Syntes: Abel et al. (2019) — with corrective feedback, no retrieval-over-restudy advantage at a 12-hour delay

Källa: Abel, M., Haller, V., Köck, H., Pötschke, S., Heib, D., Schabus, M., & Bäuml, K.-H. T. (2019). Sleep reduces the testing effect—But not after corrective feedback and prolonged retention interval. Journal of Experimental Psychology: Learning, Memory, and Cognition, 45(2), 272–287. https://doi.org/10.1037/xlm0000576

Verified numbers (full text read 2026-10-05)

• Design (Exp 1a/1b): 36 unrelated paired associates studied once (5 s each); one third restudied (intact pair, 8 s), one third retrieval practice without feedback (cue + first letter, 8 s), one third retrieval practice with corrective feedback (5 s attempt + 3 s intact pair). Equal 8 s per item across practice types. Final cued recall (cue + first letter) after 5 min, 12 h wake (9 a.m.→9 p.m.) or 12 h including nocturnal sleep (9 p.m.→9 a.m.); n = 36 per delay group.
• Exp 1a (one practice cycle, N = 108): 5 min: restudy 74.3% > retrieval with feedback 65.1% (d = 0.65) > without feedback 52.8%. After 12 h sleep restudy was still higher than retrieval with feedback (all ts(35) ≥ 2.06, ds ≥ 0.34); after 12 h wake recall no longer differed between the three practice types (F(2,70) = 1.24, p = .295). No test × delay interaction for restudy vs retrieval with feedback (wake p = .153; sleep p = .307).
• Exp 1b (three practice cycles, N = 108): retrieval success with feedback 70.3% → 95.3% → 99.2% across cycles. 5 min: restudy 92.7% vs retrieval with feedback 93.4% (ns). After 12 h sleep: restudy 85.8% vs retrieval with feedback 88.9%, no difference; after 12 h wake: restudy 73.6% vs retrieval with feedback 75.4%, and recall did not differ between practice types (F(2,70) = 1.08, p = .337). No test × delay interaction for restudy vs retrieval with feedback (Fs < 1.0).
• Exp 2a/2b (no feedback): without feedback, 12 h sleep removed the reduced forgetting after retrieval; at 24 h (Exp 2b, practice success ~87%) restudy and retrieval did not differ; at 7 days a small direct testing effect appeared only when Exp 2a and 2b were pooled (34.1% vs 29.6%, p = .031).
• Authors' reading: the absent direct testing effect at 12 h is "very likely attributable" to restudy's advantage at short delay (Exp 1a) — but in Exp 1b the short-delay levels were equal and the 12-h levels still did not differ.

Role for C0

Bound on the meaning of "delayed": at a 12-hour delay, equal-time restudy matched retrieval practice with corrective feedback even after three cycles with near-perfect retrieval success. The retrieval-over-restudy advantage with feedback in the bank is shown at delays of about a day or more (Gupta, Pan & Rickard 2026: 24 h; Gupta, Pan & Rickard 2022: ~48 h; Higham et al. 2023 Exp 3: 2 days). C0 weakened 2026-10-05; resolves open question 1. Caveats: short unrelated word-pair lists, within-subject practice types, near-ceiling recall in Exp 1b at 5 min.