ARPG+ produces higher final prompt quality than unguided, template, and rule-based conditions among simulated learners¶

Claim · All claims
Evidence · 1 study · causal r1 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? ARPG+ achieves substantially higher final quality (Q20 = 7.82 ± 0.32) than Rule (6.48 ± 0.51), Template (5.95 ± 0.58), and Baseline (4.52 ± 0.75), corresponding to 4.60 points total gain (142.9% improvement). → Ye 2026

Evidence¶

Ye 2026¶

Ye, P.-G., Mo, K., Long, Y., Liu, M., Sang, H., & Zheng, J. (2026). ARPG+: a simulation-based study of real-time coaching for educational LLM prompting. International Journal of Educational Technology in Higher Education. https://doi.org/10.1186/s41239-026-00606-9

q2 · i? · causal · r1

Simulation study with 1000 LLM-based simulated learners evenly distributed across four coaching conditions over 20 turns. ARPG+ final quality "7.82 ± 0.32" versus Rule 6.48, Template 5.95, Baseline 4.52, a "4.60 points total gain (142.9% improvement)". No inferential test statistic is printed.

"ARPG+ achieves substantially higher final quality ( Q20 =7 .82 ± 0.32) than Rule (6.48 ± 0.51), T emplate (5.95 ± 0.58), and Baseline ( 4.52 ± 0.75), corresponding to 4.60 points total gain (142.9% improvement)."

Discussion¶

Related Claims¶

• Adaptive fading with periodic reinforcement yields higher independence than fixed-high or linear-decay scaffolding in simulation — related
• ARPG+ reaches quality milestones faster than comparison conditions in simulation — related
• ARPG+ shifts cognitive load composition toward productive effort and reduces overload frequency in simulation — related
• Theory-guided five-step prompting of AI interaction improves open-ended explanation performance in construction education review, while unprompted AI access alone does not — related
• AURA's within-session reinforcement learning improved composite response quality over non-adaptive baselines (p = 0.044, d = 0.66), with fewer specification prompts and more validation behavior — related