clearly impede the development of visual skills, these periods are sensitive rather than critical, and the opportunity for learning is not necessarily lost forever, as the term "critical" implies. While children may benefit from certain types of environmental input, for example, being taught a second language during the sensitive period for language acquisition, this does not mean that adults are unable to acquire foreign language skills later in life.

The idea of critical periods comes primarily from the work of Hubel and Wiesel.[96] Critical periods generally coincide with periods of excess synapse formation, and end at around the same time that synaptic levels stabilise. During these periods of synaptic formation, some brain regions are particularly sensitive to the presence or absence of certain general types of stimuli. There are different critical periods within specific systems, e.g. visual system has different critical periods for ocular dominance, visual acuity and binocular function[97] as well as different critical periods between systems, for example, the critical period for the visual system appears to end around the age of 12 years, while that for acquiring syntax ends around 16 years.

Rather than talking of a single critical period for general cognitive systems, neuroscientists now perceive sensitive periods of time during which the brain is most able to be shaped in a subtle and gradual fashion. Furthermore, critical periods themselves may be divided into three phases. The first, rapid change, followed by continued development with the potential for loss or deterioration, and finally a phase of continued development during which the system can recover from deprivation.

Although there is evidence for sensitive periods, we do not know whether they exist for culturally transmitted knowledge systems such as educational domains like reading and arithmetic. Further, we do not know what role synaptogenesis plays in the acquisition of these skills.

The enriched environment argument is based on evidence that rats raised in complex environments perform better on maze tasks and have 20–25% more synaptic connections than those raised in austere environments.[98] However, these enriched environments were in laboratory cages, and did not come close to replicating the intensely stimulating environment a rat would experience in the wild. Furthermore, the formation of these additional connections in response to novel environmental stimuli occurs throughout life, not just during a critical or sensitive period. For example, skilled pianists show enlarged representations in the auditory cortex relating specifically to piano tones,[99] while violinists have enlarged neural representations for their left fingers.[100] Even London taxi drivers who learn the London street map in intense detail develop enlarged formations in the part of the brain responsible for spatial representation and navigation.[101] These results show that the brain can form extensive new connections as the result of focused educational input, even when this input is received solely during adulthood. Greenough's work suggests a second type of brain plasticity. Whereas synaptogenesis and critical periods relate to experience-expectant plasticity, synaptic growth in complex environments relates to "experience-dependent" plasticity. This type of plasticity is concerned with environment specific learning, and not to features of the environment that are ubiquitous and common to all members of the species, such as vocabulary.

Experience dependent plasticity is important because it does potentially link specific learning and brain plasticity, but it is relevant throughout the lifetime, not just in critical periods. "Experience-expectant plasticity",[98] suggests that the environmental features necessary for fine tuning sensory systems are ubiquitous and of a very general nature. These kinds of stimuli are abundant in any typical child's environment. Thus, experience-expectant plasticity does not depend on specific experiences within a specific environment, and therefore cannot provide much guidance in choosing toys, preschools, or early childcare policies. The link between experience and brain plasticity is intriguing. No doubt learning affects the brain, but this relationship does not offer guidance on how we should design instruction.

Bruer also warns of the dangers of enriching environments on the basis of socio-economic value systems, and warns of a tendency to value typically middle class pursuits as more enriching than those associated with a working class lifestyle, when there is no neuroscientific justification for this.

In addition some critics of the Educational Neuroscience approach have highlighted limitations in applying the understanding of early physiological brain development, in particular synaptogenesis to educational theory.

Synaptogenesis research has primarily been carried out on animals (e.g. monkeys and cats). Measures of synaptic density are aggregate measures, and it is known that different types of neuron within the same brain region differ in their synaptic growth rates [70]. Secondly, the purported "critical period" of birth to three years is derived from research on rhesus monkeys, who reach puberty at the age of three, and assumes that the period of synaptogenesis in humans exactly mirrors that of monkeys. It may be more reasonable to assume that this period of neural growth actually lasts until puberty, which would mean until early teenage years in humans.

Periods of intense synaptogenesis are typically correlated with the emergence of certain skills and cognitive functions, such as visual fixation, grasping, symbol use and working memory. However, these skills continue to develop well after the period that synaptogenesis is thought to end. Many of these skills continue to improve even after synaptic density reaches adult levels, and thus the most we can say is that synaptogenesis may be necessary for the emergence of these skills, but it cannot account entirely for their continued refinement.[102] Some other form of brain change must contribute to ongoing learning.

Additionally, the types of cognitive changes usually seen to correlate with synaptogenesis revolve around visual, tactile, movement and working memory. These are not taught skills but rather skills that are usually acquired independent of schooling, even though they may support future learning. How these skills relate to later school learning is, however, unclear. We know that synaptogenesis occurs, and that the pattern of synaptogenesis is important for normal brain function. However, what is lacking is the ability of neuroscience to tell educators what sort of early childhood experiences might enhance children's cognitive capacities or educational outcomes.

The idea that a person can have a "male" brain or "female" brain is a misinterpretation of terms used to describe cognitive styles by[103] when attempting to conceptualise the nature of cognitive patterns in people with autism spectrum disorder. Baron-Cohen suggested that while men were better "systemisers" (good at understanding mechanical systems), women were better "empathisers" (good at communicating and understanding others), therefore he suggested that autism could be thought of as an extreme form of the "male brain". There was no suggestion that males and females had radically different brains or that females with autism had a male brain.

A common myth in the field of education is that individuals have different learning styles, such as 'visual' or 'kinesthetic'. Many individuals will state preferences for the way in which they want to learn, but there is no evidence that matching a teaching technique to a preferred style will improve learning, despite this hypothesis being tested multiple times.[104][105] There may even be harms associated with the use of learning styles, wherein learners become 'pigeonholed', perceiving that they may not be suited to types of learning that are not matched to their 'learning style'[106] (e.g. so-called visual learners may not wish to learn music). Despite this lack of evidence, a 2012 study demonstrated that a belief in the use of learning styles is widespread amongst teachers,[107] and a 2015 study showed that the majority of research papers in higher education research mistakenly endorse the use of learning styles.[106]

• OECD: Organisation for Economic Co-operation and Development

• BERA – British Educational Research Association Special Interest Group (SIG) on Neuroscience & Education
• EARLI Special Interest Group (SIG) on Neuroscience & Education
• Brain, Neuroscience and Education (a special interest group of the American Educational Research Association)
• International Mind, Brain, and Education Society
• The Jean Piaget Society
• Learning & the Brain Conference
• The London School – Neuroeducation Centre
• Oxford Cognitive Neuroscience – Education Forum

• npj Science of Learning
• Mind, Brain & Education Journal
• Neuroeducation (Journal)