Evidence for Learning styles: what the evidence really says #

Every substantive claim on the Learning styles: what the evidence really says page is checked against current research. Here is each claim, how well today’s evidence supports it, and the sources. The full, de-duplicated source list lives on the references page.

Supported · moderate evidence — People reliably report a preferred mode for taking in information (e.g. visual, auditory, kinesthetic), and these self-reported preferences are real and relatively stable.

It is uncontested that learners express stable preferences for how they receive material; Pashler et al. explicitly separate the existence of preferences (which they accept) from the meshing claim (which they reject). The 2026 consensus continues to treat modality preferences as genuine self-report phenomena.

Sources: Pashler, H., McDaniel, M., Rohrer, D., & Bjork, R. (2008), Learning styles: Concepts and evidence. Psychological Science in the Public Interest — https://doi.org/10.1111/j.1539-6539.2009.01038.x · full reference ›

Supported · strong evidence — Matching instruction to a learner’s preferred style (the ‘meshing hypothesis’) does not reliably improve learning outcomes.

Pashler et al.’s review found that almost no studies used the crossover-interaction design required to test meshing, and those that did failed to support it; this remains the firm scientific consensus in 2026. The debunk of the matching claim is the well-supported position.

Supported · strong evidence — A proper test of learning-styles matching requires a crossover interaction (each style-group learning best in its matched modality and worse in others), and studies meeting that bar are essentially absent.

The crossover-interaction requirement is the methodological standard set out by Pashler et al. and is widely accepted; the scarcity of studies meeting it is a central, uncontested point of their review and of subsequent commentary.

Sources: Pashler, H., McDaniel, M., Rohrer, D., & Bjork, R. (2008), Learning styles: Concepts and evidence — https://doi.org/10.1111/j.1539-6539.2009.01038.x · full reference ›

Supported · strong evidence — Belief in learning-styles matching is widespread among the public and educators despite the lack of supporting experimental evidence.

Nancekivell, Shah & Gelman (2020) documented that a large majority of US adults endorse learning-styles beliefs, and other surveys of teachers worldwide report similarly high endorsement rates; the gap between popular belief and evidential support is well established in 2026.

Sources: Nancekivell, S. E., Shah, P., & Gelman, S. A. (2020), Maybe they’re born with it, or maybe it’s experience: Toward a deeper understanding of the learning style myth. Journal of Educational Psychology — https://doi.org/10.1037/edu0000366 · full reference ›

Supported · strong evidence — Learning-styles theories that posit fixed, instruction-relevant style types lack a credible scientific basis.

Willingham, Hughes & Dobolyi (2015) reviewed the major learning-styles theories and concluded none has adequate empirical support for the prediction that matching improves learning; this assessment is the mainstream view in 2026.

Sources: Willingham, D. T., Hughes, E. M., & Dobolyi, D. G. (2015), The scientific status of learning styles theories. Teaching of Psychology — https://doi.org/10.1177/0098628315589505 · full reference ›

Supported · moderate evidence — Distinct sensory modalities (vision, hearing, movement) are processed in characteristic brain regions, but this reflects the type of content, not a learner-specific ‘visual brain’ or ‘auditory brain’ that learns better from its matched modality.

Modality-specific cortical processing (occipital for vision, temporal for audition, motor/cerebellar for movement) is solid neuroscience, but Willingham et al. and the broader literature reject inferring person-level style brains or a matching benefit from it; localisation of modality processing does not license the meshing claim. Rated ‘qualifies’ because the page accepts the neuroscience while denying the matching inference.

Sources: Willingham, D. T., Hughes, E. M., & Dobolyi, D. G. (2015), The scientific status of learning styles theories — https://doi.org/10.1177/0098628315589505 · full reference ›

Supported · moderate evidence — Howard Gardner’s ‘Multiple Intelligences’ is best treated as a useful vocabulary for talking about strengths rather than an empirically validated taxonomy of separate, brain-localised intelligences or a prescription for matching teaching to a learner profile.

Critical reviews (e.g. Waterhouse 2006) find no adequate empirical validation of MI as a set of independent intelligences, and Gardner himself has distinguished MI from learning-styles matching; treating MI as descriptive vocabulary rather than validated biology is the cautious, evidence-aligned reading in 2026.

Sources: Waterhouse, L. (2006), Multiple intelligences, the Mozart effect, and emotional intelligence: A critical review. Educational Psychologist — https://doi.org/10.1207/s15326985ep4104_1 · full reference ›

Supported · moderate evidence — Material is best presented in the mode the content itself demands (maps seen, melodies heard, procedures performed), independent of any individual learner’s stated style preference.

Pashler et al. and later authors note that the optimal modality is typically dictated by the to-be-learned content rather than by the learner; this content-over-learner framing is well accepted, though it is a reasoned position more than a single quantified effect.

Sources: Pashler, H., McDaniel, M., Rohrer, D., & Bjork, R. (2008), Learning styles: Concepts and evidence — https://doi.org/10.1111/j.1539-6539.2009.01038.x · full reference ›

Supported · strong evidence — Combining a verbal explanation with a corresponding visual representation improves learning for learners generally (the multimedia / dual-coding principle), not selectively for self-identified ‘visual learners’.

The multimedia principle — that combining words and matching pictures aids learning across learners — is one of the most replicated findings in instructional research (Mayer 2009 and the meta-analytic multimedia-learning literature) and is the sound reason multimodal presentation helps, distinct from learning-styles matching.

Sources: Mayer, R. E. (2009), Multimedia Learning (2nd ed.). Cambridge University Press — https://doi.org/10.1017/CBO9780511811678 · full reference ›

Supported · strong evidence — The largest, most general gains in durable learning come from how material is engaged — retrieval practice, spacing, interleaving, and elaboration — rather than from the sensory channel used.

Dunlosky et al.’s (2013) review rates practice testing and distributed practice as high-utility across conditions and learners, with broad subsequent replication; redirecting effort from modality matching to these methods is strongly evidence-supported in 2026.

Sources: Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013), Improving students’ learning with effective learning techniques. Psychological Science in the Public Interest — https://doi.org/10.1177/1529100612453266 · full reference ›

Supported · moderate evidence — Receiving material in a preferred or familiar format feels easier and more fluent, but that subjective fluency is a poor indicator of how much durable learning has occurred.

The dissociation between processing fluency / subjective ease and actual long-term retention is well documented in the metacognition and desirable-difficulties literature (Bjork, Dunlosky & Kornell 2013); fluent-feeling study often overestimates learning.

Sources: Bjork, R. A., Dunlosky, J., & Kornell, N. (2013), Self-regulated learning: Beliefs, techniques, and illusions. Annual Review of Psychology — https://doi.org/10.1146/annurev-psych-113011-143823 · full reference ›

Supported · moderate evidence — The simple ’left-brain / right-brain’ dichotomy misrepresents how the brain works; while some functions are lateralised (e.g. language tends left, some music processing right), most cognition involves both hemispheres working together.

Large-scale connectivity analyses (e.g. Nielsen et al. 2013) found no evidence that individuals are globally ’left-brained’ or ‘right-brained’, though specific functions are lateralised; the popular dominance dichotomy is not supported, which matches the page’s claim.

Sources: Nielsen, J. A., Zielinski, B. A., Ferguson, M. A., Lainhart, J. E., & Anderson, J. S. (2013), An evaluation of the left-brain vs. right-brain hypothesis with resting state functional connectivity magnetic resonance imaging. PLOS ONE — https://doi.org/10.1371/journal.pone.0071275 · full reference ›