Expose examples before naming the rule (inductive first)
Present several cases before stating the principle; let the learner induce the pattern.
Why it works
When a learner encounters examples before the rule, they engage in active hypothesis-testing, trying to infer the pattern. This effortful induction — even when it is partially wrong — is a desirable difficulty: the struggle strengthens encoding of the rule once it is revealed, compared to learners who received the rule first and merely verified it.
How to do it
- Present two or three varied examples without explaining what they have in common.
- Ask yourself or the learner: "What do these share? What’s the rule?"
- After the attempt (right or wrong), reveal the principle and debrief where the induction matched.
- Apply the rule to one more novel example immediately to consolidate.
Evidence
Research on productive failure (Kapur, 2016) finds that learners who struggle with problems before instruction outperform those who receive instruction first on transfer tests, even though they perform worse on immediate tests. (rct)
The effect depends on the learner having enough prior knowledge to make any meaningful attempt. True novices may need a worked example first.
Sources
- Kapur (2016), productive failure in learning math, Educational Psychologist
Common mistake
Revealing the rule immediately after the examples without pausing for the learner’s own hypothesis — this collapses induction into passive receipt and loses the encoding benefit.
Practice this with IX Coach
7 days free, then $40/month (~$1.30/day).
More practices for Concrete Examples: How to Make Abstract Ideas Stick
- Generate your own example after learning a concept
Immediately after encountering a concept, create a fresh example from your own life or field.
- Use multiple varied examples to map the concept’s true shape
Three different examples reveal what one example hides — the edges of the concept.
- Cycle from concrete to abstract and back
Start with a story, name the principle, then return to a new application.
- Use analogies to transfer structure from known to unknown
Map an unfamiliar concept onto a familiar one that shares the same relational structure.
- Label what each example is an example of
Explicitly tag every example with the concept it illustrates to prevent "inert knowledge."
- Use non-examples to sharpen concept boundaries
Study what a concept is not — cases that almost fit — to prevent over-generalization.
Related concepts
- Make It Stick: The Science of Learning
Retrieval, spacing, interleaving, and elaboration — the evidence-backed core
- Desirable Difficulties: When Harder Means Better Learning
The difficulties worth keeping — and the ones to remove
- Dual Coding: Learn With Words and Pictures Together
Two channels, two routes to memory — when visuals actually help
- The Feynman Technique, Step by Step
Explain it simply, find the gaps, and close them