In one sentence
For required 8th-grade science terms, writing down what you recall beat silent mental recall — and quiet retrieval looked about as helpful as restudying the definitions.
What the researchers did
Witherby, Northern, and Tauber asked whether response mode — overt versus covert retrieval — changes how well retrieval practice works for adolescents in a real classroom. Eighth-graders learned science concepts required for their course inside middle-school classrooms.
Across five practice sessions, students either recalled and wrote terms (overt retrieval), mentally recalled them without writing (covert retrieval), or restudied definitions. Final tests came two days later. The design compares these strategies in the same curricular setting rather than in a decontextualised lab list.
What they found
- Covert retrieval produced learning gains similar to restudy.
- Both covert retrieval and restudy were less beneficial than overt retrieval.
- How well the full treatment package worked varied somewhat across students and terms.
- Outcomes align broadly with a retrieval-dynamics account: producing an answer (here, in writing) can matter for classroom retrieval practice, not only “thinking of” the answer.
What this means for learners and educators
- In class, prefer brief written or spoken recall over “review it in your head” when time allows — silent self-testing may feel productive while tracking closer to rereading.
- Low-tech prompts (exit tickets, blank-paper recall) can capture the overt advantage without new software.
- Do not assume every student and every term will show the same gap — monitor a few items and adjust.
- Pair overt recall with feedback so wrong written answers do not harden.
Limitations and what we don't know yet
This was one grade level, science terms, and a two-day delay — longer retention and other subjects remain open. Abstract-level reporting does not give exact means, sample size, or how feedback was delivered after each mode. “Covert” is hard to verify: some students may have peeked or under-engaged. Effects that differ by student and term complicate one-size classroom rules. We also lack detail on whether speaking answers aloud would match writing.