Interleaving: How to Mix Practice for Lasting Learning

Interleaving means mixing related examples or problem types so learners must identify what they are looking at and choose an appropriate method. A blocked worksheet can present ten questions that all use the same procedure. An interleaved worksheet mixes several familiar types, so the question itself must provide the cue.

This is not a reason to shuffle every lesson or switch subjects every few minutes. Interleaving works best when the items are worth comparing and learners have enough prior knowledge to notice the important differences. The strongest applied classroom evidence concerns mathematics problems where learners must select a solution strategy. A systematic review found useful effects but also important limits (Firth et al., 2021).

‌

What is interleaving in education?

Interleaving is an arrangement of learning examples or practice questions in which an item from one category is followed by an item from another. Blocking groups examples of the same type together. The change in order matters because learners can no longer assume that the next question needs the same method as the last one.

For example, a blocked maths set can contain five area questions, five perimeter questions and five volume questions. An interleaved set mixes all three. Learners must first diagnose the problem, then calculate. In concept learning, examples from different categories can be alternated so that their diagnostic features are easier to compare.

Firth, Rivers and Boyle (2021) define interleaving as varying the order of examples so each is surrounded by examples from a different category or concept. Their review also warns that the value depends on what is being learned and how similar the examples are.

‌

Why can interleaved practice improve learning?

Interleaving can improve learning by making learners discriminate between similar ideas and retrieve the method needed for each item. Blocking supplies an accidental cue because the answer method often repeats. Interleaving removes that cue and makes the learner inspect the features of the current problem.

This is called discriminative contrast. When related examples sit near one another, attention can move to the differences that define each category. In mathematics, this can mean noticing whether a question asks for area, perimeter or volume before choosing a formula. Dunlosky and colleagues (2019) tested this account directly.

The learner must read the cue, bring a method to mind and decide whether it fits. Our Information Processing Theory guide explains how attention and stored knowledge support that choice.

Interleaving also distributes encounters with each item type, but this does not make it identical to spacing. Chen and colleagues (2021) argue that spacing and interleaving require distinct explanations. Spacing includes a period away from deliberate learning. Interleaving fills that interval with a different task.

‌

What does the evidence say?

The evidence supports a conditional claim, not a universal rule. Interleaving has produced useful gains in mathematical problem selection and concept learning. Results vary by material, learner knowledge and test. Much of the wider evidence comes from laboratory studies rather than ordinary classroom teaching.

In a mathematics experiment where spacing was held constant, interleaving reduced performance during practice but doubled scores on a test one day later (Taylor and Rohrer, 2010). Classroom studies have also found delayed gains from interleaved mathematics practice (Rohrer, Dedrick and Stershic, 2015; Rohrer et al., 2020).

A meta-analysis across varied tasks found a moderate average benefit, with important moderators (Brunmair and Richter, 2019). A later systematic review of concept learning found benefits for memory and transfer, especially where differences between items were subtle, but noted that the literature was dominated by university laboratory studies (Firth et al., 2021).

The Education Endowment Foundation's 2021 review of cognitive science in the classroom found moderate positive evidence for learners aged roughly 8 to 14 selecting solutions in mathematics. Eleven of the twelve classroom studies reviewed were in maths. It found little applied evidence for other subjects or age groups. Classroom examples elsewhere on this page are therefore design suggestions, not proven outcome claims.

‌

Blocked practice and interleaved practice

Blocked practice keeps one item type together, which supports initial fluency and reduces the need to choose a method. Interleaved practice mixes established item types, which makes method selection part of the task. Neither format is always best. Their purposes differ.

Blocked is not the same as massed. Blocking describes the order of item types. Massing describes repeated practice with little time between sessions. A task can be blocked but spaced across several days.

‌

When is blocked practice the better choice?

Blocking is often the better choice when learners are meeting a method for the first time, when each step still needs close guidance, or when the examples within one category are so varied that learners have not yet formed a stable category. Interleaving cannot retrieve a method that has not been taught.

For a new algebra procedure, a teacher can model two examples, guide three similar examples and let learners complete a short blocked set. Once the method is secure, it can be mixed with another familiar procedure. The transition depends on evidence from learner work, not a fixed number of questions. The Deliberate Practice guide shows how feedback and focussed repetition can strengthen a weak step before it joins a mixed set.

Blocking may also help learners notice what members of one category have in common. Firth and colleagues report that interleaving is not reliably helpful when items are unrelated or when attention needs to remain on within-category similarities. The useful question is, “What distinction should this sequence teach?”

‌

Interleaving and spaced practice

Interleaving changes which kind of item comes next. Spaced practice changes when the same learning returns. A sequence can use both, but the two decisions should be planned separately. Mixing three problem types today is interleaving. Returning to those problems next week is spacing.

A retrieval starter can include ratios, fractions and percentages from earlier lessons. The gap since first teaching supplies spacing. The mixed order supplies interleaving. Our Spaced Practice guide explains how to plan the time gaps, while the Retrieval Practice guide covers bringing knowledge to mind.

Do not assume that more difficulty is always more desirable. If the time gap has made the underlying method unavailable, add a cue or worked example before asking learners to choose among methods.

‌

How to design mixed practice sets

A useful mixed set starts with a specific choice. Pick two or three related item types that learners can already complete but sometimes confuse. Vary the order, remove labels that reveal the method and ask learners to explain the cue they used. Where a model needs a diagram, the Dual Coding guide helps teachers keep the visual tied to the key relationship.

1. Name the confusion. Record the methods, categories or representations that learners mix up.
2. Check each prerequisite. Use a short blocked example to confirm that the learner can execute each method separately.
3. Choose diagnostic variation. Vary the features that should determine the method. Avoid irrelevant surface changes that only add noise.
4. Mix without a visible pattern. Do not alternate so neatly that position becomes the new cue.
5. Ask for the choice before the answer. Learners can label the problem type, name the method or underline the deciding information.
6. Check after a delay. Reuse new examples later and compare method selection, accuracy and explanation.

Use the Working Memory guide when a mixed task has too many simultaneous steps. The aim is productive selection, not avoidable overload.

‌

Classroom examples across subjects

Mathematics has the clearest classroom evidence, but the design principle can inform other subjects where learners must discriminate between related cases. Outside the strongest evidence base, treat these examples as hypotheses to test through delayed work rather than guaranteed effects.

• Mathematics: mix area, perimeter and volume questions after each procedure has been taught. Ask learners to name the deciding feature before calculating.
• Science: mix examples of physical and chemical change. Ask learners to identify the evidence that justifies the category.
• English: mix short examples of explanation, analysis and evaluation. Ask which function each sentence performs and why.
• History: mix sources that require provenance, inference or contextual knowledge. Ask learners to select the first analytical move.
• Languages: mix familiar tense forms only when the learner can produce each one separately. Require a reason for the tense choice.
• Physical education: alternate established responses where the learner must read a cue and select an action. Do not replace the blocked acquisition of a new movement.

‌

Interleaving activities teachers can use

Interleaving activities should reveal whether learners can tell related demands apart. The format matters less than the choice it requires. Begin with a small mix, ask learners to name the clue, check their work after a delay and adjust the next set.

• Mixed problem sets with two or three familiar methods.
• Error sorting, where learners match worked errors to the method that caused them.
• Example classification, with a reason for each category choice.
• Comparison pairs that place easily confused examples side by side.
• Cumulative exit tickets with one current and two earlier item types.
• Worked-example matching, where learners select the model that fits a new problem.
• Method-first questions, where the learner chooses a method before calculating.
• Shuffled flashcards that ask for a category as well as an answer.
• Source-analysis cards that mix different evidential demands.
• Grammar corrections that require the learner to identify which rule applies.
• Retrieval grids whose columns do not disclose the topic.
• Revision sets that revisit related ideas after different time gaps.
• Misconception pairs that ask why one example fits and the other does not.
• Exam-question sorting before any answers are written.
• Learner-generated mixed sets, checked by a partner for genuine contrast.

‌

How to create an interleaved study schedule

An interleaved schedule should identify related material, confirm prior learning and plan both order and delay. Avoid generic instructions to switch subject every twenty minutes. The useful unit is an example or problem whose category matters, not an arbitrary slice of time.

Start with a weekly review slot. Select one current item type and two previously taught types that learners confuse. Shuffle six to nine questions. Next week, use new examples of the same distinctions. Track whether learners choose the method correctly before recording calculation errors. Where a shared scheme fixes the sequence, agree one common mixed set at department level.

Independent learners can use the same process. List the problem types required by an assessment, practise each unfamiliar method briefly, then create a mixed set without headings. Add a delayed set to the calendar. The Desirable Difficulties guide explains why immediate ease is a poor measure of lasting learning.

‌

Common interleaving mistakes

The most common mistakes are mixing before teaching, combining unrelated topics, adding too many categories, using a predictable alternation and judging success during practice. Each mistake changes productive comparison into guessing, overload or a new superficial cue.

• Random mixing. “Variety” is not a rationale. State the distinction learners need to make.
• Premature mixing. If a learner cannot execute each method separately, return to modelling and guided practice.
• Too many categories. Begin with the smallest set that exposes the target confusion.
• Obvious sequences. An A-B-A-B pattern lets position disclose the answer method.
• Confusing effort with overload. Errors can reveal learning needs, but persistent random responding is not a desirable difficulty.
• Using immediate accuracy as the verdict. Include a delayed and slightly changed check.
• Claiming every subject is proven. Distinguish classroom evidence from a plausible application of the principle.

‌

Plan an interleaved practice set

Use this mini app to define the distinction, check readiness and plan a delayed measure. It produces a printable design brief. It does not generate questions or send learner data anywhere.

Mini app · mixed practice · lesson planning

‌

Plan one useful mix

Name the distinction first. Then decide what evidence will show that learners can choose.

Related item types Target confusion ReadinessChoose oneEach method still needs modellingEach method works with guidanceEach method works independently Mixed sequence Choice prompt Delayed check Build the plan Clear

Your entries stay in this browser and are not saved or sent.

‌

Limitations and evidence boundaries

Interleaving does not guarantee stronger learning. Its effect depends on the similarity of categories, variation within each category, prior knowledge, timing and the outcome measured. It may help when contrasts reveal diagnostic features. It may hinder when learners first need to notice similarities within one category.

The school evidence is narrower than many online explanations imply. The Education Endowment Foundation review found that nearly all classroom studies concerned mathematics and highly scripted tasks. Firth and colleagues found encouraging concept-learning results, but most studies used university learners in controlled settings. Claims about every subject, age group and classroom format go beyond that evidence.

Judge an interleaved task against a delayed sample of new problems. Record method selection separately from execution. If learners choose at random, reduce the mix or restore a model. If they choose correctly but calculate poorly, the method needs more practice. This diagnosis links interleaving with Cognitive Load Theory rather than treating difficulty as automatically productive.

‌

Frequently asked questions

These answers distinguish interleaving from task switching, spacing and random variety. They also set a boundary around the evidence and explain how teachers can decide when to move from blocked to mixed practice.

‌

When should teachers start interleaving?

Start when learners can complete each target method with reasonable independence but still need practice choosing between them. Use blocked modelling and guided examples first when a method is new.

‌

How many topics should teachers mix?

There is no universal number. Begin with the smallest set that exposes the intended confusion, often two or three item types, then add complexity only when learner work justifies it.

‌

Is interleaving the same as switching subjects?

No. Switching from chemistry to history may create variety, but it does not necessarily create useful comparison. Interleaving needs a clear relationship between the items being mixed.

‌

Why does interleaving feel harder?

Blocked practice often reveals the method through repetition. Interleaving removes that cue, so learners must retrieve and select a method. The extra demand can lower immediate accuracy.

‌

How should teachers measure its effect?

Use new examples after a delay. Record whether learners identify the item type, choose an appropriate method, execute it accurately and explain the deciding feature.

‌

Further reading and references

These sources support the definition, the account of how interleaving works, the mathematics findings and the limits used in this article. Each DOI link points to the publisher record, so readers can check the study design and findings for themselves.

• Kornell, N., & Bjork, R. A. (2008). Learning concepts and categories: Is spacing the enemy of induction?
• Taylor, K., & Rohrer, D. (2010). The effects of interleaved practice.
• Rohrer, D., Dedrick, R. F., & Stershic, S. (2015). Interleaved practice improves mathematics learning.
• Brunmair, M., & Richter, T. (2019). Similarity matters: A meta-analysis of interleaved learning and its moderators.
• Foster, N. L., Mueller, M. L., Was, C., Rawson, K. A., & Dunlosky, J. (2019). Why does interleaving improve math learning?
• Rohrer, D., Dedrick, R. F., Hartwig, M. K., & Cheung, C. N. (2020). A randomized controlled trial of interleaved mathematics practice.
• Firth, J., Rivers, I., & Boyle, J. (2021). A systematic review of interleaving as a concept learning strategy.
• Chen, O., Paas, F., & Sweller, J. (2021). Spacing and interleaving effects require distinct theoretical bases.