You've almost certainly been told you have a learning style. Visual, auditory, or kinesthetic. Maybe read/write. Maybe some combination. At some point in your education — grade school, college, corporate training — someone told you that you learn...
In This Chapter
Chapter 12: Learning Styles — The Myth That Won't Die
You've almost certainly been told you have a learning style. Visual, auditory, or kinesthetic. Maybe read/write. Maybe some combination. At some point in your education — grade school, college, corporate training — someone told you that you learn best when information is presented in your preferred modality. See it, hear it, or do it, depending on your type.
This idea is one of the most believed and most debunked claims in all of psychology.
A 2014 survey found that 95% of teachers in the UK believed that students learn better when instruction matches their learning style. Studies across multiple countries have found similar rates. Learning styles is not a fringe belief — it is the dominant framework through which most educators understand learning differences. It is embedded in teacher training programs, educational curricula, and corporate training design.
It is also almost certainly wrong.
The specific claim that people learn better when instruction is matched to their preferred learning style — the meshing hypothesis — has been tested repeatedly and has failed every rigorous test. The most comprehensive review of the evidence, by Pashler and colleagues (2008), found that the existing research "does not adequately support the hypothesis" and that the few studies that did support it had significant methodological problems.
This chapter is not about whether people have preferences. They do. Some people prefer to learn by reading; others prefer to learn by watching a video. The question is whether these preferences predict actual learning outcomes — and the answer, consistently, is no.
Before You Read: Confidence Check
Rate your confidence (1–10) that each statement is true.
- "People learn better when instruction matches their learning style (visual, auditory, kinesthetic)." ___
- "I know my own learning style." ___
- "Learning styles are supported by brain science." ___
- "Teachers should tailor instruction to different students' learning styles." ___
- "If learning styles don't work, then all students learn the same way." ___
The Claim: Matching Instruction to Style Improves Learning
The learning styles hypothesis has two components:
Component 1: People differ in their preferred way of receiving information. This is true and uncontroversial. Some people prefer reading to listening. Some prefer diagrams to text. Preferences are real.
Component 2: Learning outcomes improve when instruction is matched to the learner's preferred style. This is the testable claim — the meshing hypothesis. If you're a "visual learner," you should learn more from visual instruction than from auditory instruction, while an "auditory learner" should show the opposite pattern. This is the claim that has been tested and has failed.
The distinction between these two components is crucial. Learning style advocates often defend the concept by pointing to Component 1 (preferences exist), when the actual question is about Component 2 (matching improves outcomes). Yes, preferences exist. No, matching instruction to those preferences does not reliably improve learning.
The Evidence Against the Meshing Hypothesis
Pashler et al. (2008): The Definitive Review
Pashler, McDaniel, Rohrer, and Bjork published the most comprehensive review of learning styles evidence in Psychological Science in the Public Interest — the same prestigious review journal that later published the brain training review. Their conclusion:
"Although the literature on learning styles is enormous, very few studies have even used an appropriate research design to test the validity of learning styles applied to education. Moreover, of those that did use an appropriate method, several found results that flatly contradict the popular meshing hypothesis."
The key finding: when properly controlled studies (using the crossover interaction design — the gold standard for testing the meshing hypothesis) are conducted, matching instruction to learning style does not improve learning outcomes.
What a Proper Test Looks Like
To test the meshing hypothesis, you need a specific experimental design:
- Assess students' learning styles
- Randomly assign students to receive instruction in either their preferred or non-preferred style
- Test all students on the same material
- Look for an interaction: visual learners should do better with visual instruction, AND auditory learners should do better with auditory instruction
Most studies that claim to support learning styles don't use this design. They show that students prefer one style (Component 1), or that one instructional method works better for all students — but not the critical interaction that the meshing hypothesis predicts.
Subsequent Tests
Since Pashler et al., multiple rigorous studies have continued to test and fail to support the meshing hypothesis:
Rogowsky, Calhoun, and Tallal (2015): Assessed 121 adults' learning styles, then randomly assigned them to learn material through either audiobook or e-text (matching or mismatching their preferred style). Result: no interaction. Learning style preference did not predict which format produced better learning.
Husmann and O'Loughlin (2019): Tested 426 anatomy students. Most students who identified as visual learners did not actually study in visual ways (they used the same study strategies as everyone else). And students who did study in their "preferred" style showed no performance advantage.
Knoll et al. (2017): Random assignment, proper crossover design, measured learning outcomes. No support for the meshing hypothesis.
The pattern is consistent: when the meshing hypothesis is properly tested, it fails.
Why 95% of Teachers Still Believe It
If the evidence is this clear, why does the myth persist among the people who would most benefit from knowing the truth?
1. It Feels Right
The learning styles framework has enormous intuitive appeal. Of course people are different. Of course they have different preferences. Of course instruction should be personalized. The logical step from "people differ" to "instruction should be matched" seems obvious — but the intuitive logic doesn't match the empirical reality.
2. Confirmation Bias in the Classroom
Teachers who believe in learning styles interpret classroom events through that lens. A visual student who does well after a diagram-heavy lesson confirms the theory. A visual student who does poorly after a lecture is attributed to the style mismatch. The teacher never systematically compares matched vs. mismatched instruction with controlled assessments, so the belief is never tested against the data — it's tested against selective memory.
3. Teacher Training Programs
Learning styles are taught in many teacher training programs as established practice. Teachers learn the framework from authority figures (professors, training materials) and are told it's evidence-based. The myth is institutionally transmitted.
4. The Learning Styles Industry
The learning styles framework supports an industry: assessment tools (VARK, Kolb's Learning Styles Inventory, Honey and Mumford's Learning Styles Questionnaire), training programs, educational materials, and consulting services. The VARK questionnaire alone has been taken millions of times. The industry has financial incentives to maintain the framework regardless of the evidence.
5. It's Appealing to Students
"I'm a visual learner" provides students with a self-understanding framework (Chapter 1's need for self-knowledge) and, sometimes, an excuse for poor performance ("I did badly because the teaching didn't match my style"). The label serves identity functions that persist independently of its accuracy.
What Actually Works: Evidence-Based Learning Strategies
If learning styles don't predict learning outcomes, what does? Decades of cognitive psychology research have identified strategies that reliably improve learning for all students, regardless of preferences:
1. Retrieval Practice (Testing Effect)
Actively retrieving information from memory — through practice tests, flashcards, or self-quizzing — is one of the most robust findings in the science of learning. It improves long-term retention more than re-reading, highlighting, or summarizing. The effect has been replicated hundreds of times.
Why it works: Retrieving information strengthens the memory trace. Each retrieval makes the information easier to access in the future. Passive review (re-reading) creates a feeling of familiarity that is often mistaken for understanding.
2. Spaced Practice (Distributed Practice)
Spreading study sessions out over time produces better long-term retention than cramming the same amount of study into a single session. This is the spacing effect, one of the most replicated findings in experimental psychology.
Why it works: Spaced practice forces the brain to repeatedly reconstruct the memory, which strengthens it. Massed practice (cramming) relies on short-term memory that fades quickly.
3. Interleaving
Mixing different topics or problem types during practice improves the ability to discriminate between concepts and apply the right strategy. This is counterintuitive — it feels harder during practice but produces better transfer.
4. Elaboration
Connecting new information to existing knowledge, generating explanations, and asking "why" and "how" questions deepens processing and improves retention.
5. Dual Coding
Combining verbal and visual representations of the same information improves learning for everyone — not because some people are "visual learners" but because having two ways to access the same information provides redundancy.
The key insight: Dual coding helps everyone, not just "visual learners." This is the important distinction. The question was never "are visuals useful?" (they are). The question was "do visual learners benefit more from visuals than auditory learners do?" (they don't).
The Nuanced Truth
Learning preferences are real. People genuinely differ in what they prefer. Some prefer videos, some prefer text, some prefer hands-on activities. These preferences are worth respecting for engagement and motivation.
Learning style matching doesn't improve outcomes. The meshing hypothesis — the specific, testable claim that matching instruction to style improves learning — has failed rigorous testing repeatedly.
Universal strategies work better than style-matching. Retrieval practice, spaced practice, interleaving, elaboration, and dual coding improve learning for all students, regardless of preferred style.
Multi-modal instruction benefits everyone. Presenting material in multiple formats (text + diagrams + discussion + practice) is good pedagogy — not because it "hits all learning styles" but because redundancy and multi-modal encoding enhance memory for everyone.
The replacement is more useful than the myth. Instead of "figure out your learning style and demand matched instruction," the evidence says: "use retrieval practice, space your studying, interleave topics, and engage with material in multiple formats." This advice works for everyone and is supported by decades of rigorous research.
Verdict: "People learn better when instruction matches their learning style" ❌ DEBUNKED — The meshing hypothesis has been tested in multiple rigorous studies using appropriate crossover designs and has consistently failed. Learning preferences are real, but matching instruction to preferences does not improve learning outcomes. The most authoritative review (Pashler et al., 2008) found no adequate support for the hypothesis. Origin: Multiple competing models (VARK, Kolb, Honey & Mumford) developed commercially and academically. The idea has been taught in teacher training for decades. Evidence: Pashler et al. (2008), Rogowsky et al. (2015), Husmann & O'Loughlin (2019), Knoll et al. (2017) — all fail to support the meshing hypothesis.
Verdict: "If learning styles don't work, then all students learn the same way" ❌ DEBUNKED — This is a false dichotomy. Students differ in many ways (prior knowledge, motivation, cognitive abilities, engagement), and good teaching is responsive to those differences. But the specific claim that modality-matching improves outcomes is not supported. Universal evidence-based strategies (retrieval practice, spaced practice, dual coding) work better than style matching.
Fact-Check Portfolio: Chapter 12
If any of your 10 claims involve learning, education, or cognitive strategies: - Does the claim distinguish between preferences (real) and outcomes (what matters)? - Is the claim tested with a proper crossover design, or does it just show that people have preferences? - Does the claim involve an evidence-based learning strategy or a debunked one? - Who profits from the learning styles framework?
After Reading: Confidence Revisited
- "People learn better when instruction matches their style." — What does the crossover interaction design show?
- "I know my own learning style." — You may know your preference. Does that preference predict better outcomes?
- "Learning styles are supported by brain science." — Is there any neuroimaging evidence for the meshing hypothesis?
- "Teachers should tailor instruction to different learning styles." — What should teachers do instead?
- "If learning styles don't work, then all students learn the same way." — What is the false dichotomy here?