Why Rereading Notes Doesn't Work (And What Actually Does)

Try this with your child tonight: hand them a coin — any coin they've seen a thousand times — and ask them to pick the correct drawing of it out of a handful of near-identical fakes. Most people get it wrong. Not because they've never looked closely at a coin, but because looking isn't the same as remembering.

That's the "penny test," and it's a clean demonstration of something a lot of study habits get wrong: re-encountering information is not the same as learning it.

The myth of repetition

Rereading notes, highlighting a textbook, watching a lecture a second time — these all feel productive. They're also, according to learning-science research, close to useless on their own. The reason is that rereading is cognitively easy. Your brain recognizes the material ("yes, I've seen this before") without ever being forced to reconstruct it from scratch. And recognition is a much weaker, much less durable form of memory than recall.

What actually works is retrieval practice — being asked to produce the answer, not just re-see it. Closing the book and trying to explain a concept out loud. Answering a question without the notes in front of you. Getting it wrong and having to work out why. All of these are cognitively effortful, and that effort is exactly what makes the memory stick.

Why effort matters

Psychologists Robert and Elizabeth Bjork call this "desirable difficulty" — the idea that a certain amount of struggle, at the right level, is what actually drives learning. Too easy, and there's no benefit. Too hard, and it's just frustrating. But retrieval that's hard enough to require real effort is precisely what turns a fragile, fading memory into a stable one.

There's real neuroscience behind this. A 2018 fMRI study (Zhan, Guo, Chen & Yang, published in Frontiers in Human Neuroscience) had adults learn a set of face–scene pairs — some just once, others six times, spaced out. A month later, the once-learned pairs had degraded badly, and the brain had shifted to relying on effortful, less reliable prefrontal "search" circuitry to reconstruct them. The six-times-learned pairs, by contrast, were still stable and hippocampus-backed — the kind of low-effort, reliable memory you actually want. The behavioral gap was large: roughly 75% recognition accuracy for the repeated group versus 55% for the single-exposure group.

The takeaway isn't "repeat things more." It's that a single practice session, no matter how well it felt, is not durable learning. The repetition has to be spaced out, and each pass has to involve actual retrieval — not just looking again.

What this means for how kids should actually study

• Close the book before testing yourself. If the material is in front of you while you "check" your understanding, you're not retrieving — you're just recognizing.
• Get it wrong sometimes. A wrong answer during practice isn't a failure signal — it's often a sign the difficulty was calibrated correctly.
• Space it out. Reviewing the same thing five times in one sitting is far less effective than reviewing it five times across five days.
• Vary the question, not just the schedule. Being asked to define a concept, then apply it to a new problem, then explain it to someone else, is stronger than seeing the identical question three times.

None of this requires expensive tools — a stack of flashcards used correctly (cover the answer, force recall, check after) applies the same principle a sophisticated app would. What matters is the mechanism, not the medium: effortful, spaced, varied retrieval — not passive re-exposure.

---

Further reading: Zhan, J., Guo, R., Chen, R., & Yang, Y. (2018). "Effects of Repetition Learning on Associative Recognition Over Time: Role of the Hippocampus and Prefrontal Cortex." Frontiers in Human Neuroscience. Bjork, R. A., & Bjork, E. L. (2011). "Making things hard on yourself, but in a good way: Creating desirable difficulties to enhance learning."