Learning styles are a myth: what to do instead

Learning styles are real in one sense only: people will tell you, consistently, how they like to take in information. What the research does not support is the step that matters, that you learn more when teaching or study is matched to that preference. Calling yourself a visual learner describes a taste, not a better way for you to learn.

This guide sets out what the main studies found, each in its authors' terms and with its limits, what VARK says about itself, and what to do instead. It is plain about the verdict and not unkind to anyone who believed it: most of us were taught it.

EdCitation publishes this guide. Every paper was read in full in September 2026 except one, read as an abstract (we say which). The studies were found with EdCitation's Find sources, each reference was built in Cite a source, and the list was checked in Verify references. For which study methods work in general, read how to study effectively: what the research shows.

Are learning styles real?

Learning styles are real as stated preferences and unsupported as a method. Pashler et al. (2008) drew that line. They accepted that people, if asked, will say how they prefer to receive information, that the questionnaires recording this give fairly stable scores, and that people differ in specific aptitudes, such as reasoning with spatial material.

What they tested was the claim that diagnosing a learner's style and teaching to it produces better learning. The journal commissioned them to judge the evidence, in their words, "as dispassionately as we could".

What the meshing hypothesis says

The meshing hypothesis says instruction works best when its format matches the learner's preference: diagrams for a "visual learner", talk for an auditory one. It is the most common version of the learning-styles idea, not the only one; Coffield et al. (2004) counted 71 models.

What evidence would prove learning styles work?

Learning styles would be proved by one kind of experiment, which Pashler and colleagues spelled out.

1. Sort learners into groups by learning style, using a stated measure.
2. Randomly assign learners in each group to different teaching methods.
3. Give everyone the same final test.
4. Find a crossover: the method that works best for one style must not be the one that works best for another.

The fourth step is where claims fail. If visual and verbal learners both do better with diagrams, diagrams are simply the better method for that material, and nobody needs sorting.

What Pashler and colleagues found

The review found virtually no evidence of that crossover. Very few studies used a design that could test it; the authors found "at most one arguable piece of evidence" for the hypothesis in general, and none for the tests then marketed to teachers.

Several studies with the right design contradicted it. In Massa and Mayer's computer-based electronics lesson, help screens matched to a learner's verbal or visual preference gave no benefit across nearly 20 measures of style. In Cook and colleagues' study of 123 internal medicine residents, the predicted match between style and the order of teaching did not appear. We read these two as the review describes them, not in the originals.

Where the review stops

The authors wrote that it would be an error to conclude every version of learning styles had been tested and failed; many had never been tested at all. They also agreed that the best method varies by subject: writing needs words, geometry needs diagrams. What they found no support for was sorting students into types and teaching each type differently.

What did the big review of learning-styles models find?

The largest review of the models found that most questionnaires fail basic tests of quality. Coffield et al. (2004), a report published by the Learning and Skills Research Centre on post-16 education in England, identified 71 models and 3,800 references, and examined the 13 it judged most influential.

Each was checked for independent evidence of four minimum standards: internal consistency, test-retest reliability, construct validity and predictive validity. Only Allinson and Hayes' Cognitive Style Index met all four; Apter's and Vermunt's models met three; Dunn and Dunn, Gregorc, Honey and Mumford, and Kolb met one. On matching, the authors called the evidence "equivocal at best and deeply contradictory at worst", citing a review that found nine studies favouring a match and nine favouring a mismatch.

The limits: it covers UK post-16 education, dates from 2004, and judges the instruments rather than every classroom use. VARK is not one of the 13 models it examined in detail.

Do visual learners learn better from pictures?

Visual learners do not learn better from pictures than other people do, in the studies that tested it directly. Two tested the popular versions.

Listening against reading

Rogowsky et al. (2015) followed Pashler's design with 121 college-educated adults aged 25 to 40. An inventory classed them as auditory or visual-word learners; 53 fitted neither. In the main test, 61 were randomly given a nonfiction chapter as an audiobook or as e-text, then tested at once and two weeks later.

There was no significant interaction between learning style and format at either test. Measured listening comprehension predicted how much people learned; stated preference did not. The limits: the four groups compared had ten or eleven people each, every test was written (a limitation the authors name), and the task was verbal comprehension only.

VARK and an anatomy course

Husmann and O'Loughlin (2019) gave the VARK questionnaire and a study-strategies survey to 426 undergraduate anatomy students in 2015 and 2016. According to the abstract, most students did not report study strategies that matched their VARK results, course performance was not correlated with any VARK category, and studying in line with one's results had no correlation with the grade. Some strategies, such as using the virtual microscope, went with better grades whatever the VARK result. We read only the abstract; the paper is paywalled. It is a correlational study of one course, not an experiment.

The 2024 meta-analysis that found a small effect

A fair account includes the other side. Clinton-Lisell and Litzinger (2024) pooled 21 studies, 101 effect sizes and 1,712 participants comparing matched with unmatched teaching, and found an overall benefit of matching (g = 0.31). But only 26% of outcome measures showed the crossover the hypothesis needs, study quality was low, and the authors concluded the benefits were too small and too infrequent to justify widespread adoption.

What does VARK say about itself?

VARK says it measures preferences and has never claimed that teaching should be matched to them. Its page, published by VARK Learn Limited (n.d.), quotes its creator, Neil Fleming, calling it "a catalyst for reflection", and says the matching hypothesis is "not a claim that VARK makes". It lists what VARK is not, including a fixed label and a prescription for how teachers must teach.

The page also says VARK works because it encourages active learning, such as drawing diagrams and explaining ideas aloud. None of the studies we read tested the questionnaire for that effect; the nearest is Husmann and O'Loughlin's, in which studying in line with VARK results was unrelated to grades.

What does the evidence say, claim by claim?

The evidence supports the idea that people have preferences, and nothing beyond it.

Why do so many people believe in learning styles?

People believe in learning styles because the parts they can see are true: learners have preferences, and one student can grasp what another did not. Pashler and colleagues say teachers are right to notice that, and wrong to leap from it to sorting students into types.

Newton and Miah (2017) surveyed 114 academics in UK higher education: 58% agreed that individuals learn better in their preferred learning style, 64% said they try to organise teaching around it, and 33% had ever given students a learning-styles questionnaire. Shown the lack of evidence, 32% said they would keep using learning styles. It was a convenience sample, as the authors say; they compare it with an earlier survey, which we did not read, in which 93% of 137 UK schoolteachers agreed.

What should you do instead of studying to your learning style?

Instead of studying to a learning style, use the methods that help nearly every learner, and fit the format to the material rather than the person. This routine is ours, built from the studies above; no study tested it as a whole.

1. Test yourself. Close your notes, write what you remember, then check. Dunlosky et al. (2013) rated practice testing high utility; the leader guide has the detail.
2. Spread your study out. Return to each topic after days, not minutes; distributed practice was also rated high.
3. Pair words with a picture where the material allows. Draw the process, label the diagram, then explain it in words.
4. Choose the format by the subject. Anatomy wants diagrams; a poem wants its words.
5. Keep your preference as comfort, not a limit. Listen to the podcast if you like, then test yourself; never skip the diagram because a quiz called you auditory.
6. When you write about learning, cite the study, not the myth. Find the papers in Find sources and build each reference in Cite a source.

Dual coding, done properly

Dual coding means learning with words and images together, and it is for everyone, not only "visual learners". Clark and Paivio (1991) argue that memory keeps verbal and imagery codes separately, so that having both gives an additive benefit over words alone.

Its limits: Dunlosky et al. (2013) rated imagery for learning from text low utility, because the benefits were largely confined to material easy to picture and to memory tests. Draw the diagram when the subject has a shape; do not force a picture onto an abstract argument. Drawing it from memory and then checking adds self-testing; that pairing is our suggestion, not a tested method.

What if one format really is easier for you?

One format can be easier for you, and the research treats that as an ability, not a preference. Clark and Paivio describe a 1981 study by Riding and Calvey that classed 10- and 11-year-olds by how fast they answered imagery and verbal questions: the fast imagers remembered a visual passage better, the fast verbalisers a verbal one. That is one small, old study, read here only as described, and it measured speed, not preference. Rogowsky and colleagues drew the same line in adults. If one format is a real struggle, your university's study-skills or disability service is the place to ask.

How can you check learning-styles claims yourself?

EdCitation is the best tool for checking a claim about learning at its source, because it looks every reference up in the publisher's record and never writes one. On a topic full of second-hand summaries, that is the point: every study here was traced to its record, and one reference we had worded loosely was caught.

Find the studies

Find sources searches about 300 million published works by topic or by the claim a sentence needs to support. A search for "learning styles meshing hypothesis", with no filter set, returned 2,741,985 works. The first was a 2025 study in Pedagogies: An International Journal, "The meshing hypothesis revisited", cited 0 times. The fifth was Clinton-Lisell and Litzinger's 2024 meta-analysis, cited 28 times, which is how it entered this guide. Pashler's review was not among the 97 results returned.

Cite them

Cite a source builds a reference in APA 7, MLA 9, Chicago 18 author-date, Harvard, IEEE and Vancouver from a DOI, title, PubMed ID, ISBN or web address, and screens it for retraction. From the DOI 10.1111/j.1539-6053.2009.01038.x it returned, in a rerun on 27 September 2026:

Pashler, H., McDaniel, M., Rohrer, D., & Bjork, R. (2008). Learning styles: Concepts and evidence. Psychological Science in the Public Interest, 9(3), 105–119. https://doi.org/10.1111/j.1539-6053.2009.01038.x

The record has the title in title case, and Cite a source set it in sentence case, as APA 7 asks for an article title. Husmann and O'Loughlin's and Newton and Miah's titles came back in sentence case too, and Newton and Miah's article number came back correctly as "8, Article 444". Rogowsky's and the 2024 meta-analysis needed nothing either. None of the five was flagged as retracted.

Verify the list

Verify references checks a pasted list or an uploaded paper against the publishers' records, flags retractions, and never shows "could not check" as "not found". The VARK page first came back as "check this": "The address works, but the page there has a different title. Check it is the page meant." We had left out its subtitle; with it added, the tool reported "The page is at this address, and its title matches." On the final list all nine came back verified, none retracted; the Coffield report, which has no DOI, with "Title, author and year match a library catalogue record."

What it costs

Find sources, Cite a source and Verify references are free, with no account, and a free account keeps your reference lists. The Library of books is part of Max, $24 a month (Pricing). EdCitation never writes any part of your work.

Quick questions

Is VARK scientifically valid?

Studying to your VARK result has not been shown to help: in Husmann and O'Loughlin's anatomy course it had no correlation with grades. VARK itself says it records preferences and does not claim matching works.

Am I a visual learner?

You may prefer visual material, but no good evidence shows you learn more from it than anyone else. Use diagrams where the subject has a shape, and test yourself whatever the format.

Should teachers use learning styles?

Pashler and colleagues found no adequate evidence for learning-styles assessments in general education, and the 2024 meta-analysis judged any benefit too small and infrequent to adopt.

Is dual coding the same as learning styles?

Dual coding is not the same: it combines words and images for every learner, while learning styles gives different learners different formats.

How can I check that a learning-styles study is real?

Paste the reference into EdCitation's Verify references, free with no account. Each entry comes back verified, "check this" or not found, with retracted papers flagged.

References

• Clark, J. M., & Paivio, A. (1991). Dual coding theory and education. Educational Psychology Review, 3(3), 149–210. https://doi.org/10.1007/BF01320076
• Clinton-Lisell, V., & Litzinger, C. (2024). Is it really a neuromyth? A meta-analysis of the learning styles matching hypothesis. Frontiers in Psychology, 15, Article 1428732. https://doi.org/10.3389/fpsyg.2024.1428732
• Coffield, F., Moseley, D., Hall, E., & Ecclestone, K. (2004). Learning styles and pedagogy in post-16 learning: A systematic and critical review. Learning and Skills Research Centre. https://www.leerbeleving.nl/wp-content/uploads/2011/09/learning-styles.pdf
• Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving students' learning with effective learning techniques: Promising directions from cognitive and educational psychology. Psychological Science in the Public Interest, 14(1), 4–58. https://doi.org/10.1177/1529100612453266
• Husmann, P. R., & O'Loughlin, V. D. (2019). Another nail in the coffin for learning styles? Disparities among undergraduate anatomy students' study strategies, class performance, and reported VARK learning styles. Anatomical Sciences Education, 12(1), 6–19. https://doi.org/10.1002/ase.1777
• Newton, P. M., & Miah, M. (2017). Evidence-based higher education – Is the learning styles 'myth' important? Frontiers in Psychology, 8, Article 444. https://doi.org/10.3389/fpsyg.2017.00444
• Pashler, H., McDaniel, M., Rohrer, D., & Bjork, R. (2008). Learning styles: Concepts and evidence. Psychological Science in the Public Interest, 9(3), 105–119. https://doi.org/10.1111/j.1539-6053.2009.01038.x
• Rogowsky, B. A., Calhoun, B. M., & Tallal, P. (2015). Matching learning style to instructional method: Effects on comprehension. Journal of Educational Psychology, 107(1), 64–78. https://doi.org/10.1037/a0037478
• VARK Learn Limited. (n.d.). What VARK is (and isn't): A catalyst for better learning. VARK. https://vark-learn.com/about-vark/what-vark-is-and-isnt/