Interleaving
Mix topics in one session — don’t block them.
What the Research Says
Most students revise by blocking — an hour on networks, an hour on algorithms, an hour on data representation. It feels productive and the topic seems clear by the end. Interleaving is the opposite: you jump between topics inside a single session. It feels worse at the time, and you’ll make more mistakes. But when the real exam arrives, you will be dramatically better at knowing which technique to apply — and that is usually the hardest part of any exam question.
Evidence Base
| Study | Finding | Impact |
|---|---|---|
| Rohrer & Taylor (2007) | Students who interleaved maths topics scored 43% higher on delayed tests than blocked-practice students. | Same time spent — huge difference in exam performance. |
| Birnbaum et al. (2013) | Interleaved learners felt LESS confident during practice but performed BETTER in the test. | If revision feels too easy, interleave. |
| Bjork & Bjork (2011) | Desirable difficulties — including interleaving — create durable long-term learning at the cost of short-term performance. | Short-term struggle is the price of long-term gain. |
| Taylor & Rohrer (2010) | Benefit was largest when topics looked superficially similar — students learnt to DISTINGUISH them. | Most useful where students confuse similar techniques. |
Step-by-Step Method
- 1Pick 3–5 related topics — topics that share a theme or question type. Similar enough that students confuse them.
- 2Make a mixed problem set — 10–15 questions drawn randomly from all the topics. Don’t tell yourself which topic each question belongs to.
- 3Attempt each question cold — first step is identifying what’s being asked. That identification is the skill you’re training.
- 4Mark after every question — note which topic each one was. Pay attention to mis-identifications — those are the highest-value gaps.
- 5Rebuild a new mixed set tomorrow — same topics, different questions. Different order. Over time, misclassifications drop.
- 6Expect to feel worse than blocked practice — that’s the right feeling. Your exam self will thank you.
Write ten short-answer questions that mix: tracing pseudo-code with a FOR…ENDFOR loop, tracing one with REPEAT…UNTIL, and tracing one with IF…THEN…ELSE…ENDIF — all without labelling which construct each one uses. Attempt them cold and identify which loop type applies before writing any output. Students who block-practise loops often freeze in their exam when the loop type is unfamiliar; interleaving trains the vital “what kind of structure is this?” diagnostic skill directly.
Mix five topics that students frequently confuse: Boolean AND, OR, NOT, and XOR operators (including three-input truth tables); malware types (virus, trojan, spyware — noting worms are not on the specification); UK legislation (Data Protection Act 2018, Computer Misuse Act 1990, Copyright Designs and Patents Act 1988); SQL query syntax (SELECT, FROM, WHERE, ORDER BY); and network protocols (TCP/IP layer responsibilities). Create a 15-question mixed set from these areas and rotate daily — the confusion between, say, a trojan and a virus, or between the Computer Misuse Act and CDPA, disappears quickly through interleaving.
Common Pitfalls
- Random, not related. Interleaving Computer Science with History is just subject-switching. Interleave WITHIN a subject — related topics that students confuse.
- Giving up when it feels hard. The discomfort is the signal that it’s working. Don’t revert to blocked practice because interleaving feels slow.
- Labelling the topic at the top of each question. Defeats the point — the identification IS the training.
- Using it too early in learning. You need baseline familiarity with each topic first. Interleaving is for consolidation, not first exposure.
When to Use This
Once you’ve covered a handful of related topics — usually partway through a term — start weaving them into mixed practice sessions. Very useful in the final 8–10 weeks before your exam when you’re moving from learning to applying. Full practice papers are the ultimate interleaving exercise, because they force you to identify question types cold.