Cerdán, R., Vidal-Abarca, E., Martinez, T., Gilabert, R., & Gil, L. (2009). Impact of question-answering tasks on search processes and reading comprehension. Learning and Instruction, 19(1), 13–27

The authors compare the effects of low-level and high-level Adjunct questions on subsequent recall and inferential questions. They hold constant the textual coverage, i.e. writing more low-level questions to ensure that the same set of material is interrogated by the two question groups. They find no significant difference in recall performance, but ==students with high-level questions perform much better on inference questions which require deep comprehension==. Also, the students answering high-level questions take less total time. I don’t really understand this—maybe it’s because they’re required to compose answers, unlike as in Implicit retrieval practice?

This seems consistent with the finding in Agarwal, P. (2019). Retrieval Practice & Bloom’s Taxonomy: Do Students Need Fact Knowledge Before Higher Order Learning? Journal of Educational Psychology, 111, 189–209 that performance on higher-order test questions requires higher-order practice. It also seems consistent with the finding in Pan, S. C., & Rickard, T. C. (2018). Transfer of test-enhanced learning: Meta-analytic review and synthesis. Psychological Bulletin, 144(7), 710–756 that “broad encoding methods” produce more transfer.

This calls into question Spaced repetition memory prompts should usually focus on one atomic unit in the context of Retrieval practice and transfer learning.

Q. Main finding?
A. High-level Adjunct questions produce similar recall performance to low-level ones, but much better performance on transfer tasks.