AI-powered PBL/CBL outperforms traditional lectures for knowledge acquisition, but shows no significant advantage over AI-free PBL/CBL¶

Claim · All claims
Evidence · 1 study (2 entries) · quant-synthesis r2 · q3
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i? Subgroup analysis versus traditional lectures showed knowledge acquisition 70% higher in the AI group, statistically significant (95% CIs [0.24-1.15], P = .003); no point estimate was printed. → Wei H 2025 q3 i? Subgroup analysis versus traditional PBL/CBL showed a 31% advantage for AI that was not statistically significant (95% CIs [-0.33 to 0.66], P = .07). → Wei H 2025 (2)

Evidence¶

Wei H 2025¶

Wei H, Dai Y, Yuan K, Li KY, Hung KF, Hu EM, Lee AHC, Chang JWW, Zhang C, Li X. (2025). AI-Powered Problem- and Case-based Learning in Medical and Dental Education: A Systematic Review and Meta-analysis. International Dental Journal. https://doi.org/10.1016/j.identj.2025.100858

q3 · i? · quant-synthesis · r2

Subgroup meta-analysis pooling two RCTs (CC-Cruiser and ChatGPT-assisted PBL) against traditional lectures; the AI group's advantage was significant at P = .003 with 95% CIs [0.24-1.15]; no point SMD printed.

"Subgroup analysis of the AI intervention group and the traditional lectures group18,32 showed that the knowledge acquisition in the AI group was 70% higher than that in the traditional lectures group, and the difference was statistically significant (95% Cls [0.24-1.15], P = .003)."

Wei H 2025 (2)¶

Wei H, Dai Y, Yuan K, Li KY, Hung KF, Hu EM, Lee AHC, Chang JWW, Zhang C, Li X. (2025). AI-Powered Problem- and Case-based Learning in Medical and Dental Education: A Systematic Review and Meta-analysis. International Dental Journal. https://doi.org/10.1016/j.identj.2025.100858

q3 · i? · quant-synthesis · r2

Subgroup meta-analysis pooling two RCTs (IIAS-supported CBL and WFO-integrated CBL) against AI-free PBL/CBL; the 31% advantage was not statistically significant (P = .07), so equivalence was not tested.

"Subgroup analysis of the AI intervention group and the traditional PBL/CBL group31,33 indicated that the knowledge acquisition in the AI group was 31% higher than those in the traditional group, but this difference was not statistically significant (95% Cls [-0.33 to 0.66], P = .07)."

Discussion¶

Related Claims¶

• AI-powered PBL/CBL shows favourable but heterogeneous effects on clinical reasoning capability that could not be statistically pooled — related
• AI-powered PBL/CBL significantly improves students' knowledge acquisition compared with PBL/CBL without AI or traditional lectures — possibly the same claim (merge candidate)
• AI-supported PBL students reported higher computational thinking on a self-report scale than traditional-lecture students, controlling for pretest scores — related
• AI-supported PBL students obtain significantly higher posttest academic achievement scores in robotics programming than traditional-lecture students when controlling pretest scores — related
• TBL alone and TBL combined with other methods both outperform LBL, with no significant difference between the two TBL formats — related
• In a randomised trial of 80 first-year postgraduate oncology trainees, CBL produced significantly higher examination scores and satisfaction than traditional lecture-based learning, though confidence intervals crossed the null for examination results — related
• Student preference for CBL, PBL or TBL over traditional teaching was reported in 15 of 22 studies (68.1%), but only two studies used a controlled comparison with lecture — related
• Problem-based learning shows a small but significant pooled advantage over lecture-based learning in dental education (SMD = 0.33) — related
• Most reviewed studies found active learning improved satisfaction and knowledge acquisition and was superior to traditional lectures — related
• PBL methods produce significantly better knowledge scores than traditional teaching among dental students (pooled MD 4.15) — related