Learning styles – the myth persists

Asher Cashdan highlights some issues.

Emily Reynolds’ account of recent developments in learning styles (LS) (April issue) was most welcome. Reynolds summarised the massive review by Philip Newton and Atharva Salvi (2020) of the ‘neuromyth’ surrounding LS, mainly as practised by teachers in Primary Schools. Although all the evidence is against the usefulness of LS, the myth persists. 

A number of further issues are pertinent:

  1. There is no tested definition of the nature of a LS, although some 70 of them have been proposed.
  2. A pupil’s preference for a particular style is virtually always derived solely from their answers to questionnaires – neither a reliable nor a valid source of information. In fact, it is likely that many pupils have not understood the questions or have given the answers they believe are the ‘right’ ones. 
  3. Something like 90 per cent of teachers, in the US at least, believe in tailoring their teaching to suit individual children’s preferred styles, using the now dominant VARK classification (visual, auditory, reading/writing, kinaesthetic). Thus, they would provide different materials for a visualiser from those for an auditory learner. However, very few of them actually attempt to carry this out in practice. Teachers have their own ways of working. One said to me recently, ‘I provide a variety of approaches to much of our work and pupils may choose which they wish to use’. When teachers were asked which method they were currently using to teach reading (see Bullock, 1975, Table 30), they were expected to show a particular preference for either look-and-say or phonics. In fact, they used a mixture of methods. 
  4. There is overwhelming evidence, as assembled by the Newton team, that where such teaching is employed, no good outcomes result. It may even do harm at times.
  5. There is no evidence that learners retain their preferences over time and little that such preferences extend over all curriculum subjects.
  6. Teacher-trainers seem still to be offering LS to their trainees as a desirable feature of good practice, thus perpetuating belief in them even in the youngest teachers.
  7. After teachers have been shown the deficiencies of LS practice, some 37 per cent of them persisted in supporting their use.

What does all this amount to? Should we tell students and trainers to drop the whole idea? Or should they go for a different, better array of LS? I doubt this. At the Open University we introduced students to a few other LS as far back as 1971, including field dependence-independence, impulsivity-reflection, holists vs. serialists, convergers vs. divergers, while explaining that none had been shown to be valid. They all suggest sensible possible procedures in the right context, but fall far short of describing good practice in teaching specific learners. The assumption that a preferred style might apply across the whole range of curriculum subjects is again unproven and I would certainly expect many children to change their stated preferences quite frequently.

This last point seems to me to bring in the myth that has long been part of psychologists’ culture: the acceptance that certain qualities are ‘given’ at birth (or earlier) and are almost unmodifiable. Such was the firm belief of Burt and the other ‘intelligence testers’ and similarly of personality theorists such as Eysenck and Cattell. Yes, it is now universally agreed that environmental factors can make a significant contribution to people’s development, creating a blend of heredity with environment. But that rather assumes that the hereditary component is itself fixed and identifiable. It may well not be so. In other words, humans have their own plasticity – a capacity for development and change which is often underestimated. 

In the end, although teachers cling for some time to exploded and outmoded practices, these do eventually release their hold. Who still remembers the Initial Teaching Alphabet, though research showed that it had some benefit for initial reading? And who still purveys Bernstein’s codes: restricted for the working-classes and elaborated for the better-off? I feel sure that LS will eventually go the same way – only to be succeeded by some other new-fangled craze.

Asher Cashdan
Emeritus Professor of Education
Sheffield Hallam University


Newton, P.M. & Salvi, A. (2020). How Common Is Belief in the Learning Styles Neuromyth, and Does It Matter? A Pragmatic Systematic Review. Frontiers in Education, 5, 270.
A Language for Life – The Bullock Report. (1975). London, HMSO.

BPS Members can discuss this article

Already a member? Or Create an account

Not a member? Find out about becoming a member or subscriber