30 January 2026 · Updated 13 August 2026 · Hugo Harrabin
Three hours of revision, and half of it gone by the practice test. Here's why re-reading fails, and the one change that fixes it.
Your child revises for three hours. They go over their notes, highlight the key facts, and feel prepared. Then they sit a practice test and can barely remember half of it.
That isn't a memory problem and it isn't an intelligence problem. It's a method problem, and it has a well-evidenced fix: swap reading for testing. Reading notes builds the feeling of knowing something. Being tested builds the ability to produce it when the page in front of you is blank.
Key Takeaways
- Re-reading and highlighting build recognition, the sense that something looks familiar. Exams ask for recall, producing it from nothing. They're different skills.
- This is why revision feels like it's working when it isn't. While the notes are open the answers are in front of your child, so they overestimate what they'd manage without them.
- Of ten revision methods reviewed by Dunlosky and colleagues (2013), testing yourself was one of only two rated high value. Re-reading, highlighting and summarising were all rated low.
- Good retrieval practice feels harder than re-reading. That difficulty is the mechanism, not a sign it isn't working.
- Retrieval sets how your child revises. Pair it with spacing, which sets when. Separate ideas, and the evidence supports the combination.
Because they never practised remembering. They practised recognising.
Reading a page of notes produces a feeling of familiarity. Yes, I've seen this formula, I know this one. That feeling is misleading because it isn't the same thing as writing the formula on a blank page in an exam hall. Psychologists separate the two: recognition is identifying something when it's in front of you, recall is producing it when it isn't. Revision by re-reading trains the first. Every GCSE paper tests the second.
There's a specific reason students misjudge this so badly. Koriat and Bjork (2005) showed that while you're studying, the answer is sitting there in view, which makes the material feel obvious and inflates your confidence about how well you've learned it. They called it a foresight bias: at the moment of study you can't easily imagine the conditions of the test, where the answer will be missing. Your child isn't lying when they say they knew it. They genuinely felt they did.
This Is Not Your Child's Fault: schools rarely teach students how to revise. Most default to re-reading because it's what gets modelled. The useful part is that the fix is a change of method rather than a change of effort, so it doesn't mean more hours.
This has been tested directly, and at scale. Dunlosky and colleagues (2013) reviewed ten common study techniques for Psychological Science in the Public Interest and rated each on how much it actually helps. Only two earned a high rating. Five were rated low, and those five include the three things most students spend their revision doing.
Testing yourself came top. Re-reading came bottom.
Two of the other techniques on that chart are worth a look, because they answer different questions from this one. Spaced practice shares the top rating and decides when revision happens. Interleaving sits in the middle band and decides what order topics come in. Get all three right and they compound; this post is only the first of them.
The clearest single demonstration of why comes from Roediger and Karpicke (2006), who had students read passages and then either reread them or sit practice recall tests. Tested five minutes later, the rereaders did better. Tested a week later the pattern reversed, and the students who had been tested retained substantially more. That reversal is the whole problem in one experiment: the method that looks better today is the one that loses by June. Re-reading buys short-term fluency at the cost of long-term memory.
Retrieval practice means trying to pull information out of memory instead of putting it back in front of your eyes. It takes four practical forms, and none of them need buying.
| Method | What it looks like | Best for |
|---|---|---|
| Blank page recall | Close everything, write down all you can remember about a topic, then check what's missing | Content-heavy subjects: biology, history, geography |
| Flash cards | Question one side, answer the other. Say the answer before turning it over | Definitions, formulae, vocabulary, dates |
| Past paper questions | Timed, no notes open, then marked honestly against the scheme | Exam technique, and applying a method they already know |
| Explaining it aloud | Teach the topic to someone who doesn't know it, including a parent who doesn't know the subject | Finding the gaps your child can't see themselves |
The common thread is that the answer isn't visible. The moment your child can see it, they're back to recognition.
The Key Principle: trying to retrieve something and failing, then checking, beats recognising it comfortably on the page. The effort of reaching for it is what strengthens the memory.
This is the part that trips families up, because it inverts the obvious signal. Bjork and Bjork (2011) describe desirable difficulties: conditions that make learning feel harder and slower at the time and produce better retention later.
So when your child says flash cards are horrible and re-reading their notes feels fine, they're describing the mechanism working correctly. Fluency feels like competence and usually isn't. A session that felt uncomfortable has often done more than a smooth one.
Worth saying out loud to them, because otherwise the harder method looks like the failing one and gets abandoned on day one.
Four swaps, in rough order of how much they'll help:
If only one of these happens, make it the first. It costs nothing and it converts an existing habit rather than adding a new one.
Once the method is right, the timing of those reviews is the next thing worth getting right, and mixing topics rather than working through them in blocks is the third.
Every Tugo session opens with retrieval rather than new content: questions on material from previous sessions, answered without notes. Students don't just mark them right or wrong. The tutor works through why an answer was wrong, which is where most of the learning sits.
It's deliberately structured that way because retrieval is the hardest technique to keep up alone. It feels worse than re-reading, the benefit is invisible for weeks, and almost nobody sustains it on willpower. Building it into a scheduled session removes the need to. Retrieval is one pillar of the Tugo Method, alongside spacing and interleaving.
Most likely because they're revising by re-reading, highlighting or watching videos. Those build familiarity rather than recall, so material feels learned while the notes are open and disappears when they aren't. Switching to testing without notes fixes it, usually within a few weeks.
Testing yourself without notes. In the Dunlosky review it was one of only two techniques out of ten rated high value, the other being spaced practice. For most GCSE subjects, timed past paper questions marked honestly is the single highest-value thing your child can do.
Usually they mean flash cards feel unpleasant, which is expected and is rather the point. Two things genuinely do go wrong though: cards so easy the answer is guessable from the question, and cards used only in the fortnight before an exam. Neither is a problem with the method itself.
No, and it isn't a substitute for being taught something properly in the first place. Retrieval practice strengthens and exposes what has already been understood. If a topic was never understood, testing surfaces that quickly, which is useful information rather than a failure of the technique.
Yes. Every session opens with a retrieval segment drawn from earlier material, and tutors are trained to work through the reasoning behind wrong answers rather than simply correcting them. It's built into the session structure rather than left to the individual tutor.
Sit in on a lesson led by one of our expert PGCE qualified teachers.