Do ten division problems in a row and by the eighth one your hand knows the move before your brain does. That fluency is a trap. Shuffle those same problems in with multiplication and subtraction, so you never know which operation is coming, and every problem forces you to first decide what kind of problem it is. That decision is most of the learning, and blocked practice quietly skips it.
That's interleaving: instead of studying one topic to exhaustion before moving to the next (blocking), you mix related topics so each one keeps interrupting the others. It feels worse. You make more mistakes, progress feels slower, and you leave a session less confident. It also works better. The gap between how it feels and how it performs is one of the widest in learning research.
Why blocking fools you
When you drill one topic in a block, your working memory holds the method steady across every problem. Performance during the session climbs, so it feels like mastery. But you've practised executing a method, not choosing it, and on a real test, or in real life, nothing tells you which method to reach for. That's the skill blocking never trains.
Interleaving forces the harder, more useful thing: retrieving the right approach from scratch, every time. It pairs naturally with spaced repetition, and for the same reason: a little difficulty now, applied against the forgetting curve, buys durable memory later. The strain is the mechanism, not a side effect.
Blocking vs interleaving
| Blocking | Interleaving | |
|---|---|---|
| Session feel | Smooth, confident | Effortful, error-prone |
| What it trains | Executing a known method | Choosing the right method |
| Short-term performance | Higher | Lower |
| Test / real-world transfer | Lower | Higher |
| Best for | Brand-new motor skills, first exposure | Discriminating between similar things |
Where it helps most, and where it doesn't
Interleaving shines when topics are confusable: math problem types, similar grammar rules, artists whose styles blur together, muscles with overlapping actions. The whole benefit is learning to tell near-neighbours apart, so mix things that are close enough to be mixed up.
Be honest about the limits. When something is genuinely new (a first piano scale, an unfamiliar formula), you need a little blocked practice to build the basic motion before shuffling helps. And don't interleave unrelated subjects hoping for magic; jumping from Spanish to organic chemistry to tax law is just task-switching, which fragments focus and helps nothing.
Interleaving your notes
Most of what you keep isn't flashcard material, but the principle still pays off. A knowledge base is only useful when ideas from different corners collide, which is exactly what pure blocking (one notebook, one topic, never revisited together) prevents. Two habits bring interleaving to your notes:
- Keep ideas small and separable. Atomic notes, one idea each, are the units you can reshuffle later. A wall of blended text can't be interleaved.
- Let old notes resurface next to new ones. A weekly skim, on-this-day review, or surfacing a related older note beside today's is interleaving without a deck, the same move that powers a working second brain.
This is where a capture tool earns its keep. ClairMind's AI works across your own notes, pulling an older, related note up beside a new one so ideas keep bumping into each other instead of sitting in tidy, forgotten blocks. It won't build you a flashcard deck. For deliberate exam drilling, an app like Anki that interleaves cards by default is the sharper tool, and honesty matters more than the pitch. But for the slower interleaving of your own thinking, resurfacing beats filing.
For the wider science this sits inside (retrieval, spacing, attention), start with memory & focus.