Protocol
Use near-neighbour cases to train discrimination
Some errors are not missing knowledge; they are missing contrast.
When it fits
- When two plausible methods, categories or diagnoses are repeatedly confused.
When to avoid it
- Interleaving is not uniformly beneficial. Use this design when categories must be discriminated and the comparison clarifies rather than overloads.
Why it matters
Put easily confused cases close enough that the learner must notice the feature that changes the decision. Compare one case where option A fits with a near neighbour where B fits, then ask for the discriminating cue before revealing the answer.
Steps
- Choose two commonly confused options.
- Create or select near-neighbour cases that differ on the decisive feature.
- Ask the learner to name the cue before choosing.
- Mix fresh near neighbours until the distinction survives.
An example
Two customer-master scenarios look similar, but only one has the organizational data needed for a particular process. The practice pair makes that cue explicit.
Check your result
The learner can classify fresh near-neighbour cases and explain the feature that flips the decision.
Keep this limit in mind
- Interleaving is not uniformly beneficial. Use this design when categories must be discriminated and the comparison clarifies rather than overloads.
Connected ideas
Useful withInterleave only where choosing matters
Evidence and sources
A meta-analysis of interleaving found a moderate overall benefit (g = 0.42) but strong moderation by material type and category similarity, including conditions where blocking performed better.
Interleaving should be selected for a discrimination problem rather than used as a universal mixing rule.
Similarity matters: A meta-analysis of interleaved learning and its moderators · Abstract