Without initial instruction, varied retrieval practice matches varied worked examples and beats repeated retrieval practice for generalization¶

Claim · All claims
Evidence · 1 study (3 entries) · causal r2 · q3 · i2 medium
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i2 In Experiment 2 the variability × strategy interaction was significant (ηp² = .06), unlike in Experiment 1. → Meng Cao & Paulo F. Carvalho 2026 q3 i? With varied items there was no significant difference between retrieval practice (M = 0.64) and worked examples (M = 0.67). → Meng Cao & Paulo F. Carvalho 2026 (2) q3 i? Varied retrieval practice produced significantly better generalization than repeated retrieval practice. → Meng Cao & Paulo F. Carvalho 2026 (3)

Evidence¶

Meng Cao & Paulo F. Carvalho 2026¶

Meng Cao & Paulo F. Carvalho. (2026). Striking the Balance: How Variability Shapes Retrieval Practice and Worked Examples for Transfer Learning. Educational Psychology Review. https://doi.org/10.1007/s10648-026-10169-w

q3 · i2 · causal · r2

Experiment 2's mixed-effects ANOVA on final-test accuracy showed a significant strategy × variability interaction with printed effect size ηp² = .06 ("F(1,72) = 4.78, p = .032, ηp 2 = .06"), contrasting with the non-significant interaction in Experiment 1.

"The interaction between variability and learning strategy was also statisti-cally significant, F(1,72) = 4.78, p = .032, ηp 2 = .06, as depicted in Fig. 3."

Meng Cao & Paulo F. Carvalho 2026 (2)¶

Meng Cao & Paulo F. Carvalho. (2026). Striking the Balance: How Variability Shapes Retrieval Practice and Worked Examples for Transfer Learning. Educational Psychology Review. https://doi.org/10.1007/s10648-026-10169-w

q3 · i? · causal · r2

Simple-effects analysis in the varied-items condition of Experiment 2 found no significant strategy difference, "t(72) = −0.48, p = .635"; equivalence was not formally tested and no effect size was printed.

"varied items, there was no significant difference between those who practiced with retrieval practice ( M = 0.64, SE = 0.06) and those who practiced with worked examples (M = 0.67, SE = 0.06), t(72) = −0.48, p = .635."

Meng Cao & Paulo F. Carvalho 2026 (3)¶

Meng Cao & Paulo F. Carvalho. (2026). Striking the Balance: How Variability Shapes Retrieval Practice and Worked Examples for Transfer Learning. Educational Psychology Review. https://doi.org/10.1007/s10648-026-10169-w

q3 · i? · causal · r2

Within the retrieval practice strategy of Experiment 2, varied items beat repeated items on the final test, "t(126) = − 2.16, p = .033"; no effect size was printed.

"participants who engaged in varied retrieval practice performed significantly better than those who engaged in repeated retrieval practice, t(126) = − 2.16, p = .033."

Discussion¶

Related Claims¶

• Without initial instruction, repeated worked examples produce better generalization than repeated retrieval practice — related
• When learners receive initial instruction, retrieval practice yields better generalization than worked examples regardless of item variability — related
• Retrieval practice takes more learning time than worked examples, and variability slows improvement during practice — related
• Item variability shows no overall main effect on generalization in either experiment — related