AI-directed problem selection may diminish the metacognitive dimension of self-testing while preserving retrieval practice¶

Claim · All claims
Evidence · 1 study · theoretical r3 · q1
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q1 i? When students generate their own practice problems they exercise metacognitive judgment that is itself generative; when an AI selects and delivers problems through an adaptive algorithm, this metacognitive dimension may be diminished even though retrieval practice is preserved. → Hoyun Cho 2026

Evidence¶

Hoyun Cho 2026¶

Hoyun Cho. (2026). Relocating the Locus of Generation: The AI-GLM Framework for Generative Learning in AI-Mediated Mathematics Education. Educational Psychology Review. https://doi.org/10.1007/s10648-026-10210-y

q1 · i? · theoretical · r3

Conceptual analysis of the self-testing strategy under AI mediation. The article states the tension is between retrieval-practice benefits, which AI can preserve or enhance, and metacognitive self-regulation benefits, which AI may diminish; no empirical test of this dissociation is reported.

"When students generate their own practice problems or decide what to test themselves on, they exercise metacognitive judgment about what they know and need to practice, which is itself a form of generative activity."

Discussion¶

Related Claims¶

• Agentivism generates six empirically testable propositions linking delegation structure, verification, reconstruction, and reduced-support transfer to learning outcomes — a broader claim this one bears on
• Reviewed literature reports risks of overreliance, AI errors in complex tasks, and students' uncritical acceptance of AI-generated outputs — related
• Students exercise ethical agency in GenAI use through human-AI collaboration strategies, personal ethical frameworks, strategic calibration of use, and resistance to over-reliance — related
• Learners misjudge which learning strategies are effective — related
• Advanced learners may benefit from learner-controlled selection of learning tasks and levels of generality — related