Interleaving: why mixed practice beats blocked practice

Evidence grade: STRONG for discrimination-heavy material, PROMISING more broadly. Meta-analytic support is solid (Brunmair & Richter, 2019), with the largest, most consistent effects where the learner must choose which method or category applies — mathematics problem types, visual categories. Effects are smaller or absent where tasks aren't confusable, which is itself informative about the mechanism.

Standard practice structure is blocked: ten problems of type A, then ten of type B. Interleaved practice shuffles them: A, C, B, A, B, C. Blocked practice feels better, runs smoother, and produces higher scores during the session. On the delayed test, the ordering routinely flips — often dramatically.

The evidence, briefly

Rohrer and Taylor (2007) had students practice volume formulas for different solids either blocked or shuffled. Blocked practicers dominated during practice; a week later, the shuffled group scored far higher — one of the cleanest demonstrations of the practice/retention reversal in the literature.
Rohrer, Dedrick and Stershic (2015) extended it to real seventh-grade classrooms over a month: interleaved mathematics practice produced substantially better delayed-test performance than the same problems blocked.
Kornell and Bjork (2008) found the same for inductive learning — students shown painters' works interleaved learned to identify the artists' styles better than students shown them blocked, while overwhelmingly believing blocking had worked better. The metacognitive mismatch is part of the finding.
Brunmair and Richter's meta-analysis (2019, Psychological Bulletin) across dozens of studies: a reliable overall benefit, moderated exactly as the mechanism predicts — strong for confusable categories and problem types, weak where discrimination isn't the challenge.

Why it works

The leading account is discriminative contrast: interleaving forces the learner to notice what distinguishes problem types, because every item poses the hidden question blocked practice removes — which kind is this? In blocked practice, the strategy choice is made once at the top of the page; students execute methods without ever selecting them. Then the exam arrives — mixed by construction — and asks for selection they never practiced. Interleaving also automatically adds spacing between same-type items, and the two mechanisms compound.

What the evidence doesn't say

It doesn't abolish blocking. Early acquisition of a brand-new skill benefits from a short blocked run — you can't discriminate between methods you don't yet have. The supported pattern is block briefly, then interleave.
It doesn't promise gains everywhere. Interleaving unconfusable content (spelling words with history dates) adds shuffle without discrimination benefit — the meta-analysis shows the effect tracking confusability.
It doesn't feel good, and won't. Expect pupils (and your own instincts) to rate interleaved sessions as worse; that rating is the performance–learning dissociation on schedule.

In the classroom

Rebuild worksheets, not lessons: teach topic-by-topic as usual, but make practice sets cumulative and shuffled — today's type mixed with the last three weeks'.
Make "which method?" an explicit question: before solving, have pupils name the problem type. That's the discrimination step, surfaced.
Interleave within skills that get confused: reply vs reply-all, when-to-SUM vs when-to-COUNT, cp vs mv — confusability is the criterion, not subject boundaries.
Warn about the feeling, once, in child language: "mixed practice feels wobblier and works better — that's the wobble doing its job."

How Wiz Kids applies this

The daily review queue is interleaved by construction — due skills arrive from every realm in one session, so a child retrieves a spreadsheet formula, then a keyboard chord, then a safety judgment, each time re-selecting which knowledge applies. Capstone projects interleave at the task level (one job spanning five applications), and confusable pairs (reply/reply-all, copy/move) are deliberately reviewed near each other.

References

Rohrer, D., & Taylor, K. (2007). The shuffling of mathematics problems improves learning. Instructional Science, 35.
Rohrer, D., Dedrick, R. F., & Stershic, S. (2015). Interleaved practice improves mathematics learning. Journal of Educational Psychology, 107(3).
Kornell, N., & Bjork, R. A. (2008). Learning concepts and categories: Is spacing the "enemy of induction"? Psychological Science, 19(6).
Brunmair, M., & Richter, T. (2019). Similarity matters: A meta-analysis of interleaved learning and its moderators. Psychological Bulletin, 145(11).
Taylor, K., & Rohrer, D. (2010). The effects of interleaved practice. Applied Cognitive Psychology, 24(6).

© Glu IO Pty. Ltd. — Wiz Kids (wiz.kids). Link freely; republication requires permission — see terms. Found an error in our reading of the research? We correct fast: tell any teacher piloting Wiz Kids.