Case Study 1: Learning Styles — The Zombie That Won't Die
The Debunking Timeline
| Year | Event | Impact on Zombie |
|---|---|---|
| 2004 | Coffield et al. review: 71 learning style models, none validated | Minimal — review not widely read by teachers |
| 2008 | Pashler et al. systematic review: "virtually no evidence" | Moderate among researchers; negligible among practitioners |
| 2010 | Riener & Willingham popular article in Change magazine | Some reach to higher education |
| 2015 | Rogowsky et al.: direct experimental test fails | Further confirmation; no change in practice |
| 2017 | Newton & Miah: 89% of recent papers still endorse learning styles | Evidence that debunking has failed |
| 2020 | Nancekivell et al.: 90% of teachers still believe | Zombie confirmed as unkillable by current methods |
Why Evidence Failed
The learning styles case is the clearest demonstration that evidence is necessary but not sufficient for killing a zombie idea. The evidence is as strong as debunking evidence ever gets: systematic reviews, direct experimental tests, meta-analyses, and expert consensus all agree. Yet the idea persists at essentially the same prevalence as before the debunking began.
The evidence failed because it targeted only the evidential basis of the belief. It did not target: - The institutional embedding (teacher training curricula) - The commercial ecosystem (EdTech products and assessment tools) - The intuitive appeal (personal experience of learning preferences) - The narrative satisfaction (the inclusive story of "everyone learns differently") - The simplicity (three categories vs. complex, multifactorial learning) - The identity investment (teachers who have built practices around it) - The social cost of change (challenging learning styles risks being seen as anti-student)
Each of these non-evidential supports maintains the zombie independently of the evidence. The debunking removed one pillar (evidential support); the structure stands on six others.
The Replacement Challenge
The most promising approach to killing the learning styles zombie is replacement: providing teachers with evidence-based differentiation strategies that satisfy the same needs.
The needs learning styles serves: - Personalization: Teachers want to tailor instruction to individual students - Explanation: Teachers want a framework for understanding why some students struggle - Action: Teachers want specific, implementable strategies - Identity: Teachers want to be the kind of teacher who cares about individual differences
Evidence-based replacements exist: retrieval practice, spaced repetition, interleaving, elaborative interrogation, and instruction calibrated to prior knowledge rather than "learning style." These approaches are more effective than learning style matching and can satisfy all four needs.
But the evidence-based approaches are more complex, require more training, and don't come with the simple categorical framework that learning styles provides. The replacement doesn't fit through the simplicity filter as easily as the zombie does.
Discussion Questions
- If evidence can't kill this zombie, what can? Design a comprehensive strategy targeting all seven persistence mechanisms.
- Is there a harm in letting the learning styles zombie live? If teachers believe in learning styles but also use evidence-based methods, does the zombie matter?
- Why did the learning styles zombie infect education but not (for example) aviation training or medical education? What structural features of the educational profession made it vulnerable?
- Apply the prebunking strategy: design a teacher education module that would prevent learning styles from infecting the next generation.
References
- Pashler, H. et al. (2008). "Learning Styles: Concepts and Evidence." Psychological Science in the Public Interest, 9(3), 105–119. (Tier 1)
- Coffield, F. et al. (2004). Learning Styles and Pedagogy in Post-16 Learning. Learning and Skills Research Centre. (Tier 1)
- Nancekivell, S. E. et al. (2020). "The Persistence of Learning Style Beliefs in the US." Frontiers in Education, 5, 94. (Tier 1)