The Worked Example Effect: Why Studying Solutions First Beats Struggling

There is a deeply held belief that the best way to learn something is to wrestle with it. Hand a student a problem, let them struggle, and the struggle itself will teach them. It sounds rigorous and it feels virtuous. For complete beginners, though, a large body of research suggests it is often the slower path.

The alternative sounds almost too easy: show them the solution first. Walk through a fully solved problem, step by step, and let them study it. This is the worked example effect, and it is one of the most consistently replicated findings in instructional research.

Where the idea comes from

The worked example effect grew out of cognitive load theory, developed by the Australian educational psychologist John Sweller from the 1980s onward. The theory starts from a simple fact about the mind: working memory, the mental space where we hold and manipulate information right now, is small. It can juggle only a handful of new elements at once. Long-term memory, by contrast, is vast.

We cover the theory in more depth in our piece on cognitive load. The key point here is that learning means building organised knowledge structures, often called schemas, in long-term memory. Once you have them, they let you treat a whole pattern as a single unit, which frees up working memory for new material.

Why unguided problem solving overloads novices

Imagine you are new to algebra and are asked to solve an equation with no guidance. You have to keep the goal in mind, hold the current state of the equation, consider which operations might help, try one, check whether it moved you closer, and backtrack if not. That is a lot of juggling for a small workspace.

Sweller and colleagues argued that novices typically fall back on a general strategy called means-ends analysis: look at the gap between where you are and where you want to be, and try moves that shrink it. This can eventually produce a correct answer, but it consumes so much working memory that little is left over for noticing the underlying pattern. You can solve the problem and still not learn how problems like it are solved.

A worked example removes that burden. The search is already done. The learner's working memory is free to follow each step and, crucially, to see why each step was taken. That is exactly the material schemas are built from.

What the research found

In a series of experiments, mainly in mathematics, physics and other structured domains, students who studied worked examples tended to perform better on later tests than students who spent the same time solving equivalent problems unaided. Often they also took less time and reported less mental effort. The effect has since been replicated across many subjects and age groups.

Not all worked examples are equally good. Research on the effect identified several things that help:

• Integrate words and diagrams so learners do not have to flip between them and hold both in mind.
• Break the solution into clearly labelled steps, which makes the structure visible.
• Pair an example with a similar practice problem, so the learner studies one and then tries one.
• Prompt self-explanation: asking learners to explain each step in their own words tends to deepen understanding.

The expertise reversal effect

Here is the twist that keeps this from becoming a simple rule. As learners gain knowledge, the advantage of worked examples shrinks. Eventually it can disappear or even reverse. Researchers, including Slava Kalyuga working with Sweller, called this the expertise reversal effect.

The reason follows from the same theory. Once a learner has built the relevant schemas, a fully worked example becomes redundant. They already know the steps, so reading them is extra processing that adds load without adding much learning. At that stage, solving problems, and especially retrieving the method from memory, does more good than reading another solution.

This means the right teaching method depends on who is learning. The same worked example that helps a beginner may slow down someone with more experience.

Faded examples: the bridge

If novices need full examples and more experienced learners need practice, what about everyone in between? One well-studied answer is faded worked examples. You begin with a fully worked solution. The next example leaves the final step blank for the learner to complete. The one after leaves the last two steps blank, and so on, until the learner is solving whole problems alone.

Fading lets guidance shrink in step with growing skill, so the learner is always working near the edge of what they can handle without being overwhelmed.

How this fits with desirable difficulty

If you have read about desirable difficulty, this may sound like a contradiction. Desirable difficulty says that some struggle, such as retrieval practice or spacing, improves long-term learning. The worked example effect says struggle can hurt. Which is it?

The two fit together once you add timing. A difficulty is desirable only if the learner can respond to it successfully. For a true beginner, unguided problem solving is not a productive challenge. It is simply overwhelming. Once some schemas exist, effortful retrieval and independent practice become the kind of difficulty that strengthens memory. Guidance first, struggle later.

This also connects to Benjamin Bloom's observations about tutoring, discussed in our article on the two sigma problem. A good tutor naturally adjusts how much help to give, demonstrating when a student is lost and stepping back when the student is ready. Worked examples and fading are ways of building some of that adjustment into materials.

Practical takeaways for self-learners

1. 1When a topic is brand new, look for fully solved examples before attempting problems. Textbooks with worked solutions are valuable for exactly this reason.
2. 2Study examples actively. Pause at each step and explain to yourself why it was taken.
3. 3Alternate: study one example, then try a similar problem on your own.
4. 4Fade the support as you improve. Cover the later steps of an example and try to complete them before checking.
5. 5Once problems feel familiar, stop reading solutions and switch to practice and self-testing. If examples start to feel boring, that is often a sign you have outgrown them.

One idea at a time

The same principle shapes how we design MindSnap. MindSnap is our app: each daily lesson walks through a single idea step by step, usually through a story, before a quick quiz asks you to recall it. The lesson is the worked example and the quiz is the first, small attempt on your own. It is a light version of the pattern, not a full course, but the order is deliberate.

Struggle has its place in learning. It just works best once you have something to struggle with.