Design AI tutors for complementarity across role paradigms and move beyond static prompting toward dynamic learner modeling¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The article's forward-looking design recommendations are threefold: use factorial designs crossing role paradigms with content domains to disentangle how an agent teaches from what it teaches; establish educational value with learning-outcome designs using pre- and postlearning assessment and individual-level linkage; and address persistent deficits in adaptive difficulty and feedback by moving "beyond static prompting toward dynamic learner modeling and memory-augmented tutoring." Rather than seeking a single best role, it points toward orchestrating different pedagogical strengths across an interaction.

Design Implications¶

Context¶

Requirements¶

• Factorial designs crossing role paradigms with content domains
• Learning-outcome designs with pre- and postlearning knowledge assessment and individual-level linkage

Constraints¶

• The study confounded role paradigm with content domain, so role-related observations are descriptive rather than causal

Target Learners¶

• medical students in role-play-based curricula

Target Learning Goals¶

• adaptive difficulty calibration
• formative feedback quality
• balanced affective and cognitive teaching functions

Affordances¶

• Eight Dimension Pedagogical Rubric Ai Teaching Agents

Related Strategies¶

• Move beyond static skill-targeted rewriting toward interaction-aware adaptive error explanations
• Use dynamic user models to recognize non-optimal learning behaviors and redirect students toward more effective behaviors
• Design AI-supported VET environments that augment human judgment and strengthen learner agency rather than replicating human instruction (avoiding the Turing Trap)
• Complement outcome measures with continuous process measures to assess social effects and enable adaptive pedagogical agents

Examples¶

-

Key Sources¶

• Zhang Hui, Qu Lihui, Zheng Jianmin, Xiong Yi, Bai Hongbo, Ji Ruiying, Liu Guohui, Chen Wanling, Cheng Zirui, Chen Youbang, Yang Chun-tao. (2026). What Platform Scores Miss: Multidimensional Evaluation of AI Teaching Agents in Medical Education. JMIR Medical Education. https://doi.org/10.2196/96819