Saturday 25 January 2014

What is educational neuroscience?

©CartoonStock.com

As someone who works at the interface of child development and neuroscience, I've been struck by the relentless rise of the sub-discipline of 'educational neuroscience'. New imaging technologies have led to a burgeoning of knowledge about the developing brain, and it is natural to want to apply this knowledge to improving children's learning. Centres for educational neuroscience have sprung up all over the place, with support from universities who see them as ticking two important boxes: interdisciplinarity and impact.

But at the heart of this enterprise, there seems to be a massive disconnect. Neuroscientists can tell you which brain regions are most involved in particular cognitive activities and how this changes with age or training. But these indicators of learning do not tell you how to achieve learning. Suppose I find out that the left angular gyrus becomes more active as children learn to read. What is a teacher supposed to do with that information?

As John Bruer pointed out back in 1997, the people who can be useful to teachers are psychologists. Psychological experiments can establish the cognitive underpinnings of skills such as reading, and can evaluate which are the most effective ways of teaching, and whether these differ from child to child. They can address questions such as whether there are optimal ages at which to teach different skills, how motivation and learning interact, and whether it is better to learn material in large chunks all at once or spaced out over intervals. At a trivial level, these could all be designated as aspects of 'educational neuroscience', insofar as the brain is necessarily involved in cognition and motivation. But they can all be studied without taking any measurements of brain function.

It is possible, of course, to look at the brain correlates of all of these things, but that's unlikely to influence what's done in the classroom. Suppose I want to see whether training in phonological awareness improves children's reading outcomes. I measure brain activation before and after training, and compare results with those of a control group who don't get the training. There are various possible patterns of results, as laid out in the table below:


As pointed out by Coltheart and McArthur (2012), what matters to the teacher is whether the training is effective in improving reading. It's really not going to make any difference whether detectable brain changes have happened, so either outcome A or B would give good justification for adopting the training, whereas outcomes C and D would not.

Well, you might say, children differ, and the brain measures might show up differences between those who do and don't respond to training. Indeed, but how would that be useful educationally? I've seen several studies that propose brain scans might be useful in identifying which children will and won't benefit from an intervention. That's a logical possibility, but given that brain scanning costs several hundred pounds per person, it's not realistic to suggest this has any utility in the real world, especially when there are likely to be behavioural indicators that predict outcomes just as well.

So are there actual or potential examples of how knowledge of neuroscience - as opposed to psychology - might influence educational practice? I mentioned three examples in this review: neurofeedback, neuropharmacology and brain stimulation are all methods that focus directly on changing the brain in ways that might potentially affect learning, and so could validly be designated as educational neuroscience. They are, however, as yet exploratory and experimental. The last of these, brain stimulation, was described this week in a blogpost by Roi Cohen Kadosh, who notes promising early results, but emphasizes that we need more experimental work establishing both risks and benefits before we could consider direct application of this method to improving children's learning.

I'm all in favour of cognitive neuroscience and basic research that discovers more about the neural underpinnings of typical and atypical development. By all means, let's do such studies, but let's do them because we want to find out more about the brain, and not pretend it has educational relevance.

If our goal is to develop better educational interventions, then we should be directing research funds into well-designed trials of cognitive and behavioural studies of learning, rather than fixating on neuroscience. Let me leave the last word to Hirsh-Pasek and Bruer, who described a Chilean conference in 2007 on Early Education and Human Brain Development. They noted: "The Chilean educators were looking to brain science for insights about which type of preschool would be the most effective, whether children are safe in child care, and how best to teach reading. The brain research presented at the conference that day was mute on these issues. However, cognitive and behavioral science could help."

References
Bishop, D. V. M. (2013). Neuroscientific studies of intervention for language impairment in children: interpretive and methodological problems Journal of Child Psychology and Psychiatry, 54 (3), 247-259 DOI: 10.1111/jcpp.12034

Bruer, J. T. (1997). Education and the brain: A bridge too far. Educational researcher, 26(8), 4-16. doi: 10.3102/0013189X026008004

Coltheart, M., & McArthur, G. (2012). Neuroscience, education and educational efficacy research. In M. Anderson & S. Della Sala (Eds.), Neuroscience in Education (pp. 215-221). Oxford: Oxford University Press.

This article (Figshare version) can be cited as: 
Bishop, Dorothy V M (2014): What is educational neuroscience?. figshare.
http://dx.doi.org/10.6084/m9.figshare.1030405

39 comments:

  1. great post - I made some similar arguments to a group of teachers at an "Educational Neuroscience" seminar a few months ago and their responses were very interesting - several agreed when I pointed out that many of the neuro studies being discussed offered little additional explanatory power or practical implications over an equivalent psychology study without the brain measures. However, many said they found the neuroscientific evidence/concepts far more compelling - and so did their students. One example a teacher gave was introducing secondary school students to the notion of "neuroplasticity" and using this to encourage them that revision was not hopeless! Of course everybody knows full well that people can learn - but apparently telling people their *brain* can learn holds some additional meaning.

    My speculation is that 'tangible' evidence, in this case in the form of physical changes in the brain, make neuroscientifc evidence/concepts more 'grounded', more accessible, and thus more compelling.

    Now does that justify the additional cost...

    ReplyDelete
  2. Thanks Tom. It is weird that 'neuroplasticity' is such a compelling meme. After all, if the brain didn't change, you wouldn't be able to teach anyone anything.
    For another statement of the obvious, take "brain based learning". Is there another kind? (well, strictly you could have spinal-cord based learning I suppose, but in this context 'brain' is just a redundant word).

    ReplyDelete
  3. My first thought was that this neuroplasticity phenomenon could be explained under a Carol Dweck-style "Growth Mindset" approach -- if a child understands that their brain can be changed, it's easier for them to develop an explanation of how their performance could change. A little googling revealed that she's actually involved in a company trying to do exactly that: http://www.mindsetworks.com/brainology/ . Carol is, of course, a developmental psychologist, not a neuroscientist. There would be a nice irony if the "neuroscience" part of educational neuroscience became a sort of window dressing, making it easier for us to smuggle psychological theories into education.

    ReplyDelete
  4. Very thought-provoking post! I would agree that the link between cognitive science and education is much more direct than that between neuroscience and education. In this way, I would like to contend that any efforts to connect what we are doing in cognitive neuroscience with education are better described by the name "Mind, Brain and Education" than they are by "Educational Neuroscience" since the former reflects the complementary roles of the cognitive, neuronal and educational levels of explanation while the latter perhaps suggests a unique role that is to be played by neuroscience in informing education. Put differently, by using the label “Mind, Brain and Education” to refer to efforts to bring evidence from cognitive neuroscience to bear on education and to build transdisciplinary bridges, no knowledge hierarchy is implied.

    I do think there is reason to consider (and be optimistic about) the potential relevance of neuroscientific evidence (always in combination with the cognitive level of explanation) for education beyond thinking only about tools to directly change brain activity through stimulation etc. Consider, for example, the work showing a.) that the combination of behavioural and neuroimaging (both functional and structural) data explain more of the variance in reading & math outcomes than either level of measurement does alone (http://www.ncbi.nlm.nih.gov/pubmed/17592952) and b.) that neuroimaging measures can do a better job at predicting who will improve in their reading and math skills (reading: http://www.ncbi.nlm.nih.gov/pubmed/21173250 & math: http://www.ncbi.nlm.nih.gov/pubmed/23630286) than behavioral measures alone. In addition to such data, I think there are contributions that neuroscience can make to addressing educational problems, such as the usefulness and neurobiological grounding of the discrepancy criteria to classify learning difficulties like developmental dyslexia (http://www.ncbi.nlm.nih.gov/pubmed/22006060). Here evidence from neuroscientific measurements can be use to address questions that have direct educational relevance.

    As you say “if the brain did not change, you wouldn’t be able to teach anybody anything”. Does it not follow from this that if we gain a better understanding of the mechanisms of how the brain changes as a function of educational experiences (the notion of biological embedding), we will be able to use such information to provide more refined, evidence-based experiences to drive neuroplasticity? Bigger questions regarding the timing of educational input and how this interacts with age, individual differences in learning and the role of other variables such as sleep, exercise and nutrition may be constrained by evidence from neuroscience.

    Finally, beyond the ‘laboratory to classroom’ model of thinking about the influence of either cognitive science or neuroscience, there are more indirect influences that both these fields might have on education. I am thinking here, for example, of Carol Dweck’s work showing that teaching adolescents about brain plasticity can help them to change their implicit theories of their own intelligence from a ‘fixed’ to a ‘growth’ mindset (ex. http://www.ncbi.nlm.nih.gov/pubmed/17328703).

    These are certainly early days for the potential role of neuroscience (in conjunction with other levels of explanation). But I do think that rather than thinking about either ‘cognitive’ or ‘neuronal’ it is useful to consider the (potentially superadditive) role of both (in the same way as cognitive neuroscientists do when they think about how the brain enables the mind). At the same time it is important to build collaborative bridges between education, cognitive science/psychology and neuroscience in an effort to begin asking the right questions, which will hopefully generate useable results.

    Perhaps the first role of neuroscience in education is to ensure that educators get the right information and that apparent facts about the brain and learning, which turn out to be myths, be quashed.

    ReplyDelete
    Replies
    1. Daniel : many thanks for taking the time to write such a detailed comment. I had hoped I might provoke some people to give me good examples of why I might be wrong (or at least overstating the case). But, while the papers you cite address interesting questions, I don't find them all that compelling.
      Let's take the work by Hoeft et al on using brain imaging to predict outcomes in reading and maths. I actually blogged about the later Hoeft et al PNAS paper back in 2010, and noted some limitations.
      http://deevybee.blogspot.co.uk/2010/12/neuroprognosis-in-dyslexia.html
      As I stressed, I think their approach is interesting from a basic science point of view, though it is very early days, and one would want to see much larger studies that allowed estimates of sensitivity and specificity of prediction before accepting their conclusion. But the biggest issue for me is that the idea of using brain scans to predict outcomes is totally unrealistic for practical application. I mean, who is going to take a poor reader and put them in a brain scanner in order to predict how well they'll do a few years down the line? Even if the prediction were extremely accurate, many would baulk at doing this, on the grounds that it would be extremely expensive, and it might make more sense to wait and give extra help to children who failed to progress. Those who advocate this kind of thing would do well to also look at the literature on the practicalities and ethics of screening in health care, where there is much debate about the potential harms as well as benefits of attempting to identify outcomes before they occur (see McCartney, M. 2012, The Patient Paradox: Why Sexed Up Medicine is Bad for Your Health ).
      I hadn't known about the Superkar et al 2013 PNAS study, but I took a quick look at it and it seemed to have many of the problems that I discussed in my JCPP review article. Low statistical power, use of a 'super-normal' untreated group, and conclusions based on findings of a few significant correlations emerging from an unspecified large number of variables in a whole-brain analysis. In addition, they make much of the fact that a correlation is significant in the trained group and not in the untrained group, even though (a) the untrained group is a 'supernormal' group, (b) the untrained group is much smaller and (c) the confidence intervals of the correlations overlap (as far as one can tell - the presentation of the data don't make it easy). If these results were to replicate, I'd agree this would be of interest, but I don't have much confidence in them, and I don't see this finding as having application to education.
      Re the neurobiological grounding of discrepancy criteria, it's nice to see that the brain imaging data agree that we should reject these, but this general point was made convincingly by Stanovitch and others over 30 years ago (see, e.g. Stanovich, K. E. (1991). Discrepancy definitions of reading disability: has intelligence led us astray? Reading Research Quarterly, 26, 7- 29.
      I'll take a look at Dweck's work: Hugh also mentioned this in his comment, though with perhaps a more cynical take on it! I note that the abstract of the article you sent, though, seemed to suggest that it did not depend on any neuroscience.
      You suggest: "if we gain a better understanding of the mechanisms of how the brain changes as a function of educational experiences (the notion of biological embedding), we will be able to use such information to provide more refined, evidence-based experiences to drive neuroplasticity? etc" This all sounds fine in principle to me, but I really would like to see some good examples of how this pans out in practice.
      In sum, I must emphasise I am NOT opposed to using neuroscience methods in the study of children's learning, but I am worried that, rather than quashing the neuromyths that we all agree have gained a pernicious hold in the world of education, we may just add to them by promising more than can be delivered.

      Delete
    2. Dorothy: thanks for this! I completely agree that we need more examples. These are very early days and your criticisms of the examples I cited are very well taken. I guess we cannot fully predict how methods and the cost associated with them will change over time. Who would have predicted in 2001, following the publication of the first sequencing of the human genome, that whole-genome sequencing can now be achieved in less than a day and that soon the cost will be less than $100 per genome?

      Perhaps the question should not be about what neuroscience can add that is unique and fundamentally different from what we can glean from other methods/approaches but rather how neuroscience can play one important piece in a drive towards more evidence-based education (hence my preference of 'Mind, Brain & Education' rather than 'Educational Neuroscience')

      As put very nicely by Chloe Marshall in her response, the current level of interest in neuroscience among educators and funders of educational research can be used for good to build an interdisciplinary approach towards evidence-based education. For this to happen, discussions such as this one are essential.So thank you for providing a forum for exchanges on these issues!

      Delete
  5. I agree with all that you say above, Dorothy. But mightn't you go farther? Here is a (slightly) amended passage from your blog:

    "Neuroscientists can tell you which brain regions are most involved in particular cognitive activities. . But these indicators of brain activation do not tell you how people achieve these cognitive activities. Suppose I find out that the left angular gyrus becomes active when people are reading. What is a cognitive psychologist supposed to do with that information?"

    How sympathetic are you to this?

    Max

    ReplyDelete
    Replies
    1. Sympathetic but probably not as convinced as you. I think people overstate the case of how far localisation information can inform cognitive theory, with a great deal of inappropriate reverse inference. But (in agreement on this with Daniel) I think we'll go further by putting all sources of information together, rather than treating the neuro/cognitive distinction as either/or.

      Delete
    2. I must emphasise I am NOT opposed to using neuroscience methods in the study of children's learning, but I am worried that, rather than quashing the neuromyths that we all agree have gained a pernicious hold in the world of education, we may just add to them by promising more than can be delivered.buy rap beats with hooks

      Delete
  6. Cognitive psychology has been building a body of research since around 1875 that describes memory functions and predicts in many circumstances how people learn more efficiently. Most of that was accomplished through paper-and-pencil or computerized presentation methods, Now neuroscience can look inside the brain and is producing a line of research that mainly verifies what the 1940s psychologists found. This is all the more reason to heed the lessons of cognitive psychology and look for innovative applications. (e.g. for low-income populations in poor countries.) As Daniel Ansari and others wrote, neuroscience by itself is not ready for prime time in education.

    ReplyDelete
  7. There is a significant amount of skepticism among educators and education researchers about how both psychology and neuroscience may inform educational practice. I think it is well placed. I occasionally ask my fellow psychologist and neuroscientists how often they *interact* with educators. And I don't mean give a lecture to. Much of this talk about how psychology and neuroscience may be relevant to education is theoretical. For things to work in practice there needs to be substantial give and take between psych/neuro folks and education researchers. In my experience that interaction doesn't happen as much as it should.

    I am optimistic, otherwise I wouldn't be doing the research. I think that a big issue going forward will be not just the content of psych/neuro research but the interactions with educators and education researchers.

    ReplyDelete
    Replies
    1. I do think this is an important point. Some voices in the educational neuroscience can seem quite arrogant - as if they possess special knowledge that they will disseminate to grateful teachers. I know most people working in this area aren't like that: on the contrary, they do try hard to engage with educators. But I think we have a lot of work to do if we are to establish genuine two-way interaction.

      Delete
  8. A timely post, given that the Wellcome Trust and Education Endowment Foundation are about to invest £6 million in educational neuroscience, http://www.wellcome.ac.uk/News/Media-office/Press-releases/2014/WTP055264.htm.

    I agree with Daniel Ansari’s view that the interdisciplinarity between education and neuroscience is better served by them being partners with psychology, rather than the arrow going directly from neuroscience to education. Public Health outcomes have improved greatly with contributions from genetics, molecular biology, cell biology, microbiology, health psychology, epidemiology etc. Education similarly needs to be built on an evidence base from different disciplines. It would make no sense to leave neuroscience out.

    As someone who works at a university where our student body is composed largely of trainee teachers and qualified teachers returning for postgraduate study, I am shocked that, as far as I can work out, our students are offered no developmental psychology (let alone neuroscience!), as part of their initial teacher education (perhaps my colleagues can correct me if I’m wrong?). To me this is as inconceivable as training doctors with no knowledge of anatomy. I have sat in Board of Examiners’ meetings where the examiners have questioned why my colleagues and I teach Masters-level modules on literacy development within a cognitive psychology framework rather than a sociocultural framework. Education has little hope of becoming an evidence-based profession under these circumstances – it will remain subject to fads and government dictats.

    ReplyDelete
  9. This comment has been removed by the author.

    ReplyDelete
  10. Great post. It has struck me similarly that people in the neuroscience and genetics of both broad cognitive abilities (e.g. intelligence) and forms of psychopathology (e.g schizophrenia) often feel the need to justify the neuro/genetics studies of these topics with the handwavy promise that 'in the future, we will be able to better measure these traits with scanners/genetic screening'. This seems to be conceptually incoherent (if a trait is defined behaviourally, how could a scanner outperform this gold standard, other than smaller error terms perhaps), and in many cases simply misguided (even the most stark of neuroscientific differences shows greatly overlapping distributions, suggesting that classification accuracy at the individual level will be inherently low, not to mention the cost). Instead, it would be much preferred if people honestly state that the reason we study the brain and/or genetics is in an attempt to better understand the mechanisms underlying these traits. This fundamental research may, or may not, at some undefined point in the near or more distant future have real-life applications. But if we tie basic research too tightly to these short-term promises then the public (and/or funders) will, rightly, hold the inevitable lack of success (in this narrow sense) against the field, and then it might be too late to revert to the true goal, which is an increase of our fundamental understanding first, applications second.

    ReplyDelete
  11. Dorothy
    I mostly agree with you--the case is oversold, and most of what's pitched as "educational neuroscience" is uninteresting. But I do think there are techniques by which neuroscientific data can be brought to bear on educational problems (and does so through cognitive theory, but in ways that cognitive theory alone could not).
    I published a paper on this a few years ago, which can be found here
    http://www.danielwillingham.com/uploads/5/0/0/7/5007325/willingham__lloyd_2007.pdf
    I'll provide another example on my blog tomorrow!

    ReplyDelete
  12. Thanks for this much needed corrective. I will certainly add this to my list of links to send people to as an antidote to undue neuro-enthusiasm.

    I've made some similar points some time ago in my post on language learning and neuroscience http://metaphorhacker.net/2011/03/the-brain-is-a-bad-metaphor-for-language/.

    I'm all for neuroscience being done (the more of it the better) but many researchers are at fault for fanning the hype flames. At the moment, it seems to me, the outreach and public education should be about the limitations and not the potentials of neuroscience.

    But I'd go even further and say that even "well-designed trials of cognitive and behavioural studies of learning" are not what education needs more of. Learning is a very complex process that takes place over different time scales, different units of knowledge and action, and individuals as well as social groups. So far, cognitive studies of learning have mostly reinforced the notion of education as mostly remembering facts, while there is ample ethnographic and historical evidence that this is only a relatively small part of what goes on in schools, homes, and societies. Think about hidden curricula, pedagogy of the oppressed, 'My Freshman Year' year, etc. The Taylorist notion that we would have a more productive society if only we could squeeze more out of (or into) young brains while they are our captive audience is more likely to do harm than good. Be it interactive whiteboards or Ofsted-pleasing scripts.

    That is not to say that more research is not needed. Particularly in areas where there are relatively clear cognitive barriers to acquiring information. However, even here, there's the problem of overinterpreting the results that was nicely summarised by Andrew Gelman here: http://andrewgelman.com/2014/01/17/think-statistical-evidence-statistical-evidence-cant-conclusive.

    ReplyDelete
  13. I could not agree more with you Dorothy. However there is even worse than educational neuroscience: neuroeducation, that is, a kind of education science purportedly guided by neuroscience, but mostly in a rhetorical and misguided way, for the precise reasons you have pointed out.

    ReplyDelete
    Replies
    1. Precisely!

      Dorothy -- any objection to naming names (or websites) of perpetrators of "neuroeducation"? I think it would help folks to see examples of what is being sold.

      Delete
  14. I think you’re right to identify the early stage of the field of educational neuroscience (EN), and the relative paucity of examples of solid success for which it can claim sole credit. I'm not sure I agree with your broader arguments about EN.

    It’s true that neuroscience is a long way from what goes on in the classroom. Neuroscience is indeed remote from education, but pointing to the distance between levels of description isn’t a convincing argument against their useful interaction. “Physicists can’t build bridges!”. “Biochemists can’t cure sick people!”. “Neuroscientists can’t plan primary school lessons!”. Could you really have robust principles of engineering without a theory of the physics of materials? Could medicine have really got so far without a theory of how the body works? Why should education have some special immunity to the insights provided by mechanistic theories at lower levels of description?

    I'm guessing you’ll agree but insist that psychology should be a key intermediary between neuroscience and education. However, you seem to want to go further, that the only people useful to teachers are psychologists.

    Psychology is not independent of neuroscience. Indeed I think that psychology has not yet been sufficiently influenced by neuroscience. In turn, this has limited the efficacy of psychology in informing education. In fact, I’d argue that current psychological theory is, in many respects, not fit for purpose. It remains implicitly influenced by a metaphor of mind rooted in the desktop computer, with core ideas of abstraction and domain-general mechanisms.

    Instead, our psychological theory needs to be influenced by the implementation constraints of the brain – what computations the brain finds easy to do. Such a reshaping of psychology will be an important focus of EN.

    What are the grounds to doubt current psychology? Here are some clues that suggest current psychological theory has limitations with respect to education:

    - It is poor at predicting the range of transfer effects in training interventions; training often doesn’t generalise as much as you’d expect given cognitive theory

    - Psychological constructs don’t seem to match particularly well to the activation of neural structures; there are too many many-to-one and one-to-many relationships

    - Emotion is peripheral to most cognitive theories, despite its primary adaptive role for mammals;

    - Current cognitive theory doesn’t explain how and why plasticity should change with age;

    - There are phenomena that seem surprising given current cognitive theory, typically leading psychologists to cellotape post-hoc additions to their theories – such as embodiment, the role of sleep, or the effects of meditation.

    Together, I think these clues suggest we haven’t got our theory of cognition quite right, yet. I’d argue that’s because we haven’t taken the implementation constraints of the brain seriously enough. And in turn, this is because some psychologists mistakenly believe that because they can study psychology independently of the brain, it is independent.

    Must psychology always intervene between neuroscience and education? In the blog, you say brain imaging will never be used for diagnosis because behaviour is always there, and always cheaper to measure. What about electrophysiological measures such as Event Related Potentials (now sufficiently portable to be used in the classroom)? Recent evidence suggests diagnostic predictors of later disorders may be measurable via ERPs in infancy some months before symptoms appear in behaviour (e.g., the work of April Benasich looking at developmental language disorders; or the work of the BASIS network studying infants at risk of autism). Even if behavioural symptoms manifesting the disorder appear only a few months later than the brain markers, those few intervening months could be key for early intervention.

    ReplyDelete
    Replies
    1. Hi Michael - thanks for the comment. As you anticipate, I agree with a lot of what you say. My gripe is not with people doing developmental neuroscience (after all, I do some myself), it is with the over-hyping of the applied significance of the findings to education.
      When I ask people for good examples of such application, the answers fall into 3 categories:
      a) Psychological findings applied to education, where the brain research is redundant
      b) Neuroscience findings that are interesting but have questionable application to education
      c) Neuroscience studies that have potential application to education but are still unvalidated and are at least 10 yr away from being translated into practice.
      I'd put the ERP diagnosis studies in category c). If you're going to use ERP for prediction, the first thing you need is to demonstrate that the measure is reliable at the individual level. The next thing you need to do is to demonstrate adequate sensitivity and specificity for prediction in a realistic population. Most neuroscientists who work in this area aren't trained in either psychometrics or statistical epidemiology and don't have much of a clue about either of these things, but they are absolutely crucial if the technique is to be of any use. I agree this doesn't preclude the diagnostic use of such methods in the future, and you have to start somewhere. I just think many of those who talk about using neuroscience to predict outcomes gravely underestimate the kind of work that would need to be done, and the timescale that would be required, for this work to be translated into practice.


      Delete
    2. >> Indeed I think that psychology has not yet been sufficiently influenced by neuroscience.

      Unfortunately, psychology itself is in much the same position as education when it comes to neuroscience -- the accomplishments and even the possibilities are vastly oversold.

      For example, one prominent neuroscientist who should know better is optimistically flogging in the pop press and books the influence of neuroscience on actual psychotherapy. In such a book he claimed that there were already a number of success stories along these lines, and he propounded one. Unfortunately, when you read and analyze his example even a bit carefully, it falls apart -- it is a psychotherapeutic method neither inspired nor even confirmed by neuroscience. At best, the neuroscience is a metaphor, or what mathematicians call proof-by-hand-waving -- with some verbal sleight-of-hand he makes it seem, on a quick read, to validate his claim. My thought -- if that's the best he's got, then we probably have the null set on our hands.

      Delete
  15. Wonderful post, Dorothy, thank you. We see this all the time in education...in fact, it seems that anything with "brain" or "neuro" in it has automatic credibility with some educators. I agree that the fact is that, on some level, what's happening in the brain matters very little in education. For example, I find fMRI studies very interesting, but those images tell me nothing meaningful in terms of learning and performance in the classroom. I still need to conduct behavioral studies and measure relevant performance of skills in order to demonstrate that learning has occurred....regardless of what part of the brain lit up.

    Very thoughtful article and I appreciate it.

    ReplyDelete
    Replies
    1. I appreciate it, too. As a parent of a child with an atypical learning style, sometimes we feel run-over by the hype, and a bit paranoid of the frenzy. Dr. Bishop has a strong mind, but it is her stout heart that encourages me that our children will not be victimized by over-zealous intellectualism.

      Delete
  16. Terrific post -- much needed. I will be distributing this as widely as I can.

    I myself have been trying to make the same point with regard to the neuroscience of psychological disorders and well being and psychotherapy, which is demonstrably in exactly the same state as the neuroscience of eduction. But without credentials, it is often a difficult sale. I have been told that without a Ph.D. in neuroscience I need to keep my mouth shut.

    A big part of the problem, of course, that there are far too many neuroscientists (and eloquent pretenders), some very prominent, who certainly understand quite clearly the truth you expound here and would not dare to contravene it in scientific publications, but who still insist on misleading the popular press and public with empty notions such as "neurally-inspired psychotherapy" or "interpersonal neurobiology".

    But this is science, so perhaps someday -- not too far off, I hope -- the truth will prevail. Keep up the good work... and consider broadening the range of your critical artillery.

    ReplyDelete
  17. A wonderful post and very relevant discussion. I would like to add a comment in favor of the relation between mind, brain and education. As stated in some earlier comments, there has been done a lot of work (by Dweck and others) showing effects of ‘brainknowledge’ about plasticity on the growth mindset of children and adults, and subsequently on their learning outcomes. I experience in working with teachers (not only giving lectures) that this information is very valuable and relevant for teachers. Also providing teachers with very basic information about neuronal processing AND making the link between this information and learning and teaching is received with great interest and improves teachers attitudes and views towards learning and development (see for example an article by prof. J. Dubinsky, 2013 in Educational Researcher). To many researchers it seems obvious that the brain changes when you learn, but many teachers are not aware of this (many actually believe that we only use 10% of our brains). These are examples of how knowledge about brain functioning can be used as a tool in teacher professional development.
    If we want to improve education, why not use knowledge about the brain to better equip teachers? After all, it is the tool with which we learn. I would not call this approach educational neuroscience (there is no neuroscience in it). If I would have to give it a name it would be neuro-education. However, the aim of this endeavor is never to provide teachers with knowledge about the brain, but the aim is to improve education by using what we now about the brain. That is a major difference, because it forces other criteria on what kind of information is tought to teachers and on how this information is transferred to them. Education can be improved by providing teachers and student with more realistic views of and attitudes towards learning and by improving teachers’ attitudes towards active, inquiry based learning methods.
    Research on the effects of this kind of ‘neuro-education’ is not straightforward; how do you measure attitude change, or how do you know that a teacher is becoming a ‘better’ teacher. The fact that a lot of teachers are interested in brain functioning is hardly valid support. However, more studies, such as that of prof. Dubinsky, are appearing and i do believe it is worthwhile exploring this approach. Educational researchers should look at neuroscience to select those results and that knowledge that they can use to reach their specific education improvement goals.

    ReplyDelete
  18. If neuroscience really has no valid use in education at this point, I fail to see how neuro-education -- teaching people about how their brain works in relation to their mind (in the extremely primitive state of current knowledge) -- can be considered legitimate education. Either it is a lie -- it way overstates what we actually know and can use from neuroscience -- or a reassuring metaphor, an educational placebo as it were. And if it is the latter, there are many many more compelling metaphors -- from literature, the arts, contemplative traditions, other science -- that would truly enrich children's understanding of mind and their own mind, much moreso than the highly distant and abstract tropes of neuroscience. I think Bishop's example of the "left angular gyrus" is a perfect case of such a misleading, uselessly remote and abstract concept.

    ReplyDelete
  19. One can only agree that if one is interested in the behavioural impact of a teaching method, one must measure behaviour rather than provide a colourful picture of the brain. Yet:

    - Not all brain imaging studies are designed to purely localize processes or find brain correlates of a given ability. As a cognitive psychologist, I believe that ERP, MEG, fMRI, are additional tools in my toolbox which, along with behavioural measurements, can help shed light on the *functional* architecture of the mind. Opposing neuroscience to psychology is sterile. One has to be clear about the question one is interested in, and use the best tools available (and realize that obtaining converging evidence from different methods is the most satisfying situation).

    - I find this post a little bit optimistic about the actual relevance of Cognitive Psychology to teaching. In theory, obviously, it *should* be relevant, just like Physics is sometimes relevant to Engineering. Yet, I wonder if there exist many cognitive theories that are established enough to serve as solid grounds for educational practices (?). This is not to say that we know nothing, but I feel that even the best theories are too general and vague to take strong decisions on education. (By contrast Neuroscience can be relevant: at least we know that we will not acquire the knowledge of a person by eating her/his brain ;-)

    - Educators naturally want to know which methods work best. Given the current state of Psychology, my feeling that the best way is to directly assess the methods using the experimental method or observational studies. There, cognitive psychologists have a little bit to offer in terms of methodology., but we should remain modest and realize the complexity of assessing different methods in ecological settings.

    ReplyDelete
    Replies
    1. I agree with Christophe that one of the main things that psychology has to offer is methods to evaluate what works. And this in itself is a lot. Even if natural situations are immensely complex and intervening factors are numerous, we know how to simplify situations and practices for a first shot, evaluate them rigourously, and incrementally augment the number of factors taken experimentally into account to approach reality.

      Delete
    2. However I don't think that the issue of not having a satisfying grand psychological theory is such a problem. I think that we all realise that grand paradigms are dead, they were an oversimplification of the past. But the lack of such theories may not be a big loss. We still have a lot data that is relevant to education.
      Take for example the work on "the importance of retrieval for learning" by Roediger (http://www.sciencemag.org/content/319/5865/966.short). Maybe there is a grand theory for that, maybe there isn't. But who cares? This work has direct applications in the classroom. (in a way that no brain image will ever have)

      Delete
  20. Very interesting post

    ReplyDelete
  21. Thanks for a great post!
    I agree with you that there is a lot that can be done on an individual level and at home when it comes to gender balance, but I think it is also important to remeber that much can be done also on a societal level. I come from a scandinavian country, and here the father has to take at least 3.5 months parental leave during the babys first year, otherwise the family looses the payment for this. I think this has been a very important step. The baby from very early on gets used to that the father is an equal caregiver as the mother, and after going back to work it becomes easier to share not only the joys of parenthood, but also the responsibilities. I think this has made it much easier for women that often have children when they are post docs to stay in academia. Although I am sure this is not the whole explanation, the figure for female professors is steadily increasing here and is now about 25%

    ReplyDelete
  22. As someone whose career has moved from neuroscience to education, I am disappointed by a lot of the cynicism expressed here! Yes, a lot of current 'educational neuroscience' is oversold, but so are many, many other fields. Yes, a lot of the findings are a long way from being directly translated into everyday teaching practise, but isn't that how science works? How many biomedical researchers are working on research questions that could be instantly translated into everyday medical practise? Not many!

    Education suffers (IMHO), from having a weak evidence base, in some cases an outright hostility to the very concept of 'evidence'. The persistence of educational pseudoscience such as 'learning styles', 'left brain right brain' and 'brain gym' are examples. The fact that there are some researchers who are keen to ask questions about how the neuroscience of learning can be extended a little further into education is, overall, I think, a positive development given where we are now. We might have to hold our noses a bit as the field becomes established, and we might need to call on psychologists to translate back and forth between educators and neuroscientists, but good things will come eventually.

    Phil Newton

    ReplyDelete
    Replies
    1. I thoroughly agree with you that education suffers from lack of a strong evidence base and often there is hostility to evidence. What saddens me is when, instead of developing its own approaches to changing that, the field of education laps up supposed 'neuroscientific' evidence that actually has little bearing on what goes on in the classroom. As I've said elsewhere, either it's neuroscience that has no application, or it's potentialy useful for education but has no neuroscience content. Some of those promoting neuroscience in education are actually quite patronising to teachers; people who've never been in a classroom claim that by finding out about the brain, teacher's teaching will somehow be transformed.
      Anyhow, I can reassure you that I will be the first to rejoice if a link can be made between neuroscience and education that actually is meaningful. But so far, the evidence for that is very weak. I found it interesting that only one of the 6 projects funded by the £4 million EEF/Wellcome round on educational neuroscience seems to involve neuroscience (in the sense that it incorporates some measure of brain activity): http://educationendowmentfoundation.org.uk/news/new-research-to-investigate-if-neuroscience-can-improve-teaching-and-learni/
      It doesn't surprise me that they were not able to attract applications that actually fitted their remit better, but it confirms my suspicions that the word 'neuroscience' is just inserted into proposal to make them sound good, when really they don't need it at all.
      This result has also angered many others who would have submitted proposals to this round, but thought they were not 'neurosciency' enough.

      Delete
    2. P.S. Pedantic correction: in fact, I think the Johansen-Berg study will look at brain structure rather than brain activity.

      Delete
  23. Dear Dorothy, that is not how the panelists on the EEF/Wellcome round saw the applications. Here is a section of my rejection letter:

    “Grant rounds are always extremely competitive, and in every round there are many strong applications that meet our criteria, but which we are not able to support. For this particular funding round we were unable to support any intervention that did not have a strong basis in neuroscience and this was the reason that we were not able to take forward a large number of the submitted proposals.”

    I thought that was reasonable enough (given that there really was no neuroscience in the proposed studies), till I saw what was funded!

    Jeff

    ReplyDelete
  24. I agree, from a quick look through there appears to be little neuroscience in those projects, but i don’t think that’s cause to write off a whole field. Going back to your original post; “But at the heart of this enterprise, there seems to be a massive disconnect...........suppose I find out that the left angular gyrus becomes more active as children learn to read. What is a teacher supposed to do with that information?”. Yes, that’s true, but one could replace ‘as children learn to read’ with pretty much any behaviour (‘when recognising a loved one, ‘during a psychotic episode’, ‘when spending a lot of money’), and the same disconnect would likely still hold true for practitioners on the front line (therapists, psychiatrists, economists, etc). Does that invalidate all research which attempts to understand relationships between brain activity and behaviour? (Some would say so, but that’s nothing specific to ‘educational neuroscience’!).



    Yes, there are people patronising each other and some selling overhyped snake oil, but again I don’t think this is specific to educational neuroscience. Hopefully the hype will pass, and the backlash will not be too brutal, and we will be left with something useful – education is important, and I think it’s pretty safe to say that the brain is an important part of that important thing – attempting to understand education from a neuroscientific perspective seems (to me) like a worthy thing to do.

    ReplyDelete
    Replies
    1. Thanks Phil. It's useful to have this discussion .I know that, while many agree with me, others think I'm throwing the baby out with the bathwater.
      As someone who uses neuroscience methods in some of my research, I have to agree it is worthwhile - both for understanding mechanisms behind children's learning and in areas like psychiatry. But the hype and exaggerated talk of translation is causing real damage - not just in terms of misleading teachers, but also in limiting the kind of research that gets funded.
      The projects that were funded by the EEF/Wellcome initiative will no doubt lead to interesting findings, and obviously the researchers persuaded the funder that they met the criteria: "Projects will only be funded if there is an explicit causal hypothesis relating a finding in neuroscience to a novel intervention, not if neuroscience simply explains the basis of already established and proven education interventions." But the sad thing is that they could have had a lot more decent proposals if instead of (or in addition to) 'neuroscience' they had put 'psychology'. As it was they do not seem to have spent the 6 million allocated for this programme: and meanwhile, people with good ideas for educational interventions struggle to get funded.
      It is interesting that you drew a parallel with psychiatry. A similar process is taking place there, as I discussed here. http://deevybee.blogspot.co.uk/2014/05/changing-landscape-of-psychiatric.html. Someone on Twitter commented that all NIMH projects these days seem to require an fMRI componenet, and we have to ask if that is a sensible use of resources. Again, neuroscience is fine in its place, but its place should be complementary to other approaches, not substituting for them.

      Delete
  25. The problem is that you provide may be worth our time and also effort.Teachers Planet

    ReplyDelete