Worked-example study

Get more out of a worked solution than the feeling of having followed it, and know exactly what it will and won't buy you.

When to use this: First time through the material
Learning impact: Strong change
Learning evidence: Strong research tradition
Where you are with the topic: Building understanding
How often to use it: Worth repeating often

When to use it

You are meeting this idea for the first time and need a way in.
The reading feels dense and you want a first map before the details.
You want a dress rehearsal that feels like the real assessment, not a casual chat.

What to ask AI

Type something like: “Give me a worked solution to this one line at a time. Before each line, ask me”.

Ask it to generate material for you to attempt, problems, cues, or a schedule, not finished answers.

Keep the hard part on your side: you retrieve, attempt, explain, or decide before you ask for help.

How it helps

For a novice, studying a complete worked solution is often more efficient than attempting a problem unaided.

Sweller and Cooper (1985) found learners given problems to solve took roughly six times longer than those studying worked examples, with substantially worse post-test performance.

(Cognition and Instruction, 2(1), 59-89, verified.) That's a large effect and it's why worked examples are everywhere.

Now the part that usually gets left out, and that changes how you should use them: the benefit did not extend to post-test problems that varied even modestly from the studied ones. Worked examples buy speed and accuracy on near items.

You do not, by themselves, buy transfer.

Which is the outcome a student actually cares about, because an exam contains problems you haven't seen.

So the honest framing is: a worked example is an efficient way to acquire a method and a poor way to learn to select it.

The method here is about extracting maximum value from the first while not mistaking it for the second.

How to run it

Name a concrete target for worked-example study (narrower than a whole chapter).
Use the copyable prompt below, filling in your topic.
Do the cognitive work yourself before you ask AI to continue.
Stop when you can point to what you can now do that you could not do before.

Example exchange

You

Work this quadratic one line at a time. Before each line, ask me what comes next and why.

The equation is x squared minus 5x plus 6 equals 0. What do you expect the next move to be, and why?

You

Factor into (x minus 2)(x minus 3) equals 0, because 2 and 3 multiply to 6 and add to 5… wait, the middle term is minus 5, so both factors should be negative.

Right idea, check the signs once more, then we'll take the next line.

Copyable prompt

Give me a worked solution to this one line at a time. Before each line, ask me
to predict what comes next and why. Don't move on until I've answered.

The Tell

Here is how you know this method has flipped: you can follow any solution and can't start any problem. That's the exact signature of worked examples used without the transfer half.

In that moment the AI (or the schedule, or the story) did the thinking, and you only recognised a finished product.

Tighten the prompt for worked-example study so you attempt, decide, or retrieve before anything is handed to you.

Principle evidence

Strength of the underlying learning idea, not a claim about AI products.

The underlying learning idea is rated strong. Worked-example effect for novices is a cornerstone of cognitive load theory (Sweller & Cooper, 1985, large efficiency gains on near items; transfer boundary and expertise reversal are load-bearing caveats). explaining the idea in your own words while studying examples (Chi et al.; Renkl) further improves outcomes.

AI delivery evidence

Whether an AI tutor delivers this method well is a separate question.

Principle: strong for novice near performance; AI worked-example generation quality remains speculative.