Mix topics during retrieval practice rather than blocking by subject

When reviewing material from multiple topics, alternate between them rather than finishing one before moving to the next.

Why it works

Interleaving retrieval practice across topics forces the learner to identify which strategy or framework applies to each problem — a discrimination challenge that blocked practice avoids. This discrimination training builds a more flexible schema: you know not just how to solve a type of problem, but how to recognize which type of problem you’re facing. Short-term, interleaving feels harder and produces more errors; long-term, it produces substantially better transfer to novel problems.

How to do it

1. When reviewing two or more topics, create a mixed question set rather than a topic-by-topic set.
2. For each question, before answering, identify which topic or framework it belongs to — that identification is part of the learning.
3. Accept that performance will be lower in interleaved practice than blocked; that difficulty is the mechanism, not a failure.
4. Return to blocked practice only for material you haven’t yet grasped at all.

Evidence

Rohrer and Taylor (2007) and subsequent studies showed that interleaved mathematics practice produced better performance on delayed tests than blocked practice, even when blocked practice produced better scores on immediate tests. The "interleaving effect" has been replicated in motor learning and classification tasks. (rct)

Interleaving is most effective when the learner has some competence with each component skill; for complete novices on a topic, some blocked introduction may be needed before interleaving.

Sources

• Rohrer & Taylor (2007), The shuffling of mathematics problems improves learning, Instructional Science

Common mistake

Interpreting the lower performance during interleaved practice as evidence that it isn’t working, and switching back to blocked practice — which prevents the discrimination learning that interleaving builds.