Revisit material when it is ‘still familiar but not fresh’. In UCLA physics, students who did mixed (interleaved) homework outscored the sequential group by 74% on a transfer exam. There is ‘no universal ideal spacing schedule’.
Research cited
- Samani & Pan (2021) npj Science of Learning: interleaved physics homework → +74% on unfamiliar-problem exam.
- Carpenter et al. (2022) Nature Reviews Psychology: broad benefits of spacing, but no single ideal gap.
- Dunlosky et al. (2013) Psychological Science in the Public Interest: mechanism (retrieval after delay, consolidation, varied context) (Claude: the article names no authors; this is the standard citation for that DOI).
Moves
- 5-question mini-quizzes mixing this week, last week and older material; light marking, answers reviewed.
- 90-second rapid review at the start of class.
- Spread homework across days and topics; interleave problem types so learners must choose a strategy.
- Map where core concepts recur across the term.
Limits
- US K-12 framing; the citations are as printed in the article and weren’t checked against the primary papers.
- College physics ≠ young learners’ vocabulary.
Apply in my context (Claude)
- Interleaving grammar is underused in ELT: mix past simple, present perfect and past continuous items in one homework set after each is taught, so learners must choose. This targets the KET/PET/FCE “which tense?” decision directly.
Connections
Spaced retrieval is the best-supported technique · Familiarity is not mastery - retrieve, explain, contrast, transfer