Trap-based problem chains score slightly lower than context-based and summary-based problem chains in usability¶

Claim · All claims
Evidence · 1 study · design r2 · q2
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Usability scores for trap-based problem chains were slightly lower than for the other two chain types, and did not exceed 6.5 outside the Statistics module. → Bihao Hu 2025

Evidence¶

Bihao Hu 2025¶

Bihao Hu, Jiayi Zhu, Yiying Pei & Xiaoqing Gu. (2025). Exploring the potential of LLM to enhance teaching plans through teaching simulation. npj Science of Learning. https://doi.org/10.1038/s41539-025-00300-x

q2 · i? · design · r2

Evaluation of refined context-based, trap-based, and summary-based problem chains across knowledge modules (Fig. 4b); the article reports trap-based chains scored above 6.5 only in Statistics and above 6.0 elsewhere, i.e., "slightly lower than those forother types of questions."

"However, the analysis also reveals that the usability scores for trap-based problem chains are slightly lower than those forother types of questions, suggesting that LLMs still need improvement in designing deep and challenging questions centered around core knowledge."

Discussion¶

Related Claims¶

• Fine-grained prompting substantially improves question uniqueness and usability of LLM-generated educational questions compared with Chain-of-Thought prompting — related