The Four Conditions for Changing a Mind
A student can hand you back the exact right answer on a Friday quiz and still believe the wrong thing underneath it. That's not a rare failure of your teaching. It's the default outcome of most instruction, on most topics, most of the time — and it turns out science educators and misinformation researchers have been independently proving the same point for over forty years, from two different directions, without citing each other. This page is the theory behind the classroom move Teaching #6 already walked through. If that page was the thirty-second response, this one is why the thirty-second response works.
Two research communities, one finding
Science-education researchers have spent decades studying why a correct explanation so often fails to dislodge a student's existing wrong idea. Misinformation researchers have spent the last fifteen years studying why a correct fact so often fails to dislodge an adult's false belief. Read side by side, their conclusions are close to identical: telling someone the right answer is necessary but nowhere near sufficient. Something specific has to happen to the wrong answer first, or the correction just gets filed next to it instead of replacing it.
The four conditions
In 1982, George Posner, Kenneth Strike, Peter Hewson, and William Gertzog published what became the founding paper of “conceptual change” theory in science education. Their claim: a learner only trades an old conception for a new one when four conditions are all met.
- Dissatisfaction. The learner has to genuinely doubt the idea they're currently holding — not just hear that it's wrong, but run into a case where it visibly fails.
- Intelligibility. The replacement idea has to be something the learner can actually picture and state back in their own words, not just repeat.
- Plausibility. The replacement has to seem like it could be true, given everything else the learner already believes. An idea that contradicts too much at once gets rejected outright, however correct it is.
- Fruitfulness. The replacement has to do something useful — explain a case, predict an outcome, solve a problem — that the old idea couldn't.
Miss any one of the four, and the old idea usually survives, sometimes hiding directly underneath a correct-sounding answer.
Documented caseThe Force Concept Inventory (1992)
David Hestenes, Malcolm Wells, and Gregg Swackhamer built a thirty-question survey of commonsense beliefs about motion and gave it to students before and after a full semester of introductory physics. The finding held across instructors, institutions, and teaching styles: conventional lecture-based instruction produced almost no change in students' underlying, non-Newtonian beliefs about force and motion. Students had learned to produce the textbook answer. The intuitive physics underneath it — the belief that a moving object needs a continuous force to keep moving, for instance — was untouched. Nothing in a standard lecture had created real dissatisfaction with it.
Documented caseSynthetic models of the earth (1992)
Stavros Vosniadou and William Brewer interviewed elementary schoolers about the shape of the earth, including generative questions like “if you walked in a straight line for many days, where would you end up?” Very few children simply believed a flat earth after being taught otherwise, and very few gave the textbook-correct answer either. Most built what Vosniadou and Brewer called a synthetic model — a hybrid that kept the flat, supported ground their direct experience insisted on, while accommodating the new claim that the earth is round: a flat disk inside a hollow sphere, or two earths, one flat to stand on and one round in the sky. The children weren't refusing to learn. They were doing exactly what a mind does when a replacement idea isn't yet plausible enough to fully displace the old one — they merged the two rather than swapping them.
What the misinformation researchers found, working separately
Stephan Lewandowsky and John Cook's Debunking Handbook (2020) synthesizes the psychological research on correcting false beliefs in adults, and its practical guidance maps onto the same four conditions almost exactly, aimed at a different age group and a different subject matter. Their core recommendations: warn the person before the false claim is even repeated, so the myth doesn't get reinforced by its own retelling (a dissatisfaction move — undercut the old belief's credibility before offering anything new); and always pair a correction with a real alternative explanation that accounts for whatever the false belief was doing for the person, rather than leaving a gap where the wrong answer used to sit (intelligibility, plausibility, and fruitfulness, in adult form). Simply repeating “false, false, false” next to the correct fact is one of the least effective corrections tested — the same failure mode the Force Concept Inventory documented in a physics classroom forty years earlier.
Turning this into a four-step classroom sequence
- Elicit the existing belief out loud, before correcting anything. You can't create dissatisfaction with an idea nobody has said yet. Ask “what do you think is happening” before “here's what's actually happening” — the same move Teaching #6's fan story uses.
- Build or find a real discrepant event. A demonstration, a data set, or a direct question that the student's current belief can't actually explain. This is the dissatisfaction step, and it has to be experienced, not just asserted.
- Hand them a replacement they can state back in their own words, and that connects to something they already accept. If the new idea requires throwing out too much at once, expect a synthetic model — a hybrid, not a replacement — rather than outright rejection.
- Let them use the new idea to do something the old one couldn't. Predict a result, solve a problem, explain a second discrepant case. This is the step most lesson plans skip, and it's the one that makes the change durable instead of temporary.
A diagnostic table: when a correction doesn't take, which condition is usually missing
| What you observe | Likely missing condition |
|---|---|
| Student states the correct answer verbatim on a test, reverts to the old belief in conversation | Dissatisfaction — the old idea was never actually challenged, just outvoted |
| Student can't restate the new idea in their own words | Intelligibility |
| Student builds a hybrid answer that tries to keep both ideas at once | Plausibility — the new idea contradicted too much existing belief at once |
| Student can state the new idea correctly but doesn't reach for it when it would actually help | Fruitfulness — the new idea hasn't yet done anything the old one couldn't |
Questions for reflection
- Name Posner, Strike, Hewson, and Gertzog's four conditions, in order. Remember
- Explain why a student can pass a quiz and still hold the misconception the quiz was testing for. Understand
- Pick one misconception common in your own course. Design a single discrepant event that would create real dissatisfaction with it — not just tell students it's wrong. Apply
- Compare the Force Concept Inventory finding and the synthetic-earth-model finding — what does each reveal about a different one of the four conditions failing? Analyze
- Evaluate your own most recent attempt to correct a misconception in class against all four conditions. Which one, if any, did you skip? Evaluate
- Design a short correction — for a student misconception or a piece of public misinformation on a topic of your choice — that satisfies all four conditions in under two minutes of class time. Create
George J. Posner, Kenneth A. Strike, Peter W. Hewson, and William A. Gertzog, “Accommodation of a Scientific Conception: Toward a Theory of Conceptual Change,” Science Education, 1982; David Hestenes, Malcolm Wells, and Gregg Swackhamer, “Force Concept Inventory,” The Physics Teacher, 1992; Stavros Vosniadou and William F. Brewer, “Mental Models of the Earth: A Study of Conceptual Change in Childhood,” Cognitive Psychology, 1992; Stephan Lewandowsky and John Cook, The Debunking Handbook 2020. Companion piece: Teaching #6, “Talking About Misinformation Without Talking Politics: A Classroom Guide to Chapters 11–12,” for the in-the-moment classroom version of this same research.