Interleaving: How mixed practice works and when to use it

During practice, a tennis player can hit 50 forehands in a row and look great doing it. In a match, though, the ball comes to the forehand, then the backhand, then short over the net, and the hard part is picking the right shot before you swing. Interleaving is a way to practice for that moment: You mix different kinds of shots or problems in one session, in an order that keeps you guessing, instead of finishing a whole batch of one kind before you start the next.

The same idea works for everyday math, like moving between a "what do I pay after 25% off?" problem and a "what was the full price?" one.

Quick answer: What is interleaving?

• Interleaving, also called mixed practice, means shuffling different kinds of problems or skills together in one session, so you rarely do the same kind twice in a row.

• The opposite is blocked practice, where you do all of one kind, then all of the next.

• Mixing helps most when the kinds look alike and are easy to confuse, such as similar math problems or the paintings of different artists.

• Mixed practice feels harder and scores lower while you do it, yet people who practice this way tend to remember more on a test a week or a month later.

• Blocked practice still has its place, since it did better for learning words, and a few tries in a row also help when a skill is brand new.

How is interleaving different from blocked practice?

With blocked practice, you never have to ask: What kind of problem is this? Interleaving makes you ask it every time. When every problem is the same kind, you stop asking and just repeat the steps you used a moment ago. In real life, answering that question is often the hard part. In the words of the researchers behind a large classroom trial, an interleaved assignment "is arranged so that no 2 consecutive problems require the same strategy."

Here's what that looks like for a tennis player practicing nine shots of three kinds.

The same researchers suggest mixing can also lead you into retrieval practice, since you have to try to recall the right method from memory yourself.

Interleaving isn't the same as spacing, which is about when you come back to something over hours or days. It isn't multitasking either since you still work on one problem at a time.

What are some examples of interleaving?

Interleaving fits any skill where you'd normally drill one kind of thing at a time and later have to tell the kinds apart.

• Everyday math. Shuffle tip, discount, and price-increase problems instead of doing a page of each.

• Telling things apart by sight. Flip through photos of similar birds, plants, or painters' works shuffled together, not one group at a time. In a 2008 study, 120 university students matched new paintings to the right artist more often after mixed study than after grouped study.

• Sport. Have a friend throw fastballs, curveballs, and change-ups in random order, the way they come in a game.

• Music. Play two passages of a piece that start the same way and then split, taking turns between them instead of drilling one of them for half an hour.

• Study for a test. Mix questions from the last few chapters into one practice set, instead of reviewing chapter by chapter.

Is interleaving always better than blocked practice?

Interleaving is not always better than blocked practice, but when the problems are easy to mix up, the gap can be big. In a 2020 trial across 54 seventh-grade math classes, 787 students practiced either with mixed problem types or in blocks. On a surprise test a month after a review, the mixed group scored 61% and the blocked group 38%.

Other material gives different results. A 2019 review of 59 studies found a moderate benefit overall, biggest for paintings and other visual material and small for math. For reading passages that explain an idea, mixing made no clear difference, and for learning words, blocked practice did better.

Skip interleaving, or save it for later, in these cases:

• Word lists and new vocabulary. For new words, use a mnemonic device instead.

• A skill that's brand new. Do a few of the same kind in a row first. The 2020 trial's authors say mixing "might be less effective or too difficult" if you haven't had at least a little blocked practice on a new skill.

• Things that don't look alike. The 2019 review found mixing helped most when the kinds looked alike, so shuffling percent problems with French verbs misses the point.

• A hard new movement. Nate Kornell and Robert Bjork, who ran the painting study, note that grouped practice can work better for learning complex motor skills, so practice a tricky passage or swing on its own until it holds together.

For a list you need to remember in order, a memory palace is the better fit, and you'll find more techniques in these memory training exercises.

How do you plan an interleaved practice session?

Pick three kinds of problems you tend to confuse, try each one on its own a couple of times, and then work through them shuffled so no two problems in a row are the same kind. The session below uses three kinds of percent problems that all talk about "%" or "off" but need different steps.

Work through them in order, in your head or on paper, and check each answer before you move to the next one.

Answers: 1) $30, 2) 20%, 3) $50, 4) 25%, 5) $102, 6) $80, 7) $72, 8) $40, 9) 25%.

Problems 1 and 3 both say "% off," but you multiply for one and divide for the other. The blocked version of this set would be problems 1, 5, and 7, then 2, 4, and 9, then 3, 6, and 8. It would feel easier, and by the third problem you'd stop asking which kind it is.

Mixed problems usually take longer, so a short set like this one is plenty. When one kind gets hard, keep working on it instead of skipping ahead to an easier one.

For a quick way to work out percents in your head, see these 10 mental math tricks. For a mixed round you don't have to write yourself, The Mind Company's free mental math practice game shuffles adding, subtracting, multiplying, dividing, and percents in one 60-second round.

How can you tell interleaving is working?

Your score during practice is the wrong thing to watch. In a 2007 study, 18 university students learned to find the volume of four unusual solids. During practice, the blocked group scored 89% and the mixed group 60%. A week later, the mixed group scored 63% and the blocked group 20%. The researchers tie this to what Robert Bjork and his colleagues call "desirable difficulties," practice that costs you something now and pays off later.

The feeling fools most people. In the 2008 painting study, 78% of the students did better after mixed study, yet 78% said grouped study was as good or better, even after the test. In a 2021 study of 350 university physics students, homework that mixed topics led to better scores on two surprise tests, but the students still rated it harder and believed they learned less from it.

So test yourself instead. A week after your session, write a fresh mixed set with new numbers and do it without warming up on one kind first. If you get more right than you did near the end of the first session, it's working.

If the check shows one kind is still shaky, practice it on its own for a bit before you mix it back in. For percents, Elevate, a brain training app, can help. The Percentages game has you work out markups and discounts in your head, and the Discounting game is built around the math you do while shopping. They're two of Elevate's 12 math games, and its 40+ games also cover reading, writing, speaking, and memory, with difficulty that adjusts as you get better. The free version gives you one three-game workout a day, with the games picked for you. Elevate Premium opens every game, math and memory games included, so you can go straight to Percentages when one kind of percent problem keeps tripping you up.

To brush up on one kind of problem before your next mixed set, see Elevate's math and memory games.

Frequently asked questions

What is another word for interleaving?

Mixed practice is the plainest other word for interleaving. Sport and motor-skill researchers call it random practice and its effect contextual interference, and the 2008 painting study called it spaced presentation. The opposite is blocked practice, which some researchers call massed practice.

How long should you spend on each topic before you switch?

For a set of problems, switch with every problem, as in the session above. For bigger topics, the Learning Center at Washington University in St. Louis suggests 30 to 90 minutes on one subject for students and warns that switching too often can leave you feeling scattered.