AI-generated revision podcasts produce statistically significant immediate learning gains in medical students at both study centres¶

Claim · All claims
Evidence · 1 study · causal r1 · q2 · i2 medium
Currency · newest study 2026 · checked against its sources 2026-10-20 (automated, 1 of 1 entry) · not reviewed by a person

Subclaims¶

q2 i2 Paired pre-to-post MCQ gains were significant at both centres, with mean differences of 3.9 at Centre 1 (dz 0.42) and 7.3 at Centre 2 (dz 0.55). → Joshi U 2026

Evidence¶

Joshi U 2026¶

Joshi U, Sarkar A, Shukla A, Shrimali S, Bezbaruah NK, B Sanjay Kini, Baisil S, Sheth Z, Bhargava P, Shah S. (2026). Ears for examinations: a multicentre quasi-experimental evaluation of AI-generated revision podcasts on learning outcomes and retention among medical students. BMC Medical Education. https://doi.org/10.1186/s12909-026-10080-6

q2 · i2 · causal · r1

Paired within-participant t-tests among linked completers (Centre 1 n = 154, Centre 2 n = 36) in a one-group pre-post-delayed quasi-experiment. The article prints paired effect sizes "dz 0.42" and "dz 0.55" (medium), with the quoted mean differences of 3.9 and 7.3.

"Immediate learning gains from pre-test to post-test were observed at both centers, with mean differences of 3.9 at Centre 1 (dz 0.42, p < 0.001) and 7.3 at Centre 2 (dz 0.55, p = 0.002)."

Discussion¶

Related Claims¶

• Retention trajectories diverged by centre: Centre 1 retained gains at 30 days while Centre 2 showed significant post-to-delayed decay — related
• Podcast gains extend beyond recall to higher-order application and analysis items at both centres — a narrower finding that bears on this claim
• Higher episode completion (>50%) is independently associated with greater learning gain, and the centre effect on gain is mediated by engagement — related
• Learners valued AI revision podcasts for portability and 'dead time' use but identified absent pauses and AI voice monotony as design barriers — related
• Topic-wise gains were heterogeneous, with some topics showing no meaningful change — related
• For within-subjects designs, the author recommends reporting Cohen's d_rm or Cohen's d_av rather than Cohen's dz — related
• Students show large immediate pre-to-post knowledge gains within a single LLM-supported learning session — related
• Students improved significantly from pre- to post-test in both topics, with a medium effect in experimental design and a small effect in graphing — related