Two selected real studies illustrate annotation decisions. They are not an exhaustive evidence review, student submissions or tests of an AI tool. The annotations are original editorial examples; their relevance statements refer to the example question.

Example question: How do practice conditions shape retention in university learning?

A1
Wiklund-Hörnqvist, C., Jonsson, B., & Nyberg, L. (2014). Strengthening concept learning by repeated testing. Scandinavian Journal of Psychology, 55(1), 10–16. https://doi.org/10.1111/sjop.12093
Full text: https://pmc.ncbi.nlm.nih.gov/articles/PMC4235419/

Summary: In a randomized comparison involving 83 cognitive-psychology undergraduates, repeated testing with correct-answer feedback produced better key-concept performance than rereading. Assessments occurred after five minutes, about 18 days and five weeks; 63 students returned for the last assessment.

Evaluation: The course-based material makes the comparison relevant to university study. However, the authors acknowledge longer item exposure in the feedback condition. The experiment therefore cannot isolate feedback from the rest of the practice procedure, and its experienced student sample limits generalization.

Relevance: For this question, I would use the study as evidence about a combined practice procedure, while avoiding a claim that feedback alone caused the advantage.

Evidence locator: Methods → Participants; Materials and design; Procedure → Learning phase and Delayed tests. Results → Figure 2 and following analyses. Discussion → paragraphs beginning A limitation and In addition.
2014 journal issue; online publication was December 7, 2013.

---

A2
Greving, S., & Richter, T. (2018). Examining the testing effect in university teaching: Retrievability and question format matter. Frontiers in Psychology, 9, Article 2412. https://doi.org/10.3389/fpsyg.2018.02412
Full text: https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2018.02412/full

Summary: This lecture experiment compared short-answer and multiple-choice review without feedback against summary restudy. Tests occurred 1, 12 and 23 weeks after the last lecture. A short-answer advantage emerged for highly retrievable items; no multiple-choice advantage was detected.

Evaluation: The field setting connects the question to ordinary teaching, but outside studying and voluntary participation complicate interpretation. The authors caution against declaring multiple-choice questions generally ineffective. The participants section reports 137 students; Table 2 models use 92 or 91, so those counts describe different scopes.

Relevance: I would use this source to qualify a general recommendation about testing: the kind of question and initial retrievability matter. It cannot establish a universal ranking of quiz formats.

Evidence locator: Materials and Methods → Participants, Procedure and Design. Results → Effects of Testing With Short-Answer Questions; Effects of Testing With Multiple-Choice Questions; Table 2. Discussion → final limitations paragraphs.