You have probably seen this happen at your kitchen table. Your child works through a page of apostrophe questions. By question five they have got the hang of it. By question twenty they are getting every single one right, quickly and confidently, and everyone feels good about it.

A week later, in a mixed piece of writing, the apostrophes are gone.

Nothing went wrong. Your child was not being lazy and you did not pick bad questions. The problem is the shape of the practice itself, and it is one of the best documented findings in the research on how children learn.

What a page of the same question actually teaches

Here is the thing about twenty apostrophe questions in a row: after the first one, your child never has to work out that it is an apostrophe question.

They know what the page is about. Every answer uses the same rule, so the only work left is applying it. The hardest part of the job, noticing which rule this particular question needs, has been done for them by the layout of the worksheet.

Then the exam arrives, or the piece of writing, and every question is jumbled up. Now they have to spot what kind of question it is before they can answer it. That is a completely different skill, and a page of identical questions never asks for it once.

Researchers call practising in single-topic blocks blocked practice. Mixing different types of question together is interleaved practice. The difference between them is much bigger than it sounds.

The evidence

In a study by Kelli Taylor and Doug Rohrer, published in Applied Cognitive Psychology in 2010, a group of nine and ten year olds learned four different formulas to do with prisms. Half practised in blocks, all of one type before moving on. Half did exactly the same problems, mixed up.

Tested a day later, the mixed group scored 77 percent. The blocked group scored 38 percent.

That is roughly double, from the same problems, in the same amount of time. The only difference was the order. And when the researchers looked at where the blocked group lost their marks, it was mostly on choosing the wrong formula: exactly the skill their practice had quietly done for them.

It is not just a laboratory finding. In 2020, Rohrer and colleagues ran a proper randomised controlled trial in real schools, published in the Journal of Educational Psychology. Seven hundred and eighty seven pupils across fifty four classes did four months of ordinary maths homework, either blocked or interleaved. A month after it finished they sat an unannounced test.

The interleaved group scored 61 percent. The blocked group scored 38 percent.

Same curriculum, same teachers, same amount of homework. Different order.

Why nobody chooses it

If mixing is so much better, why does almost nobody do it by choice?

Because blocked practice feels better, and it feels better in a way that is very hard to argue with. Your child gets faster within the page. They make fewer mistakes. They look and sound more confident. Everything you can see while it is happening says it is working.

Mixed practice looks worse on all of those measures. It is slower, there are more mistakes, and it feels harder for everyone involved. The gain does not show up until later, which is precisely when you are no longer watching.

Psychologists Nate Kornell and Robert Bjork demonstrated how strong this illusion is in a 2008 study. Learners who had just been tested, and who had personally done better on the interleaved material, still reported believing that the blocked version had taught them more. Their own results were sitting in front of them and it did not change their minds.

So this is not a case of parents doing it wrong. Everyone's instincts point the same way, and the instinct is mistaken.

The honest limits

Mixing is not magic, and anyone selling it that way is overselling it.

You cannot mix something your child has not met yet. Interleaving is for consolidating things that have been taught. When a topic is genuinely new, blocked practice is the right tool: they need to get the idea in place before they can practise telling it apart from anything else. Block to learn it, mix to keep it.

The classroom evidence is more modest than the laboratory evidence. The Education Endowment Foundation reviewed the research on cognitive science in real classrooms and found interleaving promising, but with smaller and less consistent effects once it leaves controlled conditions. Their conclusion is worth borrowing: these principles are useful, and they are not a silver bullet.

Mixed sessions cover less ground per topic. If your child needs to shore up one specific weakness, focused practice on that weakness is still the right answer. Mixing is what you do afterwards, so the repair holds.

What this looks like in The Revision Wizard

We have added Mixed practice to the play screen, and it is on by default.

Tapping it starts a session that draws across the areas your child has already covered rather than drilling one of them. The part that took the work is underneath: each area is served at the level your child has actually reached in it. If they are secure on punctuation but still finding spelling hard, a mixed session gives them harder punctuation questions and gentler spelling ones in the same game. It mixes the topics without flattening them to a single difficulty, because a mix that ignores where a child actually is would just be a worse version of both.

Their recommended next area is still there, one tap away, and it is still the first thing on the screen. Mixed practice sits underneath it as the second option, because for something genuinely new, focused practice is still right.

And if you would rather your child stayed on the focused path, you can turn it off, either for your whole account or for one child at a time, in Settings.

The short version

If your child is getting everything right and it feels easy, the practice is probably too tidy. Ten out of ten on a page of one thing tells you less than seven out of ten on a page of everything.

Mix it up. It will feel worse. That is rather the point.

Sources: Taylor, K. and Rohrer, D. (2010), "The effects of interleaved practice", Applied Cognitive Psychology. Rohrer, D., Dedrick, R. F., Hartwig, M. K. and Cheung, C.-N. (2020), "A randomized controlled trial of interleaved mathematics practice", Journal of Educational Psychology 112(1). Kornell, N. and Bjork, R. A. (2008), "Learning concepts and categories: is spacing the enemy of induction?", Psychological Science. Education Endowment Foundation (2021), "Cognitive science approaches in the classroom: a review of the evidence".