What is interleaving and why does it feel worse but work better?
Interleaving is mixing different topics or problem types within a study session, rather than practising one thing until you move on. Instead of doing twenty problems of type A, then twenty of type B, you mix them: A, C, B, A, B, C.
The counterintuitive finding is that interleaving reliably produces worse performance during practice and better performance on later tests. Blocked practice feels smooth and productive; interleaved practice feels slow and error-prone. Students who interleave consistently rate the experience as less effective — while outperforming blocked learners, sometimes substantially, on delayed assessments.
Why it works:
It forces discrimination. In blocked practice you already know which method applies, because you just did nineteen of them. The hard part of a real exam is not executing a method but recognising which method a problem calls for. Interleaving practises that recognition every single time.
It creates desirable difficulty. Retrieving a strategy you have not used for a few minutes is harder than repeating one you just used, and that extra retrieval effort strengthens memory. This is the same principle underlying spaced repetition, and the two combine well.
It prevents fluency illusions. Blocked practice produces a strong feeling of mastery that does not survive a mixed test.
Where the evidence is strongest: mathematics and problem-solving with distinct but confusable categories, and perceptual learning such as recognising artists' styles or classifying specimens. Evidence is more mixed for some other domains.
Practical use: mix problem types within a maths session rather than working through one exercise set. Mix topics within a subject. Use past papers, which are interleaved by design. And judge your progress by delayed testing, not by how the session felt — the feeling is systematically misleading.
One caveat: learn a method properly first. Interleaving is for consolidating, not for initial acquisition.