Why the Learning Styles myth spreads

Learning Styles theory is the most famous myth in education. Different variants of the theory exist, which all make the claim that people can be categorised by which ‘style’ of teaching they learn best from. There is no reliable evidence in favour of the theory being accurate.

Despite this, a 2020 survey found that 90% of educators believe that we each have a ‘learning style’ [1]. This finding was met with much derision by education researchers. How has a theory with little evidence become and stayed so pervasive?

For those unfamiliar with Learning Styles theory, Veritasium provides an overview of the theory here.

What he doesn’t address is why it spreads. But it can be explained by simple models on the individual level. And, beneath the surface, the debate over whether Learning Styles really exist serves as a proxy for a war of ideas, which maintains its relevance on the cultural level. It substitutes for a battle centring on agency versus conformity in education.

What Learning Style theory claims

Most academic papers supporting or proposing the theory suggest that there are discrete ‘Learning Styles’. That is, we each have a style of learning to which we are best suited. They usually claim that the number of these styles is small. In the most widely shared theory - VARK - there are 4 distinct styles [1].

• Visual
• Auditory
• Reading / Writing
• Kinaesthetic

There are now over 70 different learning style classification systems, all variants on the theme.

The application of these theories often takes the form of matching the preferred Learning Style of the student to the delivery medium. For example, here is a set of suggestions from a US University for how their students should study according to their Learning Style:

A set of tips for how to study based on your learning style. From a US University website. There is no evidence that this is effective.

As a personal aside, I remember learning this version of Learning Styles theory as a student. We had presentations on what Learning Styles were. We did workshops to find which we fitted into (I was kinaesthetic, apparently). This led my primary school teacher to teach me the words to the national anthem through the medium of dance. Cue hilarious failure.

Yet it’s true that people have differing cognitive abilities. Some of us have rapid reaction times, and some have better spatial cognition. However, the papers proposing new classifications of Learning Styles provide no evidence that cognitive differences lead to distinct, identifiable Learning Styles. That should immediately ring alarm bells.

A recent paper [2] summarises the evidence as follows:

“Two substantial reviews of the literature concluded that there was currently no evidence to support the idea that the matching of instructional methods to the supposed Learning Styles of individual students improved their learning. Subsequent reviews have reached the same conclusion and there have been numerous, carefully controlled attempts to test this “matching” hypothesis.”

The academic literature is pretty conclusive. Experiments have tested splitting people by their learning style and delivering content in that format, then observing the difference between groups. This and other such experiments find no difference. I won’t rehash that here - there are lots of good articles on this subject. I’m much more interested in why it’s still so pervasive.

So why do students, and even education professionals, report that they believe it? The first is how concise and plausible it sounds, which determines spread. The second is confirmation bias, which determines how stubbornly it persists.

Concise and plausible

Learning Styles are very easy to describe. It takes a sentence or so to convey the idea. Whereas mind palaces or mnemonics can sound complicated, Learning Styles are simple. This has a large effect on how readily the idea spreads. In a large-scale meta-analysis of the adoption of new innovations in the UK National Health Service, ideas that were perceived as simple were adopted faster [4].

Unlike other education theories, Learning Styles speak to our experience of the world. The different categories usually relate to our different perceptual experiences or cognitive faculties. When told I was a kinaesthetic learner, I could immediately relate it to my enjoyment of sports. This made me feel I could understand the theory immediately. When prompted, everyone can find a memory of a new medium of presentation leading to a breakthrough in personal understanding. This is not a scientific approach, but makes the theory feel as if it could be true. This plausibility of Learning Styles theory helps it spread as a myth.

Learning Styles are therefore particularly potent because they are concise, plausible, and sound authoritative by using scientific language. A dangerously sharable combination.

Memory and Confirmation bias

Once the seed is planted and we think an idea may be true, confirmation bias kicks in. We start to interpret the world as if an idea were true. It happens because humans have a tendency to accept evidence that aligns with what we already believe more readily. Once we have accepted Learning Styles theory, we notice more experiences that suggest it's true.

Human memory and perception is strange in this way. It’s not a faithful replay of everything that has happened or display of everything that’s happening. Rather, our memory is often reshaped by later experience and how it's indexed.

This phenomenon occurs in criminal trials with police line-ups. Memory morphs when prompted with the line-up of possible criminals. People don't recall the face of the perpetrator, they identify the person most similar. The result is that 69% of convictions overturned by DNA evidence originally relyied on eyewitness testimony of this kind [5] Similarly, asking a question about Learning Styles makes the respondent search for, or adjust, a relevant memory. This is effect is amplified when referring to subjective preferences, rather than measured performance.

To see that subjective preferences don’t reflect reality, we can survey students and compare their answers to their behaviour when studying. When asked how they learn, people give specific and cogent answers. They describe how they prefer to learn through particular mediums. But when learning, people don’t do what they say they do. From the same article again:

“The identification of supposed student Learning Style does not appear to influence the way in which students choose to study, and does not correlate with their stated preferences for different teaching methods.”

Partly this is due to the researchers asking leading questions. If asked about which Learning Style you fall into, it's easier to pick a style than question the premise. And if you’ve heard of learning styles before and have had time for confirmation bias to kick in, you may have a default answer you give.

Learning Styles as a proxy war

A striking feature of the debate over whether Learning Styles exist is how heated the language is. Detractors refer to the theory as like chasing 'unicorns'. Supporters call this 'patronising', some even call it racist [3] and compare the research used to dismiss Learning Styles as ‘eighth grade science fair experiments’.

Both sides also claim conspiracy. Learning Styles detractors claim the idea is propagated by consultancies who get paid to teach it. Supporters claim an ‘artificial scientific consensus’ is used to justify contracts from governments to corporate giants like Pearson and McGraw-Hill [6]. There seems to be something deeper at play than debates over data. What if there’s another explanation for the persistence of Learning Styles on the cultural level?

We'll call one side of our debate the empiricists. They view randomised controlled trials (RCTs) as the only valid form of evidence. The empiricists care most about ensuring a basic level of educational quality by employing evidence-based practice. This can lead to an urge to teach all students to a set curriculum which can be controlled based on this evidence. To them, Learning Styles are a perfect example of a lack of evidence-based teaching. They view Learning Styles as akin to astrology.

Let's call the other side of our debate the romanticists. They want to encourage agency and tailor to the differences between students. As they focus on individuals, they care less about RCTs. Learning Styles serve as a placeholder label to help them do so. It doesn’t matter to them if a distinct set of styles actually exists. To them, attacks on Learning Styles sound like attempts to remove human differences from teaching altogether.

When framed this way, both perspectives have their merits. One prefers reductionism with the aim of increasing the quality of teaching. The other embraces agency and human difference with the aim of inspiring creativity. It also helps to explain why, despite a lack of evidence, belief in Learning Styles persists.

This combination of a cultural level fight around a spreadable, sticky idea explains why Learning Styles theory is so widely believed.

---

Citations

[1] Fleming and Mills, 1992. Not another inventory, rather a catalyst for reflection. https://digitalcommons.unl.edu/cgi/viewcontent.cgi?article=1245&context=podimproveacad

[2] Newton and Salvi, 2020. How Common Is Belief in the Learning Styles Neuromyth, and Does It Matter? A Pragmatic Systematic Review. https://www.frontiersin.org/articles/10.3389/feduc.2020.602451/full

[3] Black, Carol. Science / Fiction. http://carolblack.org/science-fiction/

[4] Greenhalgh, Robert and Bate, 2004. How to Spread Good Ideas: A systematic review of the literature on diffusion, dissemination and sustainability of innovations in health service delivery and organisation. https://www.cs.kent.ac.uk/people/staff/saf/share/great-missenden/reference-papers/Overviews/NHS-lit-review.pdf

[5] Mistaken Identifications are the Leading Factor In Wrongful Convictions. https://innocenceproject.org/eyewitness-identification-reform/

[6] Grunwald, Michael. 2006. Billions for an Inside Game on Reading. ****https://www.washingtonpost.com/wp-dyn/content/article/2006/09/29/AR2006092901333.html