Correcting Science misconceptions through contrastive teaching

The honest truth about reteaching is that the obvious version of it does not work very well. You notice a misconception in the marking, you set aside time, you explain the topic again — often more slowly and more carefully than the first time — and a few weeks later the same wrong answer is back. The effort was real. The result was thin. It is one of the most demoralising patterns in teaching, and it is not a sign that you explained badly.

The problem is structural. A second clear explanation, on its own, tends to add a correct idea next to the wrong one rather than removing it. The student leaves holding both. Contrastive teaching is the alternative — and it is less about explaining better and more about arranging the lesson so the student's own wrong idea is the thing that gets challenged. This guide is how to do it.

Why correction has to start with the wrong idea

The most cited account of how wrong ideas change (Posner et al., 1982) sets out four conditions. For a student to give up a misconception, they first have to become dissatisfied with it; then the correct idea has to be intelligible, plausible, and fruitful. The order is the whole point. Dissatisfaction comes first — and a fresh explanation of the right answer does nothing to make the student dissatisfied with their existing one. Nothing has put the old idea under pressure.

So contrastive teaching does something that feels slightly counter-intuitive: it brings the wrong idea into the open on purpose. Not to embarrass anyone, and not to leave it lying around, but to set up a moment where the student sees — concretely, in front of them — that their idea makes a prediction that does not come true. That gap between "what I expected" and "what actually happened" is the dissatisfaction the research describes. It is what gives the correct idea a job to do.

The shape of a contrastive correction

A workable contrastive sequence has four moves. It maps closely onto Predict–Observe–Explain (White & Gunstone, 1992) and onto the explicit, structured reteaching that Rosenshine (2012) describes.

1. Surface — get the wrong idea into the open

Pose the situation and ask students to predict and commit to a reason. "When this heavier ball and this lighter ball are dropped together, which lands first — and why?" The prediction is not a test; it is the wrong idea stepping forward where everyone, including its owner, can see it. Commitment matters: a student who has said out loud "the heavy one, because it is heavier" is far more invested in the outcome than one watching passively.

2. Confront — create the conflict

Now show the case where the prediction fails. The demonstration, the data, the worked counter-example — whatever makes the gap unmissable. The aim is the small jolt of "wait, that is not what I expected." This is the active ingredient. Without it, you are back to re-explaining.

3. Replace — name the correct idea, side by side

Conflict alone is not enough; a confused student is not a corrected one. Immediately resolve it by stating the correct idea explicitly and next to the wrong one: "The idea that heavier means faster predicts X. What we saw was Y. The reason is…" Putting the two ideas in direct contrast is what makes the correct one not just heard but preferred. Leaving students to infer the correct idea on their own is where many otherwise-good reteaches quietly fail.

4. Apply — use the correct idea on something new

Finish on a fresh example — deliberately not the one you just used. Re-using the original case lets a student pattern-match the answer without re-using the idea. A new context forces the correct idea to do real work, which both strengthens it and shows you whether it actually took. This is the "fruitful" condition from Posner et al. in practice.

What this looks like in fifteen minutes

Contrastive teaching is often imagined as elaborate. It does not have to be. For a class that thinks a plant gets "all its food from the soil":

• Surface: "A plant doubles in mass over a few months. Where did the extra material come from? Commit to an answer." Most will say the soil.
• Confront: Show the classic result that the soil's mass barely changes — the extra mass cannot have come mainly from it.
• Replace: State it directly. "Most of the new material comes from a gas — carbon dioxide from the air — built up using light. The soil idea predicts the soil should shrink a lot. It does not."
• Apply: "So if we grew the same plant in the dark with plenty of soil and water, what would you now predict — and why?"

Fifteen minutes, one idea, one contrast, one fresh application. That is the unit. It fits inside the single remedial session this approach is usually delivered through.

Common correction mistakes

• Re-explaining without conflict. The most common one. Clearer is not the same as contrastive.
• Surfacing the wrong idea and leaving it there. Conflict without an explicit, named correction can deepen confusion. Always close on the correct idea, stated plainly.
• Re-using the same example to "check". It tests recall of an answer, not transfer of an idea. Change the context.
• Correcting ten things at once. One misconception, done properly, beats a tour of five. Pick the one your diagnosis flagged as most common and most central.
• Skipping the re-check. A misconception displaced today can re-form under different wording next week. The correction is not finished until a fresh-context re-check confirms it held.

An honest boundary

Contrastive teaching is a strong approach, not a guarantee. Some misconceptions are deeply held and take more than one well-built session to shift; some students will need the contrast more than once, in more than one context, before the correct idea becomes their default. The structure here makes correction more likely to stick — it does not promise that one lesson clears a stubborn idea, and it should not be presented that way. Your ongoing read of the class, and the re-check, are what tell you whether more is needed.

If turning a stack of marked scripts into the one misconception worth correcting is the slow part of your week, that is what MyScienceHOD is built to support — you keep every teaching decision; it just helps the pattern surface faster. The free Beta is open to Singapore Science teachers and departments.