Somewhere in your schooling, someone probably told you what kind of learner you are. Visual, auditory, kinaesthetic; or one of the dozens of other schemes. It felt like self-knowledge, and it has since become one of the most durable ideas in education — surveys of teachers in multiple countries routinely find that 80 to 95% believe in it.
The belief has a precise form, the psychologists who study it have a name for it, and testing it requires a specific experimental design that the education literature has almost entirely declined to use.
The meshing hypothesis
In 2008, Harold Pashler, Mark McDaniel, Doug Rohrer and Robert Bjork were commissioned by Psychological Science in the Public Interest to review the evidence. Their first contribution was to state the claim properly.
The claim is not that people have preferences — they obviously do, and those preferences are measurable and stable. The claim is not that people differ in ability — they obviously do. The claim is what the authors called the meshing hypothesis: that instruction matched to a person’s style produces better learning for that person than the same material delivered in a different mode.
That is a specific empirical prediction, and testing it requires a specific design.
What Pashler and colleagues found
Searching for studies that used the crossover design, they found almost none. The overwhelming majority of the learning-styles literature consisted of studies that classified learners and then did something other than the required comparison — measured preferences, surveyed satisfaction, tested one group one way.
Of the small number of studies that did implement the full design properly, the results contradicted the hypothesis. The predicted interaction did not appear.
Their conclusion was blunt: there is no adequate evidence base to justify incorporating learning-styles assessments into general educational practice, and the widespread use of the concept in schools is not supported by the research that exists.
What happened after the review
Philip Newton examined what the education literature did with that finding. Looking at research articles on learning styles indexed in the major education and biomedical databases in the years following the review, he found the field had carried on largely unchanged: the great majority of papers implicitly or explicitly endorsed the use of learning styles in teaching.
Among the papers he identified, essentially one tested the meshing hypothesis directly. None substantively engaged with the critiques. The published literature was not debating the question — it was assuming the answer and moving on to applications.
Newton and Miah followed up by surveying higher-education staff and found the belief held by a large majority even among academics who had encountered the criticism.
The study that checked what students actually do
A 2019 study by Polly Husmann and Valerie O’Loughlin in Anatomical Sciences Education took a different angle with undergraduate anatomy students. They administered the VARK questionnaire, then asked what study strategies the students actually used, then looked at exam outcomes.
Two findings, both awkward. First, most students did not use study strategies that matched their assessed learning style — the categories did not predict behaviour. Second, the students whose study strategies did match their assessed style did not perform better.
The instrument failed to predict the thing it was supposed to describe, and matching failed to produce the benefit it was supposed to produce.
Why it feels so true
The belief survives because it is built out of several things that are genuinely correct.
Preferences are real. People do prefer diagrams or explanations, and preferences are stable enough to be measured reliably. What does not follow is that learning is better in the preferred mode.
Material has an optimal modality. Geometry is visual. A foreign language’s pronunciation is auditory. Surgery is kinaesthetic. This is a property of the subject, not the student, and it is often mistaken for evidence of styles.
Multimodal teaching works. Presenting material in more than one form genuinely helps — but it helps nearly everyone, which is the opposite of the styles prediction. Teachers who diversify their methods see results and reasonably attribute them to reaching different learners.
It is flattering and harmless-sounding. Being told you have a distinct cognitive profile explains past difficulty without implying anything unpleasant. That is an unusually comfortable belief, and comfortable beliefs are sticky.
Does the belief do harm?
This deserves a careful answer rather than an alarming one. Teachers who use varied methods because they believe in learning styles are doing something that works, for the wrong reason. That is not a disaster.
The costs are subtler. Time and money spent on style assessments is time and money not spent on methods with strong evidence — spaced practice, retrieval testing, interleaving, worked examples — all of which have substantial support and none of which require knowing anything about the individual student. There is also a risk of a self-limiting identity: a student who has decided they are “not a reader” has acquired a reason to avoid reading.
The harder point
The interesting thing about learning styles is not that a popular idea turned out to be unsupported. It is what happened afterwards. A major review in a prominent journal identified the exact experiment needed, reported that it had almost never been run, and reported that where it had been, the result was negative. The field’s response was to continue publishing on applications.
Eighteen years on, the surveys still find teacher belief above 80%. Whatever mechanism is supposed to carry a finding from the literature into practice, this is a reasonably clean demonstration that it did not operate here.
Sources
- Pashler H, McDaniel M, Rohrer D, Bjork R. Learning Styles: Concepts and Evidence. Psychological Science in the Public Interest 2008;9(3):105-119.
- Newton PM. The Learning Styles Myth is Thriving in Higher Education. Frontiers in Psychology 2015;6:1908.
- Husmann PR, O’Loughlin VD. Another Nail in the Coffin for Learning Styles? Anatomical Sciences Education 2019;12(1):6-19.
- Willingham DT, Hughes EM, Dobolyi DG. The Scientific Status of Learning Styles Theories. Teaching of Psychology 2015;42(3):266-271.
