Neuromyths: the popular beliefs that won’t die.

Ask a room of educators or trainers three questions. Do people learn better in their preferred learning style? Are right-brained people creative? Do we use only 10% of our brains? Most hands go up for at least one.

These are not fringe beliefs held by the uninformed. They are majority positions among the people in charge of how millions learn — measured, repeatedly, in the studies this article reviews.

The stakes of the hand-count are not rhetorical. The people raising their hands design courses, buy tools, coach colleagues, and grade learners. And each myth carries a design decision inside it: sort learners by style, balance activities by hemisphere, cap videos at the length of a goldfish’s attention. Beliefs about learning are unusual among professional errors in one way: they compile directly into practice.

The field calls them neuromyths: claims about the brain and learning that sound scientific, trace loosely to real research, and are false in the form everyone repeats. One by one, each myth has its own debunking literature. Together, they form a measurable epidemiology — a map of who believes what, at what rates, protected by which factors. That map is the useful part. It explains why myths about learning beat knowledge about learning in the very professions that should know better.

The prevalence study that named the problem

Anecdotes about credulous colleagues prove nothing. The field’s contribution was to measure the problem the way it would measure any other outbreak: broad samples, standard item sets mixing myths with true statements, and analysis of who believes what and why. The modern benchmark surveyed teachers in the UK and the Netherlands with a mixed set of brain statements — some true, some myths — and asked which they believed. The results organized a decade of follow-up work: teachers believed around half of the myths presented.

The flagship items ran at rates that deserve their notoriety. Roughly 93% endorsed matching teaching to learning styles, and 91% endorsed hemisphere dominance as an explanation of learner differences (Dekker, Lee, Howard-Jones & Jolles, 2012). The study’s most quoted twist: teachers with more general knowledge of and interest in the brain believed more myths, not fewer. Enthusiasm for the brain, without the detail to tell true from false, works as a myth delivery channel.

Replications broadened the map without changing its shape. Surveys across countries found similar belief rates on every continent tested. A large U.S. study compared the general public, educators, and people with high exposure to brain science. Educators believed slightly fewer myths than the public; the trained group fewer still. But even the most trained group endorsed almost half of the classic items — training reduced belief rather than removing it (Macdonald, Germine, Anderson, Christodoulou & McGrath, 2017). The review literature’s summary is blunt: the myths are international, professional, and robust to ordinary education (Howard-Jones, 2014).

The myths, one by one

The canon is worth walking one myth at a time. Each has a distinct anatomy — a real kernel, a specific distortion, and a direct refutation — and the anatomies teach the pattern better than any definition.

The 10% myth. No brain-science result has ever supported it. Imaging shows nearly all brain regions active across ordinary tasks, and damage to “unused” areas is never harmless. The myth’s family tree runs through misquoted early-1900s guesswork and self-help marketing, where its function is obvious: a pool of untapped potential is a product’s best friend (Howard-Jones, 2014).

Left-brain/right-brain personalities. The real finding — hemispheric specialization for certain functions, from split-brain and lesion research — was inflated into a typology of logical-left and creative-right people. The direct test at scale: resting-state imaging of over a thousand people found no evidence of globally dominant hemispheres (Nielsen, Zielinski, Ferguson, Lainhart & Anderson, 2013). Lateralization is real for functions, absent for personalities.

The Mozart effect. The 1993 original reported a short-lived improvement on one spatial task in adults after listening to a sonata (Rauscher, Shaw & Ky, 1993). Popular culture converted it into “music makes babies smarter” — a claim the original never made about babies, intelligence, or duration. The escalation was institutional as well as commercial: a U.S. state famously budgeted classical CDs for newborns, on the strength of a fifteen-minute effect in college students. The meta-analytic autopsy, memorably titled “Mozart effect–Shmozart effect,” found little to nothing left once arousal and preference were controlled (Pietschnig, Voracek & Formann, 2010). Pleasant stimulation briefly perks performance, and Mozart holds no privileged key.

Learning styles. The reigning champion, covered in full in its own article in this library. The matching hypothesis — instruction aligned to diagnosed styles improves learning — has failed where properly tested. Yet the belief holds steady above ninety percent in survey after survey (Dekker et al., 2012).

Its survival despite a dedicated debunking record makes it the type specimen for everything this article describes. The kernel is real: learners differ. The distortion: the differences are fixed channel preferences — visual, auditory — that teaching must match. The survival strategy: the claim flatters both teacher and learner while requiring no evidence either will ever check.

Why the myths win

The belief rates describe; they do not explain. The explaining — why these beliefs, in these professions, at these rates — is where the literature becomes useful in practice. The spread factors it identifies are the levers a company can actually pull. The research has isolated them, and none is stupidity.

Neuromyths are simplifications of real science. Each carries a kernel — specialization, plasticity, arousal effects — that lends borrowed credibility and makes correction feel like pedantry. They are intuitive: styles matches the genuine observation that learners differ; the 10% myth matches the felt sense of untapped potential. They arrive wearing the brain — and adding neuroscience language to weak explanations measurably increases how convincing non-experts find them (Weisberg, Keil, Goodstein, Rawson & Gray, 2008). That finding explains half of the genre’s marketing. And they spread through professional channels — teacher training, CPD workshops, train-the-trainer decks — which certify them exactly as our article on the learning pyramid describes.

The protective-factor data explain why the obvious fix fails. General exposure to brain science helps only modestly (Macdonald et al., 2017). The myths are not gaps in brain knowledge. They are confident learning claims, and the discipline that refutes them is experimental psychology: control groups, transfer measures, meta-analysis. The discriminating skill is about evidence, not anatomy. People who ask “what was the design behind this claim?” shed myths; people who pile up brain facts gain new vocabulary for the myths they keep.

The corporate strain of the infection

The survey research sampled schools, but the corporate strain is arguably worse-placed, because the workplace lacks even education’s imperfect antibodies. No education faculty reviews what enters leadership courses. Buying is faster, vendor claims face less peer review, and “brain-based” is a selling word rather than a red flag.

The corporate genre has its own additions to the canon. Dubious attention-span statistics justify content shredded into fragments. “We only remember X% after Y days” figures carry the learning pyramid’s exact sourcing profile. Hemisphere language returns repackaged as thinking-style labels for team workshops. Each buys real budget; none survives the design question.

The typology industry deserves its own sentence, because it is where hemisphere mythology went professional. Instruments sort employees into thinkers versus feelers, or left versus right quadrants, deployed for team make-up and communication training. Whatever value the workshop conversation has, the brain framing is decoration. It sits on constructs whose measurement properties would embarrass any test in our assessment cluster. And the decoration is doing the persuading, exactly as the seductive-allure experiments predict (Weisberg et al., 2008).

The organizational cost

Skeptical readers reasonably ask whether survey belief turns into anything that matters — people believe many idle things that never touch their work. For neuromyths the evidence is the buying record and the design record, both public. It is tempting to file the beliefs as harmless folklore; the accounting says otherwise. The direct costs sit in the buying: styles inventories, workshops on balancing hemispheres, and brain-branded programs are bought with budgets that real programs were denied.

The indirect costs run deeper. Every myth occupies the explanatory slot a true model should hold. A trainer who puts a struggling learner down to a style mismatch stops searching for the actual cause — missing prerequisites, no retrieval practice, overload. The myth works as a full stop on diagnosis. And design attention is finite. Hours spent tailoring modalities are hours not spent on spacing, testing, and feedback — swaps with known, large costs given the effect sizes on the true side of the ledger (Howard-Jones, 2014).

Reading a brain claim: a two-minute protocol

Because the skill is about evidence, it can be taught as a checklist, and the checklist fits on a card. First, separate the brain claim from the learning claim. “The hippocampus consolidates memory” and “therefore our product improves retention” are different claims, and the second never inherits the first’s evidence. Second, ask for the design: was the learning claim tested with a comparison group doing something equally engaging, on outcomes beyond the training itself?

Third, check the numbers’ texture. Real findings arrive with effect sizes, intervals, and boundary conditions; myths arrive with round percentages and universal scope. Fourth, trace one citation — a single spot-check catches circular sourcing more often than seems possible, as the pyramid literature shows.

The protocol’s value is that it works without expertise in either brain science or statistics. It moves the buyer’s question from “is this true?” — unanswerable without a literature review — to “has this been shown the way true things are shown?”. That second question anyone can answer from the vendor’s own materials in minutes. Most myths fail at step two. The ones that survive all four steps have, by passing, stopped being myths.

What actually reduces belief

If exposure to brain science barely helps, what does? The question has an experimental literature of its own: refutation studies comparing correction formats in real classrooms and staff-training settings. Its findings are specific enough to build training from. The intervention studies converge on a two-part prescription.

Refutation must be specific. Myth-by-myth corrections that state the false claim, the kernel it distorted, and the true finding outperform general science education. The format matters because the myth and its correction compete for the same slot, and the correction must fit the slot to win it.

Replacement must follow. Corrections decay unless the empty slot is filled with a usable model. Tell a trainer styles-matching is false and, unless dual coding, retrieval, and load theory are installed in its place, the styles model quietly returns. Practitioners need some working model and will keep the one they have over none (Macdonald et al., 2017), (Kowalski & Taylor, 2009).

The training implication is concrete. A half-day on the psychology of evidence — designs, controls, transfer, meta-analysis, taught through two or three famous myths — buys more protection than a semester of brain anatomy. It also compounds. The evidence habit carries over to the next myth, the one not yet invented — which anatomy knowledge never will.

What the evidence doesn’t show

• It doesn’t show neuroscience is irrelevant to education. The mature bridge field — sleep and consolidation, spacing’s biology, plasticity’s real limits — is productive precisely where it respects the behavioral evidence; the objection is to ornamental brain talk, not to the science (Howard-Jones, 2014).
• It doesn’t show believers teach badly overall. Endorsing a myth on a survey and building practice around it are different exposures; the measured harm concentrates where myths drive purchasing and design decisions.
• It doesn’t license smugness. The high-exposure groups still endorsed large fractions of the items (Macdonald et al., 2017); the honest response to the prevalence data is an audit, not superiority.
• Kernels remain true. Hemispheric specialization, arousal effects, learner differences, and untapped improvement are all real; the myths are what happened when each kernel was stretched into a slogan. Correcting the slogan must not discard the kernel.

Where the evidence stops

1. 1It doesn’t show neuroscience is irrelevant to education
2. 2It doesn’t show believers teach badly overall
3. 3It doesn’t license smugness
4. 4Kernels remain true

© 2026 FUTURE PROOF™

What this means for practice

The fixes here are aimed at institutions on purpose, because the survey data say individual enlightenment does not scale. Run the audit the surveys imply. List the brain claims in your company’s training materials, vendor decks, and certification courses. Check each against the pattern this article and its siblings document — round numbers, missing designs, brain imagery doing the arguing.

Where myths surface, correct them in the format the intervention data endorse: specific refutation plus installed replacement, in the same materials, occupying the same slot. And add the one buying question that filters the whole genre: what is the experimental evidence for the learning claim — not the brain claim? A product that answers with imaging pictures has answered.

Train the evidence protocol, not the anatomy. The two-minute checklist above, taught through two or three classic myths, carries over to claims not yet invented — which the survey data say is the only durable protection.

Above all, make your outcomes myth-proof by design. Suppose a company’s learning systems run the replicated effects by default: spacing, retrieval, feedback, load-aware sequencing, all automatic. That company has decoupled its results from its folklore. Individual beliefs can lag while measured practice stays correct, and the retention curves themselves become the standing rebuttal. Beliefs are slow to fix. Defaults are fast, and defaults are where the learning actually happens.