Blocks versus mixing
A common way to practice is in blocks. You do ten problems of one kind, then ten of the next kind. A language learner might drill only past-tense verbs, then only future-tense verbs. A guitar student might play one chord over and over, then move to the next.
Blocking feels smooth. By the third problem you have the hang of it, and you fly through the rest.
Interleaving means mixing it up. Instead of ten of the same, you do a mixed set. Problem one might be past tense, problem two future tense, problem three past tense again. Each one could be anything.
Why the harder way helps
Mixed practice feels slower and clumsier. You make more mistakes during practice. Many people take that as a sign it is not working.
But think about what happens in a real situation. A test, a job, or a conversation does not tell you which kind of problem is coming next. Part of the skill is spotting what kind of problem you are facing and choosing the right approach.
Blocked practice skips that step. If the last nine problems needed method A, the tenth probably does too. You never have to choose.
Mixed practice makes you choose every time. Research suggests this helps people do better later, especially when similar-looking problems need different methods. It also tends to space out your practice of each type, which brings in the benefits of spacing too.
When blocking is fine
Interleaving is not always the right tool. When something is completely new to you, a short block makes sense. You need a few tries in a row to understand how the method even works.
A reasonable plan:
- Learn a new type of problem with a few examples in a row.
- Once you can do it, add it into a mixed set with things you have learned before.
- Keep the mix to related topics. Switching between math and history every five minutes is not what this means.
Interleaving works best when the topics are related and easy to confuse, so that telling them apart is part of what you are learning.
Try it
Find a set of practice questions on something you are learning. Instead of doing them in order, pick four from different sections and shuffle them. Before you solve each one, say out loud what kind of problem it is and which approach you will use.