Thursday, 4 October 2012

Learning styles: facts and fictions

This article originally appeared in Modern English Teacher, volume 21 No.4.  Thanks to Dave Francis for allowing me to reproduce it here.  
 Learning styles (sometimes 'learner styles' hereafter LS) are pervasive in education and there are far too many articles examining the learning styles of various groups of students to reproduce here. However a quick search of Google scholar will bring up numerous articles examining the various LS of various groups, from Iranian freshmen to Taiwanese and Kuwaiti students. However, for the amount of attention they receive there is very little evidence of their efficacy:

The contrast between the enormous popularity of the learning-styles approach within education and the lack of credible evidence for its utility is, in our opinion, striking and disturbing. If classification of students’ learning styles has practical utility, it remains to be demonstrated (Pasler et al 2008:117).

 This essay will examine LS in the EFL literature and question whether they actually provide a useful tool for aiding students language learning or whether they are merely a pseudo-scientific distraction. Firstly, This essay will examine the theory of learning styles, focusing on the VAK model in particular. It will then examine references to LS in a small sample of EFL literature. Finally some of the problems with the theories of LS will be outlined.

What are learning styles?
The basic premise of LS is that 'individuals learn in different ways'(Nel 2008:51).  There is a large body of literature related to the various types, whether they be visual, audio or, kinesthetic, (VAK) left brained or right brained, concrete or communicative, levelers or sharpners, plungers or non-committers, convergers or divergers, the list, while not endless is certainly long!  The thrust of the various theories is that if a teacher caters to a student's particular LS then said student's learning will be enhanced. (Thornbury 2006)  Therefore a teacher should prepare a variety of approaches when introducing material in order to cater for these disparate needs.  

There does appear to be some conceptual confusion (Nel 2008) with regard to this term. Whereas for some, LS simply indicates the things different students prefer to do in order to learn,  For example, student A may use flash cards and student B may like writing lots of notes. For others LS indicates a scientific theory relating to 'the biologically and developmentally imposed set of characteristics that make the same teaching method wonderful for some and terrible for others' (Dunn & Griggs, 1988:3). There is quite a difference here between that which is preferential and that which is biologically constrained. This paper will specifically deal with latter of these two notions, though it is possible that the vagueness of the concept has actually aided its popularity, as a wide range of activities can be grouped under one catch-all term.   

weak and strong forms

It is useful to divide LS into two versions, for the sake of criticism. The first can be termed the 'weak' version in which it is posited that all students learn in different ways and while one student can find a activity enjoyable or an explanation clear, another student can find the same activity dull and the same explanation confusing (Pashler et al 2008:116). Also, as noted earlier, students have their own favourite ways of learning. All of this is relatively uncontroversial and probably quite familiar to most teachers. The 'strong' version however moves from this position to suggesting that not only do students learn differently, but that that difference is attributable to a certain biological difference within each student. Furthermore this difference can be reliably discovered through testing (often in the form of self-report questionnaires)  and that a teacher can then target lessons to suit the students particular LS which will in turn accelerate a student's ability to learn. In this version, LS is presented as a complete and well defined theory. This view of learning is as problematic as it is popular.  


References to LS

The weak form can sometimes be used as a wedge to introduce the strong form. Alternatively, criticism can be deflected by appealing to the more general and 'common sense' ideas of the weak form while the strong form is pushed as being a legitimate technique. An example of this can be seen in 'the practice of English language teaching' in which Harmer, despite noting that the claims of this theory have not 'been subjected to any kind of rigorous scientific evaluation'(2007:93)  suggests carrying out computer based tests of students' multiple intelligences. Harmer notes that the Coffield study has severely criticised learning styles including the following quote '[we] advise against pedagogical intervention based solely on any of the learning style instruments' (Coffield et al 2004:140) Yet continues to promote learning styles claiming that they are useful for making teachers aware  of  'self-evident truths, - namely that different students react differently to different stimulus' (2007:93). While it may be the case that learners learn differently, it does not follow that therefore LS are the answer to this. nor is it at all clear why teachers need to be made aware of things which are 'self evident'. The same tact is employed elsewhere in the section:

         It may sound as if, therefore, there is no point in reading about different learner styles at all - or trying to incorporate them into our teaching. But that is not the case. We should do as much as we can to understand the individual differences within a group (2007:89)

 Understanding that students are individuals is quite a different thing from employing specific LS tests and classroom practices. In the above quote Harmer refers to evidence based criticisms of LS and yet dismisses these criticism, without evidence, merely by asserting that 'this is not the case'.  

Thornbury is more damning of LS noting that there is little evidence 'that any of these dispositions correlates with specific learning behaviours. Nor has it been shown that preference in one area predicts success in language learning.' (2006:116-7) Yet despite Thornbury's  caution LS are presented uncritically in a large amount of EFL literature, or criticisms are brushed away, as with  Nel (2008) who after acknowledging the conceptual confusion in LS and listing criticism of LS, then goes on to suggest, similarly to Harmer, that teachers should still test their students to find out their LS in order to 'maximise the learning opportunities of their students'(2008:57). However, as the next section will show, even if we accept LS as a legitimate theory, it is hard to see how it's implementation would actually aid students learning in any way. Ellis is equally critical of the research noting that his original conclusion of their limited worth did not require updating for the second edition, published 14 years later. He also apologies for the lengthy treatment of LS noting that this merely reflected the huge amount of "attention it has received from researchers"(2008:671-2).

 The evidence problem

There is very little credible research to support LS (Coffield et al 2004:140). Specifically in the field of L2 acquisition Ellis notes that there is uncertainty over whether "any useful generalisations can be based on the research undertaken to date"(2008:669 ) Pashler et al. (2008:116)  in relation to research on the subject which reaches an acceptable level of credibility note that :

only a few fragmentary and unconvincing pieces of evidence that meet this standard, and we therefore conclude that the literature fails to provide adequate support for applying learning-style assessments in school settings... several studies that used appropriate research designs found evidence that contradicted the learning-styles hypothesis

Despite this there is a very lucrative industry built around testing and providing materials for different learner types. Another unfortunate trend as noted is, when faced with disappointing results, or criticism of LS,  some in the EFL world seem reluctant to accept them, possibly noting that there is not enough research yet or that the ideas are still, on some level, useful.

It is also worth noting that what I have mainly described (VAK) is just one version of the LS theory. Coffield et al note that 'learning style researchers do not speak with one voice' (2004:140) For example is a pragmatist the opposite of a theorist or a reflector?  It depends on which theory you choose. The proliferation of theories, all lacking direct comparability and each with their own technical terms, 'is both bewildering and off-putting to practitioners and to other academics who do not specialise in this field.' (Coffield 2004:136)

 Another issue is, as Pashler et al (2008) point out, the subject being taught may well define which approach to the content is best. Whereas they give geometry as an example, for EFL perhaps the idea of a listening lesson for non-audio learners would be sufficient to highlight the absurdness of LS. Or perhaps trying to think of ways to teach reading to non-visual learners.  But if moving around the classroom with bits of text or reading out loud comes to mind, I have to admit I despair for our profession.

The feasibility problem
Another problem with LS is that even if it were a viable theory, it is doubtful that it would be of any use to use in the EFL field. As Harmer notes (2001:90) different intelligences mean some learning tasks might 'not be appropriate for all of our students.'  He then suggests activities which might appeal to different types. There is something of a catch 22 here though which seems to remove all the supposed benefits of the approach. If we imagine a class with three learners, with LS X,Y and Z doing an activity targeted at style X will perhaps be less than optimal for Y and Z. so any benefit of supposed accelerated learning is instantly lost when we are inclusive. Any benefit gleaned from knowing the student's LS would be lost. We could divide the students up by their individual LS as we divide students up by ability now but the cost of 'interventions built around learning styles' would be huge as students would need to be tested, customized materials made and teacher's retrained. (Pashler 2008 116) And how would we know which version of LS to use in the first place?   Can we ethical test students for their optimum VAK LS while ignoring their right-brain/left brain potential?  But that's not all, some students may, according to the theory, have two LS or a mix of all 3 (in the VAK model). You maybe for example be 30% visual and 70% kinaesthetic.  If the purpose of this knowledge is to inform a teachers about the method of delivery it is hard to see how it is anything but general to the point of useless. Even if LS could be proved to be effective, the benefits would have considerable to make up for these costs. Is it not infinitely more sensible to spend this time teaching all these students language in as engaging a way as is possible? 


Students have a limited amount of time in the classroom. We um and ah over whether to schedule practice tests in class time, whether to use a whole hour for review and whether the students are getting enough practice time. Despite this we seem perfectly happy to throw away lesson time on something which is untested and probably completely useless. Pashler et al (2000:117) recommend (as Coffield et al cited earlier) due to the problems citing in this essay, use of LS at present is an 'unwise and a wasteful use of limited resources.'

 There is also the question of the massive amounts of research time that is wasted on this kind of thing. Not only the researchers' and their subjects time, but the time of the trainee teacher, keenly devouring article after article on 'getting the best out of kinaesthetic learners.' Then there are the hours wasted by the likes of Coffield et al and Pashler et al debunking these ideas only to leave us back where we started knowing no more about teaching than we did before.  This time could have been spent teaching students more language, researching better ways for students to retain language or reading articles which actually tell us something of use. Not only are we wasting our student's time but we're also making ourselves look foolish. Worse, we risk the students, who may well also be teachers, uncritically adopting these ideas and spreading them in their home nations.


Coffield F., Moseley D., Hall, E., and Ecclestone, K., (2004) 'Learning styles and pedagogy in post-16 learning : a systematic and critical review' London; Learning and Skills Network.

 Dunn, R. and Griggs, S. 1988. Learning Styles: Quiet Revolution in American Schools. Reston, VA: National Association of Secondary School Principals

Ellis, R. 2008. The Study of Second Language Acquisition (2nd ed.). Oxford:Oxford University Press.

 Harmer J. 2007 The Practice of English language Teaching Essex: Pearson Education Limited

 Nel, C. 2008. 'Learning styles'. In Griffiths, C. 2008 (Ed.). The good language learner: A tribute to Joan Rubin. Cambridge: Cambridge University Press

 Pashler, H.,  McDaniel, M., Rohrer, D., and Bjork, R. 2008 'Learning Styles: Concepts and evidence' Psychological Science in the Public Interest 9/3, 105-119

Thornbury, S. (2006) An A to Z of ELT Oxford: Macmillan 








  1. "Understanding that students are individuals is quite a different thing from employing specific LS tests and classroom practices." Thank you for this. The whole post was very helpful, but this sentence is what brought it home. I think perhaps what I've been mixing up is the difference between getting to know my students and understanding them as individuals, and what has been defined as learning styles. The concept of learning styles helps me know my students a bit better. When they can tell me that they retain information better when they see images versus just hearing it, I get a glimpse into my student's life and way of being.

    I will be taking a new approach to this in the next training course I give thanks to your research.

    Thank you,

    1. Thanks for your reply! I'm really happy to hear it made you think about changing the way you employ LS. You might enjoy this video too.

  2. Could you give the author, Russell Mayne, credit?

    1. Ah sorry, I probably didn't make it clear that i am he.

  3. This comment has been removed by a blog administrator.