Learning-Style Preferences¶

Learner Variable · All learner variables
Evidence · 25 claims (16 show a difference, 3 mixed, 6 no difference) · 14 studies (4 causal, 4 review, 4 associational, 2 theoretical), q1–q3 · 7 of 14 report an effect size · 24 claims rest on one study

Bottom line¶

Matching instruction to a learner's preferred "learning style" does not improve learning. The meshing hypothesis (teach visual learners visually, auditory learners aurally, and so on) has been tested directly many times, and the studies that cross learners' styles with matched and mismatched instruction do not find the interaction it predicts (Pashler et al., 2008; Rogowsky et al., 2020): Learning styles matching does not improve learning [+S]. Learners do report preferences, and those preferences can matter for comfort and choice, but they are not a basis for designing instruction.

The single studies listed below report differences between learners who describe their style differently (in beliefs, or in gains in one condition). None of them is a test of matching instruction to style, and none overturns the finding above. What does work is matching the format to the content: a diagram for spatial relations, narration with animation, worked examples for novices.

Description¶

Learning-style preferences are learners' reported preferences for one mode or format of instruction, often classified with instruments such as VARK. The preferences are real in the sense that learners report them, but the meshing hypothesis, that teaching in a learner's preferred style improves learning, has repeatedly failed when tested properly: studies that cross style with matched and mismatched instruction find no interaction. The page records the evidence so a design can account for what learners prefer without building on the matching claim. Matching the format to the content (a diagram for spatial relations) is supported; matching it to the learner's style is not.

Implications¶

Context¶

Preferences can inform choice and comfort, but not the expectation of better learning from matching.
Time spent assessing styles is time not spent on practices with evidence.

Target Learners¶

All learners: the evidence does not support sorting learners by style.

Target Learning Objectives¶

Achievement, which matching to style has not been shown to improve.

Claims¶

Accommodator learners gained significantly more in the VR (guided exploration) mode than in VR (non-guided exploration) or Non VR modes [+M] — an effect differed by reported style in this study; matching to style is not supported overall
Assimilator learners gained significantly more in the VR (guided exploration) mode than in VR (non-guided exploration) or Non VR modes [+M] — an effect differed by reported style in this study; matching to style is not supported overall
Avoidant-style preservice teachers hold stronger certainty-of-knowledge beliefs than participant-style peers [+W] — learners reporting different styles differed here; not a test of matching instruction to style
Canfield's learning style variables predicted academic success where cognitive style did not [+W] — learners reporting different styles differed here; not a test of matching instruction to style
Chatbot preference groups differed in total EBAPS and Axis 4/5 scores at the nominal level, with small effect sizes, but not after Bonferroni adjustment [~M] — learners reporting different styles differed here; not a test of matching instruction to style
Collaborative-style preservice teachers score higher on continuous and intentional thinking than avoidant and dependent styles [+W] — learners reporting different styles differed here; not a test of matching instruction to style
Combination-preferring students scored higher than answer-preferring students on evolving-knowledge (Axis 4) and source-of-ability (Axis 5) EBAPS dimensions [+W] — learners reporting different styles differed here; not a test of matching instruction to style
Greater proportion of voice prompts used correlates moderately with greater stated preference for voice [+W] — learners reporting different styles differed here; not a test of matching instruction to style
In one author's classroom media experiments, intuitive learners tended to score above sensing learners, though not in every format comparison [~W] — learners reporting different styles differed here; not a test of matching instruction to style
In the VR (guided exploration) mode, assimilator and accommodator learners did not differ significantly in gain score [-W] — learners reporting different styles differed here; not a test of matching instruction to style
Independent-style preservice teachers score higher on open-mindedness than avoidant, collaborative, and competitive styles [+M] — learners reporting different styles differed here; not a test of matching instruction to style
Intellectual styles are partly socialized and therefore modifiable to some degree [~W] — instruction changes reported style
Learning style (visual-verbal) shows no significant correlation with the effectiveness of pictorial elucidation for adolescent EFL learners [-W] — an effect differed by reported style in this study; matching to style is not supported overall
Learning styles impact teachers, pedagogy, student engagement, and assessment in K-12 education [+W] — learners reporting different styles differed here; not a test of matching instruction to style
Learning Styles Matching Does Not Improve Learning [-S] — an effect differed by reported style in this study; matching to style is not supported overall
Learning styles measured by Kolb's inventory showed no significant change over one year of varied instruction [-M] — instruction changes reported style
No significant interaction between learning mode and learning style on gain scores, while the learning-mode main effect was significant [-W] — an effect differed by reported style in this study; matching to style is not supported overall
No statistically significant EBAPS differences were observed for Axes 1–3 or between guided-inquiry and answer-oriented chatbot preferrers [-W] — learners reporting different styles differed here; not a test of matching instruction to style
Preservice teachers' epistemological beliefs differ significantly by learning style in all four dimensions [+M] — learners reporting different styles differed here; not a test of matching instruction to style
Preservice teachers' reflective thinking tendencies differ statistically by learning style in all seven dimensions [+W] — learners reporting different styles differed here; not a test of matching instruction to style
Students preferring combination-style chatbots had higher total EBAPS scores than students preferring direct-answer chatbots (nominal significance only) [+M] — learners reporting different styles differed here; not a test of matching instruction to style
In hierarchical regression, gender and LD status contributed 6.4% to science achievements, and auditory learning preference contributed a further significant 3.1% beyond them [+M] — learners reporting different styles differed here; not a test of matching instruction to style
Science achievements correlated significantly and positively with preferences for visual, auditory, and agency-support means, but not sensorimotor means, with low coefficients [+W] — learners reporting different styles differed here; not a test of matching instruction to style
Correlational meta-analyses of learning styles and achievement average r = 0.24 but do not test or support the matching hypothesis [+M] — learners reporting different styles differed here; not a test of matching instruction to style
Meta-analyses testing the matching hypothesis show a negligible overall effect of matching teaching to students' learning styles (d = 0.04) [-M] — an effect differed by reported style in this study; matching to style is not supported overall

The evidence by question¶

What should the adaptation be based on: responses, or learner styles and preferences? Learning Styles Matching Does Not Improve Learning [+S]

Page metadata

Field	Value
`type`	`learner-variable`
`id`	`learning-style-preferences`
`title`	`Learning-Style Preferences`
`description`	`A learner's stated preference for receiving information in one form (such as visual, auditory or kinaesthetic), and the claim that matching instruction to it improves learning.`
`status`	`review`
`canonical`	`True`
`generated`	by `claude/unspecified` on `2026-10-15`
`trust`	unverified — no one has confirmed this page's content

Frontmatter is the page's machine-readable record — see the Schema & Guide.