Interleaving: What the Research Shows and How to Use It

- Interleaving means mixing different problem types or categories in one practice session (A, C, B, A, B, C) instead of finishing all of one type before the next (A, A, A, B, B, B).
- In a seventh-grade classroom study (Rohrer, Dedrick and Stershic, 2015), students who did mixed practice scored 74% on a test 30 days later, against 42% for students who did blocked practice.
- Blocked practice feels better while you are doing it. In Rohrer and Taylor (2007), blocked practice scored 89% during practice vs. 60% for mixed, then 20% vs. 63% on the test a week later.
- In Kornell and Bjork (2008), 78% of students learned painting styles better when artists were mixed, yet 78% judged blocking to be as good or better.
- The evidence is narrower than for spacing or retrieval practice. Dunlosky et al. (2013) rated interleaving moderate utility, and most studies used math problems or visual categories.
Open almost any math textbook and you will find blocked practice: a lesson on one kind of problem, followed by twenty problems of exactly that kind. You know which method to use before you read the question, because it is the method from the lesson you just finished. The practice goes smoothly and you get most of them right.
The exam does not work that way. Problems from different chapters appear side by side, and half the difficulty is figuring out which method a problem needs. Interleaving is the practice format that makes you do that during study instead of for the first time on the test.
Quick answer: Interleaving is mixing different types of problems or material within one study session instead of practicing one type at a time. In the studies below, mixed practice usually looked worse during practice and better on a delayed test. The evidence is strongest for math problems and learning to tell categories apart, and it is less tested elsewhere.
What Interleaving Is
Say you are studying three kinds of problems: A, B and C. Blocked practice means doing all the A problems, then all the B problems, then all the C problems: A A A A, B B B B, C C C C. Interleaved practice means mixing them so consecutive problems usually differ: A C B B A C B A C. You do the same problems either way. Only the order changes.
Interleaving can also apply to categories rather than problems: mixing examples of different painters, bird species, diseases or types of argument, instead of studying all the examples of one before moving on.
One detail matters for reading the research. Mixing problem types also spreads out practice of each type, because two A problems now have other problems between them. So in many studies interleaving and spacing happen together, and it is not always possible to say which one caused the result. One of the four studies below held spacing constant to separate them.
The Research: Four Key Studies
| Study | Who and what | Practice result | Test result |
|---|---|---|---|
| Rohrer & Taylor (2007), Exp. 2 | 18 participants completed; volume formulas for four solids | Blocked ahead: 89% vs. 60% | Mixed ahead a week later: 63% vs. 20% |
| Kornell & Bjork (2008) | Lab study; painting styles of 12 artists | Not the main measure | Mixed ahead on new paintings: 0.61 vs. 0.35 (Exp. 1a), 0.59 vs. 0.36 (Exp. 1b) |
| Taylor & Rohrer (2010) | Children; four kinds of math problems; spacing held constant | Blocked ahead | Mixed scores doubled a day later |
| Rohrer, Dedrick & Stershic (2015) | 126 seventh graders in 9 classes; two kinds of algebra problems | Not the main measure | Mixed ahead: 80% vs. 64% at 1 day, 74% vs. 42% at 30 days |
Rohrer and Taylor (2007): shuffled math problems
This paper in Instructional Science ran two experiments. Its first experiment, with 66 undergraduates, is often cited as evidence for interleaving, but it was really about spacing: students who split their permutation practice across two sessions scored 74% on a test a week later, against 49% and 46% for students who did all their practice in one session.
The second experiment is the interleaving one. Students learned formulas for the volume of four solids, practiced in two sessions a week apart, and took a test a week after that. The blocked group did each kind of problem in its own set; the mixed group got all four kinds shuffled together. During practice, the blocked group was far ahead, 89% vs. 60%. On the test, the mixed group was far ahead, 63% vs. 20%. Only 18 students completed this experiment, and the mixed schedule also spaced out practice of each solid, so it is a striking result rather than a definitive one.
Kornell and Bjork (2008): learning painters' styles
This study in Psychological Science moved interleaving from math to categories. Participants studied six paintings by each of 12 landscape painters, either massed by artist (all six by one painter, then the next) or mixed. Then, after a 15-second distractor task, they tried to name the artist of paintings they had never seen. On the first block of the test, mixed study did better: 0.61 vs. 0.35 correct in Experiment 1a (120 participants) and 0.59 vs. 0.36 in Experiment 1b (72 participants).
The caveats: the test came almost immediately, it was a lab task, and spacing and interleaving were confounded here too. The most quoted part of the paper is about what students believed, covered below.
Taylor and Rohrer (2010): children and math problems
Published in Applied Cognitive Psychology, this study gave children four kinds of math problems. Crucially, it held spacing constant, so the comparison was mixing vs. blocking and not mixing vs. cramming. Interleaving lowered performance during practice but doubled scores on a test one day later. An analysis of students' errors suggested the benefit came from picking the right procedure for each problem.
Rohrer, Dedrick and Stershic (2015): a real classroom
This study in the Journal of Educational Psychology took interleaving into ordinary classes. 126 seventh graders in 9 classes, taught by 3 teachers at one Florida school, did about three months of practice on two kinds of problems: graphing a linear equation and finding a slope. Classes were assigned to mixed or blocked practice. On an unannounced test, the mixed group scored 80% vs. 64% one day later (d = 0.42) and 74% vs. 42% thirty days later (d = 0.79).
The effects were smaller than in the lab studies, it involved only two problem types at one school, and the authors themselves wrote that whether the results generalize to other material "remains unknown." Of the four studies here, it is the one run in ordinary classrooms, over about three months.
Why Blocked Practice Feels Better
The most useful thing these studies show may be how misleading practice performance is. In Rohrer and Taylor's second experiment, the blocked group scored 89% during practice and the mixed group 60%. Anyone judging by practice scores would pick blocking. A week later the order flipped, 20% vs. 63%.
Students' own judgments are no better. In Kornell and Bjork's painting study, 78% of participants did better with mixed study, yet 78% said massed study was as good or better. People tend to trust the ease of blocked practice, and that ease comes partly from knowing which method or category is coming next.
What this means for you
If mixed practice feels slower and you get more wrong, that matches what these studies found and is not a sign it is failing. Judge your study method by how you do on a delayed test, such as a practice exam a week later, not by how smooth the session felt.
Why Interleaving Might Work
A leading explanation is about choosing a method. In blocked practice, the lesson tells you which procedure to use, so you only practice carrying it out. In mixed practice, each problem makes you decide which procedure applies before you carry it out. Exams require both steps. Taylor and Rohrer's error analysis fits this account: interleaving appeared to help students pick the right procedure.
For categories, the account is similar. Seeing different painters side by side may make the differences between their styles easier to notice than seeing six paintings by one artist in a row. Kornell and Bjork's design could not fully separate this from the effect of spacing, so treat it as a plausible explanation rather than an established one.
Where the Evidence Is Weaker
Interleaving has less evidence behind it than retrieval practice or spacing, and it is worth being honest about that.
- Dunlosky et al. (2013) rated it moderate utility. Their review rated practice testing and distributed practice high utility and put interleaved practice in the middle, partly because its benefits "have just begun to be systematically explored." That rating is about how broadly the evidence generalizes, not about effect size.
- Most of the evidence is math and category learning. The four studies here used geometry, algebra and painting styles. Whether mixing helps with, say, history essays or reading comprehension is much less studied.
- Spacing is often mixed in. In several studies, interleaving also spaced out practice, so part of the benefit may be spacing.
- Samples are often small. Rohrer and Taylor's interleaving experiment had 18 students who completed it.
- Classroom effects were smaller. The classroom study found real but smaller effects than the lab studies, with only two problem types.
None of this means interleaving does not work. It means the strongest case is for subjects where the hard part is telling problem types or categories apart.
How to Use Interleaving When You Study
Build mixed problem sets
When you practice for a math, physics, chemistry, economics or statistics exam, stop working through end-of-chapter sets one chapter at a time once you have learned each method. Pull a few problems from each chapter the exam covers and shuffle them. Cover the section headings if your book prints them above the problems, because the heading tells you the method. This is the closest match to the studies above.
Shuffle flashcards across units
If you keep one deck per chapter, you always know which chapter a card comes from, which is a hint the exam will not give you. Once you have learned each deck, review cards from several units in one session. This matters most for cards where the challenge is telling similar things apart: similar drugs, similar historical periods, confusable terms.
Mix old and new topics
Instead of spending each session only on this week's material, add a few problems or cards from earlier units. That mixes problem types and also brings older material back after a gap, so you get interleaving and spacing in one move.
Learn each type before you mix
Mixed practice is for telling methods apart, which means you need to know each method first. If you have never solved a slope problem, a mixed set will just be confusing. A sensible order is to learn a new method with a few worked examples, then fold it into mixed practice.
StudyEdge AI
Practice the way the exam asks.
Upload your course material and StudyEdge AI builds a practice exam of 10, 20 or 30 questions across it. You also get flashcards that come back on a schedule, and quizzes on your course topics.
Try StudyEdge AI FreeFree plan available. Pro comes with a 7-day free trial · Cancel anytime.
How It Fits With Spacing and Retrieval Practice
Interleaving works best as one part of a system, not a standalone trick. The three techniques answer different questions:
- Retrieval practice is about how you review: recall the answer instead of re-reading it. Roediger and Karpicke (2006) found that students who took recall tests remembered 61% of a passage a week later, against 40% for students who only re-read.
- Spacing is about when you review: spread sessions out, with longer gaps for longer-term goals. Cepeda et al. (2006) reviewed 317 experiments and found spaced practice beat massed practice.
- Interleaving is about what order you review in: mix types so each problem makes you choose a method.
They combine naturally. A mixed problem set done from memory, a few days after the last one, is retrieval practice, spacing and interleaving at once. For the wider picture, see our ranking of the best study techniques by science and the overview of study techniques.
Sources
- Rohrer, D., & Taylor, K. (2007). The shuffling of mathematics problems improves learning. Instructional Science, 35(6), 481-498. doi:10.1007/s11251-007-9015-8
- Kornell, N., & Bjork, R. A. (2008). Learning concepts and categories: Is spacing the "enemy of induction"? Psychological Science, 19(6), 585-592. doi:10.1111/j.1467-9280.2008.02127.x
- Taylor, K., & Rohrer, D. (2010). The effects of interleaved practice. Applied Cognitive Psychology, 24(6), 837-848. doi:10.1002/acp.1598
- Rohrer, D., Dedrick, R. F., & Stershic, S. (2015). Interleaved practice improves mathematics learning. Journal of Educational Psychology, 107(3), 900-908. doi:10.1037/edu0000001
- Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving students' learning with effective learning techniques. Psychological Science in the Public Interest, 14(1), 4-58. doi:10.1177/1529100612453266
- Roediger, H. L., III, & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science, 17(3), 249-255. doi:10.1111/j.1467-9280.2006.01693.x
- Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., & Rohrer, D. (2006). Distributed practice in verbal recall tasks: A review and quantitative synthesis. Psychological Bulletin, 132(3), 354-380. doi:10.1037/0033-2909.132.3.354
Frequently Asked Questions
What is interleaving in studying?
Interleaving is mixing different types of problems or material within one study session instead of practicing one type at a time. For example, instead of doing ten area problems and then ten volume problems, you shuffle them together. The problems are the same; only the order changes. The opposite is blocked practice.
Is interleaving better than blocked practice?
In the main studies, yes, on delayed tests. Rohrer, Dedrick and Stershic (2015) found seventh graders who did mixed practice scored 74% on a test 30 days later, against 42% for blocked practice. Taylor and Rohrer (2010) found mixing doubled children's math test scores a day later. Blocked practice usually looks better during the practice itself, so do not judge by how the session feels.
Why does interleaving feel harder?
Because each problem makes you decide which method to use, and you get more wrong while practicing. In Rohrer and Taylor (2007), blocked practice scored 89% during practice and mixed practice 60%, but on the test a week later mixed practice scored 63% and blocked 20%. In Kornell and Bjork (2008), most students learned better with mixing but believed blocking worked as well or better.
Does interleaving work for every subject?
The evidence is mostly from math problems and learning to tell categories apart, such as painting styles. Dunlosky et al. (2013) rated interleaving moderate utility because its benefits had only begun to be explored systematically. It is most likely to help where the hard part of the exam is recognizing which type of problem or category you are looking at. For memorizing facts, retrieval practice and spacing have stronger evidence.
What is the difference between interleaving and spaced repetition?
Spaced repetition is about when you review: spreading sessions out over days and weeks. Interleaving is about the order within a session: mixing problem types instead of grouping them. They often overlap, because mixing types also puts gaps between problems of the same type, which is why many studies cannot fully separate them. Taylor and Rohrer (2010) held spacing constant and still found a large benefit from mixing.