Landing : Athabascau University

Babies in the learning-style bathwater

A recent Guardian article reports on a letter sent to the paper by 30 eminent academics from neuroscience, education, and psychology disciplines, voicing concerns about the absurd popularity of learning styles among teachers.

They are, of course, correct to be concerned. There is no good evidence that being taught according to your learning style has any positive value, despite decades of spurious attempts to show a correlation. Moreover, even if there were such a correlation, it would behoove teachers to help learners to learn using different styles because real-life learning doesn't come neatly packaged in forms that fit with how we want/are constituted to learn, and teaching should primarily be concerned with supporting learners' capacity to learn. The fact that there are scores if not hundreds of incompatible learning style theories, most of which have similarly (un)compelling evidence to support them, should be a clue that there is something seriously wrong with the whole idea. And it's not a harmless foible. Not only is it a massive waste of time and money, not to mention a terrible example to set in truthiness acceptance, it can be actively harmful to learners, teaching them to believe that they can only learn properly if things are packaged to suit their style. 

What's shocking in the article is the report on the number of teachers who, despite a total lack of evidence and copious amounts of debunking, continue to use and believe in the things. To our shame, I have even seen examples of it at AU (our own Math Site mentions them) where we really ought to know better. But we are not unusual in this. Not at all. In the UK and Netherlands in 2012, 80% of teachers apparently believed that individuals learned better when doing so in a manner according with their preferred learning style. This is like discovering that 80% of the world's scientists believe that their horoscopes determine the results of their experiments.

That said, there's a baby in this very dirty bathwater that should not be thrown out.

If a belief in learning styles means that teachers feel challenged to design learning experiences in different ways to suit more diverse needs, that's not a bad thing, apart from that it increases the costs of learning development. In fairness, it would work at least as well if they used astrological star sign personality characteristics as a basis but, whatever the reasons, giving students choices is a worthwhile outcome. And, just like horoscopes, there is value to learners themselves in providing an opportunity and a framework for reflection, even if the framework itself is erroneous and based on fallacies.

I'm a sceptic, but even I use variants on the theme. For example, I often try to provide versions of learning content that are meant to cater for serialist and holist ways of learning (Gordon Pask's approach to categorizing learning strategies). Notwithstanding the extra effort and cost of designing at least two ways to approach a topic, it's a good creative catalyst for me, and it gives students greater choice and control over their own learning.

And, in fairness, not all learning-style types of theory are equally awful. Slightly less harmful variants talk of learning preferences rather than styles, which does not necessarily imply that those preferences are a good idea nor that they even need to be catered for, though it still perpetuates the myth that there are relatively fixed characteristics in such things. Much better ones, including Pask's, talk of selectable learning strategies rather than stable characteristics or preferences of learners, which seems eminently sensible to me: it's just about general pedagogical patterns. It's not about labelling learners, though (sadly) some do try to apply the labels to learners, and even Pask himself (arguably) sometimes seems to present it in that way. The best of breed models recognize that learning strategies can and should change in different learning contexts as well as over time, and make no attempt to label or pigeon hole learners themselves at all. I think it is really useful to find regularities and patterns in learning designs, and that's the baby we should not throw out when we (rightly) reject learning style theories.


  • Jon,

    Thanks for sharing this article and your commentary.

    I tend to agree with this:

    If a belief in learning styles means that teachers feel challenged to design learning experiences in different ways to suit more diverse needs, that's not a bad thing...

    Generally speaking (and, as always, the devil is definitely in the details), the learning styles conversation can be a good way in a classroom setting to have teachers and students start to talk about student directed learning. The issues I have seen (apart from the pure neuroscience component) is when this opening into personalized learning becomes a rigid pedagogy (or rigid or a pedagogy). Then things become ugly quickly.

    Gerald Ardito March 15, 2017 - 6:04am

  • I'd still love to supervise a comparative study into the use of learning styles vs use of astrology or phrenology to pick a teaching strategy. I strongly suspect there'd be no significant difference. It would be equally good fun to invent a plausible but totally unfounded learning style theory and compare that. Maybe something based on the big 5 personality types so that it seems sciencey.

    Your comment on personalization is spot on. In some ways it would actually be worse if it worked. Even if a system does increase the speed/efficiency of learning as a result (as measured in tests) the assumption that the teacher-specified outcome is the one and only point of the learning process describes pretty much everything that is wrong with our educational systems today. Not a recipe for cognitive flexibility, not transformative, not life-changing, just a better form of indoctrination.


    Jon Dron March 15, 2017 - 10:18am

  • I think we could get a grant for a phrenology study. Surely, it will be resurrected as a new fad in education any day now.

    So, you are saying indoctrination is a bad thing? Laughing

    Gerald Ardito March 15, 2017 - 12:35pm

  • Interesting article Jon. I can believe that 80% of teachers in the UK and the Netherlands believe that student learn best in their prefered style. (I wonder what the percentage would be in Canada.)

    I used various learning style inventories for quite a few Septembers for a couple of reasons. They gave me a chance to learn a lot about students and their approaches to things like following directions in a way that engaged them - who doesn't like thinking about themselves? It also gave me the opportunity to introduce the concept of metacognition and using strategies for learning. I was also curious about the whole idea and noticed that although I had inventories that were designed to be age appropriate, students didn't develop preferences until they were 10 or 11. Until then their profiles were flat. 

    In the end, if they prompt teachers to accept learning strategies that are different from their own and encourage students to think about how they learn, there is some good in them, but they're a long way from science. 

    Mary McNabb March 15, 2017 - 10:30pm

These comments are moderated. Your comment will not be visible unless accepted by the content owner.

Only simple HTML formatting is allowed and any hyperlinks will be stripped away. If you need to include a URL then please simply type it so that users can copy and paste it if needed.