Interleaving vs Blocked Practice: Which Study Method Actually Works Better

You finish a chapter on quadratic equations. You feel good. You've been drilling the same problem type for forty minutes and your accuracy is climbing. You move to the next chapter.

But here's what's actually happening: you just got better at quadratic equations today. You didn't get better at math. Those are different things, and the research has been clear about this for years.

Blocked practice, the habit of studying one topic or skill type until you feel competent, is intuitive. It feels productive. Your confidence climbs. The problem is that confidence and competence don't always track together, and the difference shows up on tests, not in your study sessions.

What the Research Actually Shows

In one study, math students practiced one problem type until they felt ready, then moved to the next. Other students mixed problem types together, switching between them as they practiced. On an immediate test, the blocked group performed better. No contest.

But the real test came a week later, when students encountered those problem types mixed together in a way they hadn't practiced. The interleavers scored 43% higher. The blocked practice group had built performance that didn't transfer.

This is from Rohrer and Taylor (2007), published in Cognitive Psychology. They've run variations of this study multiple times. The pattern holds: interleaving produces learning that sticks and generalizes, while blocked practice produces learning that looks good in the moment and falls apart when conditions change.

The mechanism matters here. When you interleave different problem types, your brain can't rely on a single strategy. It has to constantly retrieve the right approach, discriminate between similar-looking situations, and update its understanding of what distinguishes one problem type from another. That retrieval and discrimination is the learning. Repetition without discrimination is just exercise.

Kornell and Bjork (2008) found something related. People consistently rate interleaved practice as less effective than blocked practice, even when the opposite is true. They feel like they're learning less. The interleaved approach feels harder and messier. So they stop doing it. This is a deeply human response to a counterintuitive finding: we trust our feelings over the evidence, and our feelings are calibrated to immediate performance, not long-term retention.

The effect isn't limited to math. Birnbaum, Kornell, and Bjork (2013) found interleaving advantages across different perceptual and cognitive tasks. It seems to be a general property of how we learn to categorize and distinguish, which covers a lot of what you actually study.

Why It Works

Think about what blocked practice lets you avoid. You never have to ask yourself whether this is a statistics problem or a probability problem. You already know, because you've decided you're studying probability. The context does the work your brain should be doing.

Interleaving removes that context. You have to figure out what kind of problem you're looking at before you can solve it. That effort is uncomfortable, but it's also exactly where the learning happens. You're building the pattern-recognition system that will serve you on any novel mix of problems, which is what exams actually are.

There's another layer. When you study one topic for a long stretch, you build a mental set. You start anticipating the same operations, using the same strategies. That narrow activation feels like mastery but produces narrow learning. Interleaving forces you out of those mental sets. You have to stay alert, adjust on the fly, and that flexibility is part of what transfers.

Some researchers call interleaving a desirable difficulty. It feels harder. It feels slower. Your immediate performance will be lower. That's not a sign it's not working. That's the difficulty part.

How to Use This

You can start small. Pick one subject where you've been blocked-practicing and introduce one interleaved session per week. It doesn't need to replace everything you do. It just needs to show up.

Plan your study blocks around mixed problem sets rather than single-topic sets. If you're studying for a language exam, mix vocabulary, grammar, and reading comprehension within one session instead of spending thirty minutes on vocabulary, then moving to grammar. The switching is the work.

If your material naturally clusters into types, use that. In math, mix problem types within a set. In history, alternate between causation questions and comparison questions. In language learning, switch between grammar rules, vocabulary, and listening practice within one session. The key is that the types should be similar enough to require discrimination but different enough to create friction.

Expect it to feel weird. You will feel like you're worse at the material during an interleaved session than during a blocked session. You probably are, in the short term. Remember that the short term is not the goal.

If you're working from a textbook that presents topics in blocks, you can interleave by reviewing earlier chapters alongside current ones. Mix end-of-chapter problems from different sections in the same session. Create your own mixed sets. It requires a little more effort to set up, but the learning is worth it.

Notice the confusion. When you mix problem types and start mixing up your approaches, that's not a sign to go back to blocked practice. That's the signal that you're doing the difficult part of learning. The confusion is the work. Sit with it.

Track long-term results, not session-to-session performance. If you're doing practice problems, save a few from each topic type and test yourself a week later. Compare your interleaved sessions against your blocked sessions on that delayed test. The evidence suggests you'll prefer interleaving when you have that data.

The Real Question

Blocked practice is comfortable. It produces the feeling of progress. The research suggests that feeling is mostly accurate in the short term and mostly misleading in the long term.

Interleaving is uncomfortable. It produces the feeling of being worse than you are. But the learning it builds is more flexible, more durable, and more transferable to new situations.

These aren't equally valid options. The evidence favors interleaving for most forms of learning that involve discrimination or categorization. But you don't have to take my word for it. You can run this experiment on yourself and see what your own data tells you.

So what's actually happening in your study sessions right now? Are you the type who finishes one chapter before moving on, or do you naturally hop between topics? And when you test yourself a week later, do you feel like you know more than you actually do?