Study a fully worked example before attempting a new problem type

Read through a completely solved example and understand each step before you try to solve anything yourself.

Why it works

A novice attempting an unfamiliar problem without examples spends most working memory on "means-end analysis" — testing candidate moves, backtracking, estimating progress toward the goal. This exhausts capacity before schema-building can begin. A worked example keeps the solution path visible, so all available working memory is directed at understanding why each step follows logically — the very process that builds a usable schema.

How to do it

1. Find or construct a complete, well-annotated solution to a representative problem of the type you want to learn.
2. Read the solution step by step, pausing after each step to ask: "Why this step? What rule or principle drove it?"
3. Do not attempt a problem until you can read through the example and predict each step before reading it.
4. Then attempt a closely related problem without looking at the example.

Evidence

Sweller and Cooper (1985) showed in two experiments that groups studying worked examples acquired algebra skills faster and with less investment than groups who solved equivalent problems. The effect has been replicated in mathematics, physics, chess, programming, and medical diagnosis. (rct)

The worked examples effect is strongest for novices. For learners who already have relevant schemas, worked examples may actually impede learning (expertise reversal effect).

Sources

• Sweller & Cooper (1985), The use of worked examples as a substitute for problem solving in learning algebra, Cognition and Instruction

Common mistake

Treating worked examples as things to skim for the answer, rather than as material to actively interrogate step-by-step — the schema is in the why, not the what.