Cathy Moore - Training design


Learning styles: Worth our time?

We’re wasting our time talking about appealing to learning styles when research has identified far more powerful techniques. This post summarizes two meta-analyses about learning styles. Read more

Learning styles: Worth our time?

If you had time to evaluate the research on learning styles, what would you conclude?

Here’s what four cognitive psychologists concluded: “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.”

Learning styles argumentThat quote is from Learning Styles: Concepts and Evidence, an examination of learning style research. Some more quotes from the paper:

  • Studies are weak: “Although the literature on learning styles is enormous, very few studies have even used an experimental methodology capable of testing the validity of learning styles applied to education. Moreover, of those that did use an appropriate method, several found results that flatly contradict the popular meshing hypothesis.” “Meshing” refers to changing your teaching style to match a learning style. (p. 105)
  • Variation among learners doesn’t prove any specific theory: “It is undoubtedly the case that a particular student will sometimes benefit from having a particular kind of course content presented in one way versus another…. There is, however, a great gap from such heterogeneous responses to instructional manipulations … to the notion that presently available taxonomies of student types offer any valid help in deciding what kind of instruction to offer each individual. Perhaps future research may demonstrate such linkages, but at present, we find no evidence for it.” (p. 116)
  • It might seem intuitive, but that doesn’t make it scientific: “There is growing evidence that people hold beliefs about how they learn that are faulty in various ways, which frequently lead people to manage their own learning and teach others in non-optimal ways. This fact makes it clear that research—not intuition or standard practices—needs to be the foundation for upgrading teaching and learning.” (p. 117)
  • Don’t spend time on something that isn’t proven: “We conclude therefore, that at present, there is no adequate evidence base to justify incorporating learning-styles assessments into general educational practice. Thus, limited education resources would better be devoted to adopting other educational practices that have a strong evidence base.” (p. 105)

So what are those “other practices” that would be more effective? Another, even more exhaustive study of learning styles research offers some ideas.

In Learning Styles and Pedagogy in Post-16 Learning: A Systematic and Critical Review, the authors suggest that instead of adapting instruction to one of the gajillions of learning style theories, we should build learners’ metacognitive skills and use formative assessment:

  • “Marzano (1998) reported on the largest meta-analysis of research on instruction ever undertaken. He found that approaches which were directed at the metacognitive level of setting goals, choosing appropriate strategies and monitoring progress are more effective in improving knowledge outcomes than those which simply aim to engage learners at the level of presenting information for understanding and use.” (p. 143)
  • “Black and Wiliam (1998a) … concluded from their study of the most carefully conducted quantitative experiments that: ‘Innovations which include strengthening the practice of formative assessment produce significant, and often substantial, learning gains…. The formative assessment experiments produce typical effect sizes of between 0.4 and 0.7: such effect sizes are larger than most of those found for educational interventions.'” (p. 143)

So rather than creating redundant versions of the same material, such as narrating on-screen text, we would likely get better results by helping learners structure their learning and gauge their progress, and by offering contextual feedback and any necessary reinforcement.

Like the authors of Learning Styles: Concept and Evidence, the team that produced this (huge! detailed!) study found no clear evidence supporting any of the many theories about learning styles. They fault weak methodology and the commercial nature of much of the research:

  • “Research into learning styles can, in the main, be characterised as small-scale, non-cumulative, uncritical and inward-looking.” (p. 143)
  • “A thriving commercial industry has also been built to offer advice to teachers, tutors and managers on learning styles, and much of it consists of inflated claims and sweeping conclusions which go beyond the current knowledge base.” (p. 127)
  • “Instruments to measure learning styles are weak: We therefore advise against pedagogical intervention based solely on any of the learning style instruments.” (p. 134)

Whew. That’s an unusual amount of serious text for this blog. If you want to read even more, check out the studies.

Learn more

Design challenging scenarios

Want to help people change what they do, not just what they know? Give them realistic practice with scenarios. Learn how in my four-week scenario design online course. You'll have hands-on practice, lively online sessions, and my personal feedback.

Design training that matters

My book Map It helps you turn training requests into projects that make a real difference. With humor and lots of examples, Map It walks you through action mapping, a visual approach to needs analysis and training design used by organizations around the world.

64 comments on “Learning styles: Worth our time?

Comments are closed.

  1. “Small-scale, non-cumulative, uncritical and inward-looking”.

    Yep. Sounds like a lot of practice in the learning biz – progressing on the basis of heresay and intuition.

    Reminds me of the not-infrequently-quoted US Dept of Ed Strategic plan:

    “Unlike medicine, agriculture, and industrial production, the field of education operates largely on the basis of ideology and professional consensus. As such, it is subject to fads and is incapable of the cumulative progress that follows from the application of the scientific method. . . . We will change education to make it an evidence-based field.” —U.S. Department of Education Strategic Plan 2002-2007 (p48).

  2. Good post Cathy, and I support the campaign for us to focus learning design on much more solid research based foundations.

    I guess it would be helpful to identify neglected research which the learning community currently ignore perhaps because it lacks practical applied methodology and tools. That said, there is also an element of art as well as science to the creations and delivery of an effective learning experience, which makes it hard to generalize methods.

    Do you have a view on this?

  3. Interesting. Although I admit it’s probably something of a lost battle here — “learning styles” are a compulsory part of the standards for one of the qualification our learners do!

    Perhaps the appeal is simply that it’s easier. A quick survey, which the learners generally enjoy and find interesting (most people like talking about themselvesf!) , versus the ongoing slog of trying to break down an overall, vague “get better at X” into meaningful, measurable goals…

  4. The absence of a print tool is greatly increasing my cognitive load as I have to memorize your indispensable articles!

  5. Thanks for all your comments.

    Vijay, you should be able to print my blog posts using your browser’s print feature.

    Lars, I agree that instruction is part art and science. To me it can resemble marketing: both roles want to change understanding and, ideally, behavior, and they do it in creative ways.

    However, marketing isn’t considered “good” unless it has a measurable, behavioral goal and is tested to see how well it meets that goal. For example, dedicated marketers will track all clicks on a website and use A/B tests to compare how often visitors who see one paragraph of text buy compared to visitors who see a different paragraph of text.

    I don’t see that sort of testing being used in elearning. For example, we don’t offer one prototype module with narration and one without to see how that difference affects our learners’ performance, and then base our development decisions on that test.

    Worse, most of the time we don’t measure how well our materials changed performance on the job. This would be like a marketer not bothering to check if their $20,000 campaign had any effect on sales.

    In the corporate world (which I’m writing about in this blog), all major expenditures should be justified by showing how they improve or maintain the performance of the business.

    So one of the methodologies that corporate elearning developers might consider adopting would be prototyping + A/B testing: creating two slightly different *prototype* sections of the material and comparing how the learners performed on a realistic assessment or on the job. The result of that test would help guide the development of the final material.

    The factors tested could be suggested by published research like the studies cited above, and ideally, we would share our results with others as marketers do on sites like

    Of course, that would require some additional time in the development timeline, but our argument could be that it would be better to take that extra step than to develop an entire intervention that failed to work.

  6. I may be having trouble understanding what’s being said here. Perhaps because of my learning style. I only know that very little is learned if the lesson is not tied to something relevent to the student. For example, art students can learn much more about chemistry if it is tied to the paints and art materials they use. The student has to see a reason to learn what is being taught and be able to link it to something they already understand.

  7. Maery, what you’re describing is making learning relevant and contextual, which I agree is very important. What the studies looked at are claims that we should put learners into separate categories (e.g. “visual learner” or “concrete random learner”) and modify our instruction to match those categories.

    In corporate elearning, learning styles are often used to justify presenting the same information in multiple modes simultaneously, which is assumed to appeal to many learning styles at once. This redundant media use goes against recommendations from other research that suggests it can interfere with learning.

  8. Printing from browser prints everything on the page, not just the article text, and so wastes several times the ink otherwise required. I use acrobat to make the pdf of the article but converting it into a hard copy results in the same waste.

  9. Dr. Will Thalheimer has an interesting related take on his site rebuking the premise about a certain percentage of people remembering X% of what they see, Y% of they hear, etc. It is available here

    No one is exclusively one learning style or another, so I think the idea of variety might be as critical to learning as anything else. My major learning style is “please don’t bore me”.

  10. Hi Cathy,

    Thanks for the summaries.

    I’ve even seen school information management systems that contain data fields so schools can record whether students are Visual, Auditory or Kinaesthetic learners!

  11. Veda, thanks for the link to Will Thalheimer’s site. I encourage people to look at all of his publications for plenty of myth-busting information.

    Mark: yikes!

    Vijay: Instead of creating a PDF, you could just select the text you want to print, paste it into a text doc, and print that. The quickest way is to click at the beginning of the text, shift-click at the end, copy & paste. This is a WordPress blog and there appear to be no currently functioning and supported “print this post” plugins. I’m also reluctant to take the time to code a special print function for a blog that often contains Flash interactions and videos, but if many other people ask for that function, I’ll reconsider it.

  12. I see any “learning styles” classification as more of a useful brainstorming tool / framework for a qualitative check when deciding on approaches. (Well some of them are useful anyway!)

    E.g. using a Theorist/Pragmatist/Activist/Reflector (or whatever) classification can be useful in order to just qualitiatively look at and discuss the solution being developed. We’re asking the question – will this product work for the different kinds of people in the client’s organisation, or not? If not, what more variety can we provide within the budget?

    IMHO proving theories is neither here nor there as long as the end product does what it’s supposed to do.

  13. Vijay,

    One of the ways to get around printing everything whether using the browser print option or creating a PDF, which still includes everything is to use the “printfriendly” option from the “Share” plugin. It can be downloaded from at: or through the Google Toolbar. Another plugin that has the “printfriendly” option is “Add this” which is available for downloading at:

    You can also go directly to PrintFriendly and add it to your bookmarks at: Cathy could also get a button for her blog from this site that would do the same, if she wanted to.

  14. Thalheimer’s commentary is an eye opener, not just to what can happen in the training industry but how things can be accepted as fact that we hear and read in the News. With so much information coming at people and so little time to truly look into the facts to back it up, this is a little disconcerting. But a good reminder on being careful to check citations and studies before passing information along.

  15. Connie, thanks to the links to print-friendly add-ons. I’ll check them out.

    Another way to quickly create a print-friendly version of any site: Use the Instapaper “Read later” button. This has the additional advantage of creating a text-only archive of all the stuff you might want to refer to later. But, again, the Flash and videos I include on the blog won’t end up in the text version.

  16. Formative assessment involves assessing the learner and then using the results to adjust our instruction. In elearning, this could mean having the learner complete an activity and, if their answers show that they don’t get something important, providing additional information or help that other learners won’t necessarily see. Most elearning tools that I’ve seen can handle this type of branching.

    Simple example: Bob, a pharmaceuticals salesman, is completing a module on his employer’s latest drug. In one activity, he needs to help a fictional doctor understand when to use the drug. He incorrectly says that it’s good for diabetics.

    Our program notices his incorrect answer, highlights the aspects of the drug that make it bad for diabetics, and has Bob repeat the activity. If we already covered this material earlier in the module, we offer it to Bob again in a different way (different media, different way to explain it, whatever) because obviously Bob didn’t get it the first time.

    Ideally, we’re also capturing this assessment information for all our sales trainees so we can see patterns and revise our materials to make them more effective.

    More complex example: Bob completes several activities in the module. His answers throughout the module suggest that he doesn’t thoroughly understand endocrine disorders. Our program sends him into an optional module about endocrine disorders that also includes activities that help him see how his understanding is improving.

    This ERIC summary describes several classroom techniques that could be adapted for synchronous online training or async materials that include discussion forums.

  17. I saw a site a few months ago that noted a reward that was being offered if anyone could produce conclusive evidence of learning styles benefiting learners. It was a fairly substantial reward, and it had remained unclaimed for years.

    Did anyone else see that? I scanned my delicious tags and concluded I hadn’t saved the link (I was hoping to share it).

  18. @ David,

    It’s Will Thalmeir’s site – I think he first offered the US$1000 cash in 2006.

    Apropos of not much, apparently evidence isn’t generally enough to change behaviour – tertiary education still features lectures, despite the evidence against their instructional value being widely available for at least ten years.

    And the corporate world is still happy to tout them as important design considerations. Training Tech Talk recently featured Coca Cola’s sales training simulations, with the vendor claiming they were designed with learning styles in mind – as published in Training Tech Talk recently. My rebuttal and suggested alternative approaches, citing the same article that Cathy headlined with, was published last week in the same newsletter.

  19. Here is a systematic literature review from 2004 on most of the instantiations of Learning Styles in teaching and learning. This is my favorite thing to give to any proponents I come across. It speaks for itself. 😉

    “Learning styles and pedagogy in post-16 learning. A systematic and critical review”

    In searching for the latest download link, I found it referenced in the wikipedia article on Learning Styles. It has a nice historical overview and is a fine source for more references.

    Liz Dorland – @ldinstl_chimera

  20. I do want to push back a bit on the statement from the US Department of Education. Don’t forget who had just taken control of the US government in 2002. The NCLB/No Child Left Behind nightmare and the very conservative research agenda of their “new” IES – Institute of Education Sciences came out of the thinking of some of these folks. As did the “blame the teachers” meme.

    “US Dept of Ed Strategic plan:

    “Unlike medicine, agriculture, and industrial production, the field of education operates largely on the basis of ideology and professional consensus. As such, it is subject to fads and is incapable of the cumulative progress that follows from the application of the scientific method. . . . We will change education to make it an evidence-based field.” —U.S. Department of Education Strategic Plan 2002-2007 (p48).”

    Just like with Learning Styles, there is a surface interpretation and then a deeper level of understanding of the complex issues. The question of “what constitutes evidence” has been hotly debated, and the work of a lot of prominent educational researchers was marginalized in the years since then. I personally spoke with Dept. of Ed employees who did not feel free to give their names for fear of losing their jobs. They told me horror stories about those who had been driven out or demoted. The disbanding of the ERIC database wiped out access to whole bodies of research.

    It’s too complicated to explain well here, and I’m learner/observer rather than an expert when it comes to educational research. I did teach college chemistry for 35 years and in 2003-04 was a Program Director in the Division of Undergraduate Education at the National Science Foundation, so I observed these debates first-hand. The NSF education folks had a very different viewpoint from the new regime at the Dept. of Ed, and there were some definite ideological clashes.

    Do keep in mind that like Learning Styles, simplistic statements about complex issues — i.e. “application of the scientific method” — can be very seductive!

    Thanks for a thought-and-discussion-provoking post.

  21. Liz – thanks for the views on the US Dept of Education quote. The political nuances and associations with the NCLB agenda are less apparent from the distance of the east coast of Australia – however, I sympathise with your views on the unfortunate policy directions and their effects.

    The quote resonates with me after working with a wide range of public and private clients, who frequently made decisions concerning policies, acquisition and administration of training that had little or no relation to readily available, accessible evidence that could contributed to better outcomes.

    Agree that the term “evidence-based” is problematic – have been working with neurologists and other medical specialists recently, and find it creates similar difficulties across other disciplines.

  22. Thanks for the continuing discussion. You’ve helped me find a way to clarify my main concern, which is this:

    I often hear instructional designers say, “We’ve put text on the screen for visual learners, and we’ve had a narrator read it aloud for audio learners.” Or, “We should do this fact check as a drag-and-drop for the kinesthetic learners.”

    This approach continues to treat instruction as a presentation with a quiz. It’s fun to use lots of different media, and “learning styles” as commonly interpreted give us that excuse. And it’s easier to tack some narration onto boring text than it is to completely overhaul the material so it teaches through contextual activities and inspires independent, critical thought.

    There may be other learning style theories that don’t encourage such surface applications, but the only one I hear about from corporate IDs is the VARK model.

  23. My TAFE students regularly ask for material to be presented in multiple formats – text, images, video They are not influenced by learning theory – just aware of what works for them. I am always happy to oblige, and it does seem to make a difference to their learning experience.

  24. Providing visual resources, text, and sound has nothing to do with learning styles. Everyone needs to be able to learn from all of those when appropriate. Some concepts are visual, and not providing images, animations, videos, is just dumb. We used too much text for way too long. Having the students create some of their own explanations and to explain to others is even better.

  25. I don’t think introducing material in multiple ways has nothing to do with learning styles. Students need a way into new materials and will then take that material and make it their own by chewing it up and refashioning it. The point of the multiple methods of presentation is to get them chewing in the first place, to take that first byte (pun!) into new territory. If they won’t/can’t engage with the material initially, they just sit there starving.

  26. There seems to be an assumption in this discussion and the one in my previous post on audio that “don’t put faith in learning styles” = “it’s fine to put a textbook on the screen.” That’s not what I’m saying, and it wasn’t the conclusion I got from the studies.

    Obviously we need a wide range of formats and styles when presenting information. This is true because different content clearly works better in different formats. For example, a map of Argentina communicates its basic geography more clearly than do paragraphs of text.

    Different formats are also useful because learners clearly have preferences. However, the studies cited above make clear that there are many conflicting and unproven ideas about how to categorize or respond to those preferences, and no one brand of “learning styles” has been shown to be useful. There’s even debate about whether we should match someone’s alleged learning style or make them “grow” by having them work in a different “style.”

    However, presenting information is only part of what I think instructional designers should be doing. My concern is with corporate instructional designers who emphasize surface learning styles (presentation preferences) to the detriment of more proven methods of instruction that take longer to design. For example, I’ve seen designers add redundant audio due to “learning styles” instead of creating challenging, contextual activities or following up after the learning event to reinforce learning.

    Too often we view elearning as information presentation and neglect deeper approaches. Again, I’m talking about corporate elearning here, which is often expected to do the entire job on its own in one standalone 30-minute “course.” This is far different from longer, ongoing instruction that includes contact with the instructor, personal feedback, discussion groups, etc.

  27. I would agree with your thoughts here Cathy but I think there is a complete lack of political will to do anything that isn’t expedient and cost-cutting. We are always being squeezed to “do” more with “less” because funding is always an issue and I am not really sure that those shaping our society on the macro scale really want a well-educated critical public anyhow, despite their rhetoric.

  28. Having used some of the ideas behind learning styles theories, I have been curious about the research. What I see in what you have summarized, (thank you!) is that the application of these theories has some major flaws and justifications.

    Basing my own theories on what I’ve seen work in my classroom there are a few things that I find key in applying learning styles.

    1. Using a simple learning styles quiz (even with all the flaws) allows me to begin the conversation of metacognition with my students. (I’ve done this with 6-12 students, but I supposed some adult learners may need a tool to begin the conversation as well.) Once students have some language to talk about their learning they can move on to increasing their “meta-cognitive skills”. The key being able to talk about learning.

    2. Learning styles can change over time. The comment about a learning style listed in student database makes me cringe People change all the time, isn’t that what learning is all about.

    3. Teachers tend to teach in their preferred style. Information needs to be presented in multiple ways. As teachers stretch to include multiple learning styles, the lesson should become deeper. Not like the example of adding a narrator and a drag-and-drop interaction. Being able to discuss different learning styles and pushing my students to incorporate different styles in their own presentations increased the student involvement and the presentations.

    4. Learning styles is simply a theory – that needs to be further researched. And as mentioned in the second article the “Research into learning styles can, in the main, be characterised as small-scale, non-cumulative, uncritical and inward-looking.” (p. 143). So I say – do better research! As a teacher there is some intuitiveness about things that work. As educators we need to do better at using that intuitiveness to create better research studies to ultimately improve teaching.

  29. Thank you for igniting such a great discussion, and thanks to all for thoughtful comments.

    I think the issue is the amount of faith put into the concept of “learning styles” which is a sacred cow of the learning industry despite the fact that anyone arguing the point should be aware that the theory never had solid grounding.

    I trust there is something to learning styles. But the true issue was obvious to me when I recently set up an entertainment system. Youtube videos on the process got me further than documentation. Nothing to do with preference, but task. Video could provide more detail and mirrored application environment more effectively than text (and I am a reader- I would have preferred to be able to get it from the provided documentation from the pros).

    Simpler example: you won’t learn to swim with competence from a textbook or lecture or online course: the task, not user preference, is the driving factor.

    Cathy’s recommendation seemed clear- look at the task and what modalities will be create the opportunities for the person to learn and apply what they have learned with all the needed extras, like follow-up for transfer, etc… (I never took her counterpoint to learning styles as an argument to just create “the text under glass”).

    I also agree that often too much effort is spent on the presentation layer of training itself and creating new presentation modes of the same tired info instead of creating great learning and support in the appropriate modes (Obviously, 508 for those required, is an exception). Seems that too many people are busy arranging deck furniture to notice the Titanic is sinking.

    This is not to say there aren’t learner preferences having some influence on presentation, but efforts based on the theory are often taken too far. The energies put into presentation often take away from the energies that need to be put into better learning design before presentation is even considered.

    The “learning style theory” issue reminds me of another theory that is grossly misunderstood and applied inappropriately in the workplace- the Mehrabian Myth. See the bust here:

  30. Ms. Moore, I recently discovered your blog and am in love with it! Thanks for sharing all your fantastic insights!

    I’ve long suspected that learners cite learning styles based on time constraints, attention span or motivation issues. In grad school, I think people often called themselves ‘visual learners’ so that they would have to deal with a single diagram or mind map encompassing key points instead of an entire article. Or ‘audio learners’ so they would need to listen to a podcast or audio recording that lends itself well to multi-tasking or listening on-the-go. Compare this with reading a textbook or article – I am yet to encounter a single self-described ‘text learner’, even though text is usually the default. My opinion is that ‘learning styles’ = a cry to make content quick and nippy, one way or the other.

    I’ve never been a fan of narration but I find that I am invariably asked to incorporate audio and visual/text redundancy within a course for accessibility, independent of any formative assessment (since our courses are free and open to the public). We receive the occasional complaint when we don’t do so…

  31. A follow up. It is sad that we think science can help in learning but turns out a lot of time it will point you to the wrong path until it is too late…maybe one day we will get there but I am not sure if those mistakes really worth it.

  32. I’ve read many studies for and many studies against ‘learning styles’.
    Most learning styles talk about preferences to aid engagement which in turn can aid memory retention. Colby, Honey & Mumford and VARK all state emphatically that most of us have multi-modal preferences and few of us have one clear-cut all defining learning style.

    These three theories all recommend learning professionals utilise a variety of styles to actively engage learners to assist their learning process.

    There are certainly pro-learning styles studies out there that are poorly constructed and do tout the ‘one learning style preference’ myth.

    However, we are human beings, we have different taste preferences, different lifestyle preferences – yet many would claim that we have no preferences when it comes to how we learn. I have worked in Vocational training for over 13 years and before that as High School teacher. My experience is that people have learning preferences.

    That doesn’t mean that just because I hate lectures that I can’t learn from that delivery method. It just means that I’m more likely to get bored, tune out (and in my case) less likely to show up.

    Learning styles is more about learner engagement and the real lesson for learning professionals is that your learners have different preferences to you. As someone earlier in the post stated ‘don’t bore me’ and I’m more likely to get more out of your learning intervention.

  33. I love the idea of A/B testing prototypes. I wonder what your thoughts are on how to apply this idea to a recent design you created? How would you do it? How would you prove as irrefutably as is possible that design A was more effective than design B in this recent design of yours? I know one weakness that I have is that I’m asking the wrong person/people to go through the A/B test. Sure, its important to get the stakeholders blessing on the design, but the stakeholder isn’t the target audience!

    I’m thinking about a recent design of mine. It was on presentation skills. How could I have A/B tested my design. Here’s an idea:

    Give the learner an outline of material to present and then have them present it before going through any training. Specifically observe how effective they were at verbally communicating this portion of material. Use a Likert scale to rank them across the main factors of effective verbal communication (Tone, Pace, Inflection etc.) I suppose this would be the control. Then create a prototype that allows them to practice verbal communication in some meaningful way. Next, have the learner present different material. Did they do any better? If yes, test A = on the right path. If no, try test B.

    Sounds like a fascinating learner-centric approach for coming to a more effective learning experience. But, I have a sneaking suspicion that most clients wont understand. They’ll think you’re trying to boil the ocean to come up with a design.

  34. Dan I like your idea of measuring actual ability as part of an A/B test. For example, if the prototype is designed to teach a particular presentation style (e.g. Beyond Bullet Points), your before-and-after tests could measure how well the person applies that style.

    So their “before” presentation might be a lot of slides with bullet points, and your test could measure how well their “after” presentation on a different topic applied the new approach.

    It seems like the strongest test would look for a change in direction in the learner’s behavior. It might be harder to measure simple improvement in a skill, because the testing itself gives the learners practice in that skill. Their second attempt could be better than the first simply because the first round gave them some practice.

    I agree that clients would probably wonder why we want to do this testing, but if they care about ROI (or if they have marketing experience) they would probably be able to see the benefits. Testing a prototype can avoid a lot of wasted time.

  35. I remember my first formal exposure to the VARK model while in training to be an instructor. We spent several hours talking about learning styles, at the end of which the facilitator conducted an exercise to identify the learning styles of all of the participants.

    Each participant stood to tell what their learning style was and to relate an example of past learning that supported or refuted that style designation.

    When it was my turn, I stood and said “I’m a kinetic learner… We have spent alot of time on this. Can we move on?”

    The facilitator was not amused:P

  36. Certain Frustration
    My field is English instruction to non-native speakers. (EFL, ESL, etc.). Most research that actually compares two methods was done ages ago. So too are methodological proposals.
    I have research and recommendations against instruction in grammar from Sweet (1900); Britannica (1911); Charters & Miller (1915) and a Harper’s Magazine article (Johnson, 1916) that is apt but not scholarly. There are more recent studies, though mostly proposals, but nothing related to eLearning.
    I did a qualitative study into why learners preferred F2F over DE and one interviewee had “tried the rest. I tried learning…Japanese…from the internet and it didn’t work. And I tried learning Italian with a…cassette system with books and that didn’t work either.” Both were grammar based and essentially copies of tried but untrue, ancient F2F systems.
    I cannot do a thesis on comparison of the two. It’s impossible to remove grammar without completely revamping all components of the course. Since my method would use bits and pieces from many methodologies, the control group must include learners in all those using grammar. You can imagine the size of this control group.
    So it looks like research will never be done. I suspect this problem is transferable to many potential comparison studies in most fields.

    Britannica. (1911). Retrieved from,
    Charters, W. & Miller, E. (1915). A course of study in grammar: Upon the grammatical errors of school children of Kansas City, Missouri. The University of Missouri bulletin, 16(1). Columbia, MISS: University of Missouri. Retrieved from
    Johnson, Burges. (1916). Grammar, the bane of boyhood. Harper’s monthly magazine, 134, 122. Retrieved from
    Sweet, H. (1900). A new English grammar: Logical and historical; part 1; introduction, phonology and accidence. Oxford, UK: Clarendon Press. Replica by Adamant Media Corporation, New York. Retrieved from,M1

  37. @John

    I recommend that you look for research in English countries that have high long term and ongoing immigration from non-English speaking countries such as the UK and Australia. Both these these governments invest heavily into education research at both K-12, Adult Education and VET (Vocational Education Training). The following are two Australian websites that publish research:

    Australian TESOL (Teachers of English to Speakers of Other Languages) Council:

    The Australian National Centre for Vocational Educat Training (NCVER):

    The Emerald database is also a good place to search for articles and books from around the world on ESL / TESOL online learning methodologies.


  38. Cathy, first I want to thank you for posting the mega-study on learning styles and pedagogy by Coffield and his colleagues. The first thing I did when I opened it was flip to the 4MAT section, as this is a typology recently implemented in my work setting to assist new trainers in understanding learning styles. I enjoyed reading the methodical debunking of the value of this process from an application perspective.

    Essentially, we do this block of instruction on learning styles, they take the 4MAT and enjoy gratuitous self-analysis, and then what? I have seen this: they play with these ideas, have a “wow, that’s interesting” experience and then forget the entire process almost immediately. They certainly do not seem able to apply this understanding, especially in the initial period of training others.

    Yet, understanding learning styles is perceived as such an important part of train-the-trainer ideology. I enjoyed reading another well-annotated perspective.

  39. I have been a devoted follower and believer of the different learning styles, until reading this blog. Recently I have been introduced to the Cognitive Information Processing Theory which focuses on the specific things that are going on in the head of the learner. My beliefs are shifting and now focusing on the motivation of the student versus the learning style. When students are motivated, they become active. An active student seeks and processes information which in turn causes learning to take place.
    It is the meaningfulness of the material that sparks learning. When a student can relate “new information to previously acquired knowledge, and organize knowledge to make it meaningful” (Learning Theories and instruction pg. 50) retention takes place.
    Learning style test are entertaining, but I agree with an earlier post concerning how to apply this to retention and retrieval. The most important questions to an instructional designer should be how to move information from short term memory to long term memory, and how to create instruction that makes the material “stick” in the learners mind. Once these questions are answered and the answers successfully implemented, the student has an expanded base in which to draw motivation and meaningfulness.

  40. My thinking of learning style left-right brained thinking have me re-evaluating my teaching style. Normally, I observe students, checking if they are learning and how they are learning. I have always observed learning styles to reach my students. Now after researching learning theories and how the brain works, my teaching is now gravitating towards learning theories and instructional methods.
    The blog “Learning Styles: Worth Our Time brought up some interesting points. For one the blog stated that learning style studies are weak. I disagree, because in my observations of students I observe the best way to reach my students. For example, in teaching Social Studies I was teaching concepts of North Africa. In the textbook there was a picture of an 800 year-old church in Lalibela, Ethiopia. The church was carved from solid rock entirely below ground level. As always, I pulled down the map (too bad I could not have taught this class using multimedia tools) to locate Ethiopia which is in North Africa. A week later I brought in a photography that I have in my living room that my brother-in-law took of the church. The picture was taken looking down at the church, so you can see a cross on the roof. The students were able to have a discussion about Ethiopia and the church. My observation of this lesson said that the students connected with this lesson because the lesson was visual and auditory.
    The blog stated that we should build learner’s metacognitive skills and use formative assessment. As a teacher I must have an environment to help the learner develop self-regulation skills and at the same time have formative assessment. Hopefully by the end of this class I will be able to incorporate many new skills in my classroom.

  41. @LG – I think your interpretation of learning styles differs from the intent implied in the studies. The oft lamented learning styles imply that each learner has a specific learning style to which you should tune your methods.

    In your case, your tuning your methods for conveyance effectiveness, regardless of consideration for individual styles. This, in my opinion, is the right way to do it. It’s media and method selection based on a mix of theory and experience. And it works, or you’d adjust it:)

    The studies imply that accommodating the styles of individuals is futile for a range of reasons –

    It’s better to tune methods to most effectively deliver the lesson, build the skill, etc… to your audience using an aggregate of audience properties. Considering the whole audience and matching that to one set of methods (or a multi-level set of methods if some students need additional examples / representation / practice), vice attempting to play whack-a-mole with VARK style needs.

  42. Cathy, thanks for sharing the Learning Styles: Concepts and Evidence study. While I’m not completely convinced that we should move away from differentiating instruction to accommodate various learning styles, I am inclined to believe that we need to spend more time advocating and developing metacognitive strategies with our students.

    The research in support of teaching people how to think about their learning is compelling to say the least. The Cofield and associates study from 2004 demonstrates this clearly. I also came across a Dutch study involving nine schools and hundreds of students where experimental groups were trained in metacognitive strategies relative to reading comprehension. The experimental groups outperformed the control groups with large gains that were sustained when the same students were tested in reading comprehension the following year (Houtveen & Van de Grift, 2007).

    Houtveen A.A.M., Van de Grift, W.J.C. (2007), Effects of Metacognitive Strategy Instruction and Instruction Time on Reading Comprehension, School Effectiveness and School Improvement, Vol. 18, No.2, June, 2007, pp. 173 – 190

  43. @Steve: a direct quote from the VARK website:

    ‘Teachers, and others who assist with learning, should use a variety of modes if they expect to reach every learner.’

    We need to be aware that there are many odioous studies that purport to be about learning styles and preferences. I have yet to read a study that actually tests learning preferences – including VARK – in the same way the theories recommend that they be used.

    VARK, Honey & Mumford and Colby – all recommend multi-modal style learning methodologies. Not choose one style and ‘whack them’ with it. Nor an ‘A’ vs. ‘B’ methodology as others have discussed.

    None of these three theories suggest that learners can’t learn from a style that isn’t their preference. I suffered through several years of lectures at uni and still passed with distinction – but I hated those lectures passionately (and slept through more than one).

    In my experience, a multimodal approach to learning works best for groups of students. I would futher argue that Cathy Moore – based on the samples she posts – clearly employs learning theories in her design.

  44. Thanks, everyone, for continuing the discussion.

    I’m not sure what Jennifer means by saying that I employ “learning theories” in my design. I don’t subscribe to any particular theory. What I do is learn what I can about the audience and what they need to do in the real world. Then I help them practice what they need to do in a way that they have shown that they prefer.

    For example, when I worked on a cross-cultural communication project with the US military, our team visited many military bases to interview soldiers in depth about their cross-cultural duties and their training preferences. We also observed how soldiers responded during several hours of classroom instruction.

    From our research, it became clear that the audience strongly preferred group discussion over solo elearning, although we were technically supposed to design elearning for them. As a result, our online materials were short, game-like activities that were designed to prime learners for group discussion or that could be used during a discussion. The activities raised issues; they didn’t present information.

    So yes, we tailored our materials to the learners’ preferences, which I think everyone should do. However, we learned about those preferences by talking with our learners about their job roles, their attitudes about the topic, their feelings about their current training…

    I don’t think that a VARK assessment would have persuaded us to use the design approach that we ultimately chose. It might have suggested that our learners were visual or kinesthetic, but that could inspire some designer (not my client!) to, for example, create a forgettable arcade-game fact check about Afghan culture instead of the branching scenario we created.

    Again, I’m not saying (and I don’t think the researchers are saying) that different modes shouldn’t be used. As I’ve said before here, obviously some content is easier to understand in one mode vs. another (e.g. a map vs. paragraphs of geographical description).

    It’s also obvious that people have different preferences. I personally would much rather look at a self-explanatory image than read text. I also prefer to learn new music by ear rather than reading it from a score, which flatly contradicts the VARK assessment that gave me a fat 0 for audio learning.

    What concerns me is what I said in a previous comment: I often hear instructional designers say, “We’ve put text on the screen for visual learners, and we’ve had a narrator read it aloud for audio learners.” Or, “We should do this fact check as a drag-and-drop for the kinesthetic learners.”

    This approach continues to treat instruction as a presentation with a quiz. It distracts designers from the harder work of creating contextual activities that inspire independent, critical thought.

    As designers, we need to know what our learners need to do on the job, why they aren’t doing it, what tools and training they have now, what they think about those tools and training, etc. Choosing media is a very small part of our job but “learning styles” as I’ve seen them applied in corporate elearning elevate media selection beyond all other concerns.

    Again, I’m talking about short-term training for adults who are performing a job. My blog doesn’t address higher education or K-12, where the goals and timelines are often different.

  45. my main objection to discussions re “learning styles” in recent years is that they’ve often been interpreted/used in a way that is indistinguishable from astrology instead of seeing them in relation to the learning cycle (which is what i think honey and mumford and kolb might have intended) and how all steps in the process might play a role in learning – in fact, i would argue that catering to a learners WEAKNESSES would play a more important role in learning than serving their supposed preference.

    in my view it would be a pity if we threw them out altogether however throwing out the simplistic applications would be a good first move.

  46. Cathy,

    As an educator and instructional designer, I find your blog entry on learning styles extremely interesting. Your summary of the article “Learning Styles: Concepts and Evidence” by Pashler, et all (2008) inspired me to read the entire text. You did a nice job highlighting the main conclusions drawn from this article. I agree with Pashler, et al. (2008) that studies on learning styles need to be much stronger.

    I found a great article titled “Perfect Learner: An Expert Debate on Learning Styles” that provides further information concerning the flaws of research and studies on learning behaviors. In this article, Martin Delahoussaye (2002) claims that more research needs to be conducted on the specific links between different learning styles and various methods of learning (activities). Delahoussaye (2002) gives the following suggestions as to how studies on this topic can be improved:

    1. Avoid over-generalizing learning styles based on the measurement of a single construct.

    2. Use various instruments to assess learning styles on multiple occasions.

    3. Implement multiple measures of behavior change

    4. Consider interacting variables such as gender, IQ, ability or initial capability in target behavior, time on task, and teacher expectations

    Aside from considering the research that exists on learning styles, there is one important question that must be asked. The question that educators in every field must consider is: Should we adapt our teaching to reach each learner’s preferred style? After reading the articles by Pashler, et al. (2008) and Delahoussaye (2002), among others, I believe the answer is no…

    Any good educator includes multiple methods of content delivery and learning activities into his/her instruction. It is important to design curriculum so that there are a variety of ways that students can learn the material and become engaged. When learning and engagement does transpire via specific learning activities, a student can “stretch his/her learning capability in other learning modes” (Delahoussaye, p. 4). Not only is teaching individual students according to their preferred learning style nearly impossible, educators would be doing a disservice to all students by stereotyping learners based on one or two different “styles.” Educators need to expose all students to all kinds of learning approaches to create well-rounded and flexible learners that can adapt to any learning situation.


    Delahoussaye, M. Perfect Learner: An Expert Debate on Learning Styles. (2002). All Business. Retrieved November 14, 2010, from URL

    Pashler, H, McDaniel, M, Rohrer, D & Bjork, R. Learning Styles: Concepts and Evidence. Psychological Style in the Public Interest. 9, 103-119.