2026/03/31 by Eason Chen, Xinyi Tang, Yvonne Zhao +10 · 1 voice
Computer Science · Psychology · #Calculus (dental) #Clinical Practice #Control (management) #Educational Strategies and Epistemologies #Innovative Teaching and Learning Methods #Quality (philosophy) #Transfer (computing) #Transfer of learning #Visual and Cognitive Learning Processes #cs.HC
paper · pdf · open access · doi:10.48550/arxiv.2604.00142
published in arXiv (Cornell University) (Cornell University)
openalex publication_date 2026/03/31 · arxiv published 2026/03/31 · openalex created_date 2026/04/03 · arxiv updated 2026/05/30 · openalex updated_date 2026/07/28
We conducted a between-subjects experiment (N=92) comparing three conditions in a calculus learning environment: no self-explanation (control), menu-based self-explanation, and open-ended self-explanation with LLM-generated feedback. All conditions showed positive learning gains within a fixed 60-minute practice session, with no significant between-condition differences in post-test performance. On transfer questions, the open-ended condition produced significantly higher-quality explanations than control on "Not Enough Information" (NEI) problems (β=+11.9 percentage points, p=.030), though the corresponding NEI multiple-choice accuracy advantage was not significant (p=.183). Moreover, across all post-test open-ended explanations, the open-ended condition showed a marginally significant advantage (β=+7.3%, p=.057). These findings suggest that LLM-supported open-ended self-explanation can improve explanation quality on NEI transfer problems, with weaker evidence across broader transfer explanation measures. Notably, these effects emerged even though learners in the open-ended condition completed substantially fewer practice problems within the same practice time.