Case Study: Learning Styles — Anatomy of an Unkillable Zombie

The Origin

The learning styles concept has no single origin — it emerged from multiple sources in the 1970s and 1980s as researchers explored individual differences in learning. David Kolb's experiential learning model (1984), Howard Gardner's multiple intelligences theory (1983), and Neil Fleming's VARK model (1987) all contributed to the general idea that students have distinct learning preferences that should shape instruction.

The VARK model became the most popular version, perhaps because of its simplicity: Visual, Auditory, Reading/Writing, Kinesthetic. It offered teachers a clear diagnostic tool (administer the VARK questionnaire) and a clear prescription (match instruction to the student's dominant style). It was easy to understand, easy to implement, and easy to train.

By the 1990s, learning styles had become one of the most widely taught concepts in teacher preparation programs worldwide. It was embedded in professional development curricula, educational materials, and school improvement plans. An entire industry of assessments, training programs, and consultants had developed around it.

The Debunking

The systematic debunking of learning styles began in earnest in the 2000s:

Coffield et al. (2004) conducted the most comprehensive review, examining 71 different learning style models. They found that most models lacked basic psychometric properties — the instruments didn't reliably produce the same results when the same person was tested twice (test-retest reliability), and the categories didn't predict learning outcomes (predictive validity). They concluded that the "enormous" popularity of learning styles was "not justified by the evidence."

Pashler et al. (2008) published in Psychological Science in the Public Interest — one of the most prestigious psychology journals. They established a rigorous evidential standard: to validate the meshing hypothesis, you need a crossover interaction design — randomly assign students to matched or mismatched instruction and show that matched students outperform. They found virtually no studies meeting this standard. Their conclusion: "There is no adequate evidence base to justify incorporating learning-styles assessments into general educational practice."

Willingham, Hughes, and Dobolyi (2015) reviewed the evidence again and reached the same conclusion. They noted that the learning styles hypothesis remained "remarkably popular" despite a near-total absence of supporting evidence.

Husmann and O'Loughlin (2019) tested whether students classified by VARK actually studied in ways consistent with their supposed style. They didn't — and studying in one's "preferred" style didn't correlate with better outcomes.

The Persistence

Despite this overwhelming evidence, learning styles remains entrenched:

• Surveys from 2020 found 64% of U.S. higher education faculty believing in the meshing hypothesis
• International surveys consistently find 80-95% of K-12 teachers believing in learning styles
• Teacher preparation programs in multiple countries continue to teach learning styles
• Professional development workshops continue to offer learning styles training
• Educational materials continue to reference learning styles

Failure Mode Analysis

The Plausible Story Problem (Ch.6)

Learning styles offers a coherent narrative: students are different → they have different learning preferences → matching instruction to preferences improves learning. Each step feels intuitively correct. The first two steps are correct — students do differ, and people do have preferences. The third step is the unsupported leap — but because the first two steps are true, the third feels true by extension.

The Zombie Properties (Ch.16)

Learning styles possesses every property in the zombie resilience taxonomy: 1. Intuitive appeal: It aligns with personal experience 2. Actionable framework: It gives teachers something specific to do 3. Identity function: It validates both teachers (as diagnosticians) and students (as having a "type") 4. Commercial infrastructure: Multimillion-dollar industry of assessments and training 5. Training perpetuation: Taught in teacher preparation programs to new cohorts every year 6. No visible cost: Believing and acting on it doesn't produce obvious harm 7. Resistant to falsification: Any failure to find the meshing effect can be attributed to poor implementation, wrong assessment instrument, or insufficient differentiation

The Consensus Enforcement (Ch.14)

Within many schools and teacher preparation programs, learning styles has achieved the status of received wisdom. A new teacher who questions learning styles may face social pressure from colleagues and mentors who consider it a foundational element of good teaching. Questioning it can be perceived as laziness ("you don't want to differentiate instruction") rather than as legitimate skepticism.

The Sunk Cost (Ch.9)

School districts that have invested in learning styles training, assessments, and materials have institutional sunk costs. Professional development providers who have built their businesses around learning styles have financial sunk costs. Individual teachers who have spent years differentiating instruction based on learning styles have personal career investment. Acknowledging that the approach doesn't work means acknowledging that these investments were misdirected.

What It Would Take to Kill It

Based on the failure mode analysis, debunking alone will not kill learning styles. The evidence has been clear for over two decades. What would be needed:

1. Remove it from teacher preparation accreditation standards. If accrediting bodies required that teacher preparation programs teach evidence-based practices rather than learning styles, the training pipeline would change.

2. Provide an equally simple alternative. Learning styles persists partly because debunking removes a framework without providing a replacement. Teachers need actionable strategies for differentiation. Evidence-based alternatives (retrieval practice, spaced repetition, interleaving, worked examples) need to be packaged with the same simplicity and actionability.

3. Change the commercial incentive. As long as there is a market for learning styles products, companies will produce them. Procurement policies that require evidence of efficacy could shift the market.

4. Make the cost visible. The current cost of learning styles is invisible (slightly suboptimal instruction). If the opportunity cost — the evidence-based practices that are displaced by learning styles training — could be made visible, the urgency of correction would increase.

Analysis Questions

1. The Pashler et al. review was published in one of psychology's most prestigious journals. It received widespread media coverage. It reached a clear, unambiguous conclusion. And it had essentially no effect on practice. Apply the evidence-practice gap analysis from this chapter: what structural features of education prevented this high-profile debunking from changing behavior?

2. Compare learning styles to the dietary fat hypothesis (Chapter 26). Both are wrong ideas that persisted for decades despite evidence against them. But the dietary fat hypothesis is now widely recognized as at least partially wrong, while learning styles persists nearly unchanged. What structural differences explain why nutrition has corrected (partially) while education has not?

3. Design a "learning styles sunset plan" — a realistic multi-year strategy for removing learning styles from teacher preparation and professional development. Identify the stakeholders who would need to be engaged, the institutional changes required, and the resistance points. Use the Correction Speed Model to estimate how long the process would take.