Retrieval practice (taking an initial test rather than restudying) reliably improves later recall of the practiced material, but its benefits typically shrink when the final test differs from the initial one. A classroom study by Barenberg, Berse, Reimann and Dutke (2021) had eighth-graders learn English vocabulary with an initial cued-recall test or restudy, followed by a final test comprising one identical format and three transfer formats; testing benefits appeared in three of the four formats but were smaller when transfer was required. The authors specified an open research agenda: determine the boundary conditions under which retrieval practice supports transfer across diverse final-test formats, disentangle similarity-based from difficulty-driven mechanisms, establish how the initial-test format shapes the range of transfer, and control the receptive/productive and forward/backward asymmetries that can confound tests of mechanism. That agenda remains open: later work (2023–2025) addresses adjacent aspects, but no study implements the specified designs.

Background

Retrieval practice — attempting to recall studied material on an initial test — generally produces better long-term retention than restudying, a phenomenon called the testing effect (Rowland, 2014, meta-analytic estimate g ≈ 0.50). Of particular educational interest is whether this benefit transfers: does an initial test of material in one format also improve performance when the final test uses a different format, as classroom assessment routinely demands? Barenberg, Berse, Reimann and Dutke (2021) studied this question with N = 182 eighth-grade students learning English vocabulary (German–English pairs) in a quasi-experimental design with a study phase, an initial cued-recall test (retrieval practice) or restudy, and a final test comprising four formats: one identical to the initial test (the non-transfer context) and three deviating from it (the transfer context), which included formats requiring productive use of the learned words, such as generating sentences, and tests in the opposite direction of the practiced pair (backward transfer, i.e., being tested on the former cue rather than the former target). Comparing the testing and restudy conditions yielded a significant multivariate effect across all final formats and significant univariate effects in three of the four formats, with smaller retrieval-practice effects whenever transfer was required. Theoretically, the results were discussed against transfer-appropriate processing (memory benefits when the processes required at test match those engaged during learning), elaborative retrieval (the idea that retrieving a word activates elaborative semantic associates), and a distribution-based model of retrieval; the episodic context account — the proposal that successful retrieval reinstates and updates the temporal or episodic context in which an item was studied, later aiding retrieval from multiple contexts (Lehman, Smith & Karpicke, 2014; Karpicke, Lehman & Aue, 2014) — is a further candidate framework for predicting how far testing benefits generalize. A prior meta-analysis of transfer of test-enhanced learning (Pan & Rickard, 2018; 192 effect sizes, 122 experiments) found overall positive transfer (d = 0.40) that was strongest across test formats, with response congruency, elaborated retrieval, and initial-test performance as moderators — but its moderator evidence is correlational and dominated by laboratory designs, so it frames rather than settles the boundary-condition question in classroom vocabulary learning. A systematic review of retrieval practice in authentic school settings (Agarwal, Nunes & Blunt, 2021) likewise concludes that benefits are consistent but that transfer across formats and materials remains underexplored. Two methodological hazards recur in this literature: floor effects (performance so low, e.g. on context-cued fill-in-the-blank items, that a true benefit cannot manifest statistically), and confounding of transfer requirements with test difficulty, since a transfer format may simply be harder rather than demanding different processes. The source study explicitly left unresolved how these hazards and mechanisms should be separated experimentally.

Problem Statement

The open problem is to determine the boundary conditions and mechanisms of retrieval-practice transfer across final-test formats, as specified by Barenberg, Berse, Reimann and Dutke (2021) for the classroom English-vocabulary paradigm. Concretely, an answer must show, through experiments that include multiple transfer formats calibrated a priori within the same design: (1) under which final-test formats retrieval practice (relative to restudy) improves classroom vocabulary performance, with final-test difficulty manipulated orthogonally to transfer requirements so that similarity-based mechanisms (transfer determined by overlap between initial and final test processes) can be distinguished from difficulty-driven mechanisms (transfer determined by the effort or difficulty of the final format); (2) whether transfer to context-cued (fill-in-the-blank) formats is absent or merely masked, by adjusting item counts and item difficulty so that performance is measurably off the floor; (3) how the format of the initial retrieval attempt (e.g., cued-recall variants, multiple choice, short answer) shapes the range of formats to which the testing benefit transfers, testing whether the episodic-context account predicts that range; and (4) whether the conclusions hold when receptive-versus-productive retrieval direction and forward/backward association strengths of the paired-associate material are experimentally controlled rather than confounded with format. These four components are independently necessary and jointly serve one objective — a mechanistically interpreted map of when and why testing benefits generalize beyond the practiced format — and a complete answer should support prescriptive guidance for retrieval-based vocabulary instruction in school.

References

  1. Barenberg, J., Berse, T., Reimann, L., & Dutke, S. (2021). Testing and transfer: Retrieval practice effects across test formats in English vocabulary learning in school. Applied Cognitive Psychology, 35(3), 700–710. DOI: 10.1002/acp.3796, https://doi.org/10.1002/acp.3796 (source of the open research agenda)
  2. Pan, S. C., & Rickard, T. C. (2018). Transfer of test-enhanced learning: Meta-analytic review and synthesis. Psychological Bulletin, 144(7), 710–756. DOI: 10.1037/bul0000151, https://doi.org/10.1037/bul0000151
  3. Rowland, C. A. (2014). The effect of testing versus restudy on retention: A meta-analytic review of the testing effect. Psychological Bulletin, 140(6), 1432–1463. DOI: 10.1037/a0037559, https://doi.org/10.1037/a0037559
  4. Agarwal, P. K., Nunes, L. D., & Blunt, J. R. (2021). Retrieval practice consistently benefits student learning: A systematic review of applied research in schools and classrooms. Educational Psychology Review, 33(4), 1409–1453. DOI: 10.1007/s10648-021-09595-9, https://doi.org/10.1007/s10648-021-09595-9
  5. Lehman, M., Smith, M. A., & Karpicke, J. D. (2014). Toward an episodic context account of retrieval-based learning: Dissociating retrieval practice and elaboration. Journal of Experimental Psychology: Learning, Memory, and Cognition, 40(6), 1787–1794. DOI: 10.1037/xlm0000012, https://doi.org/10.1037/xlm0000012
  6. Karpicke, J. D., Lehman, M., & Aue, W. R. (2014). Retrieval-based learning: An episodic context account. In B. H. Ross (Ed.), The Psychology of Learning and Motivation (Vol. 61, pp. 237–284). Elsevier Academic Press. DOI: 10.1016/B978-0-12-800283-4.00007-1, https://doi.org/10.1016/B978-0-12-800283-4.00007-1
  7. Opitz, B., & Kubik, V. (2024). Far transfer of retrieval-practice benefits: Rule-based learning as the underlying mechanism. Cognitive Research: Principles and Implications, 9(1), Article 65. DOI: 10.1186/s41235-024-00598-y, https://doi.org/10.1186/s41235-024-00598-y
  8. Bernardi, E., Vaughn, K. E., Dunlosky, J., & Rawson, K. A. (2024). Toward mastering foreign-language translations: Transfer between productive and receptive learning. Memory, 32(9), 1235–1246. DOI: 10.1080/09658211.2024.2397043, https://doi.org/10.1080/09658211.2024.2397043
  9. Hughes, G. I., & Thomas, A. K. (2023). Retrieval practice and verbal-visuospatial transfer: From memorization to inductive learning. Journal of Memory and Language, 129, Article 104402. DOI: 10.1016/j.jml.2022.104402, https://doi.org/10.1016/j.jml.2022.104402
  10. Kranz, J., Möller, A., Kaufmann, K., & Tempel, T. (2024). Retrieval practice: A tool for teaching the control-of-variables strategy in science classrooms? The Journal of Experimental Education, 94(3), 420–445 (first published online 2024; print issue 2026). DOI: 10.1080/00220973.2024.2392684, https://doi.org/10.1080/00220973.2024.2392684
  11. Huang, S., Cao, H.-W., Wang, J., & Liu, S. (2025). The moderating role of learning rounds: Effects on retrieval practice and context dependence in digital flashcard foreign language vocabulary learning. Behavioral Sciences, 15(11), 1540. DOI: 10.3390/bs15111540, https://doi.org/10.3390/bs15111540
  12. Brunyé, T. T., Smith, A. M., Hendel, D., Gardony, A. L., Martis, S. B., & Taylor, H. A. (2020). Retrieval practice enhances near but not far transfer of spatial memory. Journal of Experimental Psychology: Learning, Memory, and Cognition, 46(1), 24–45. DOI: 10.1037/xlm0000710, https://doi.org/10.1037/xlm0000710