Tutor Mode's adaptive responsiveness was overestimated by simulation: only 3/8 pilot participants agreed LEA adapted appropriately when they expressed uncertainty¶

Claim · All claims
Evidence · 1 study (2 entries) · design r2 · q2
Currency · newest study 2026 · checked against its sources 2026-10-20 (automated, 2 of 2 entries) · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly

Subclaims¶

q2 i? Only 3/8 participants agreed LEA adapted appropriately when they expressed uncertainty (median 4, Neutral), the most consequential divergence from simulation predictions. → Rumble 2026 q2 i? Participants reported affirmation fatigue and scaffolding rigidity in Tutor Mode, corroborating the simulation's lower Learner State Match scores for low-knowledge profiles. → Rumble 2026 (2)

Evidence¶

Rumble 2026¶

Rumble, T., Zarrin, J., Lovell, P. G., Falconer, R. (2026). Learning Engagement Assistant (LEA): Cross-Course Scalability and Classroom Evaluation of an Agentic AI Tutoring System. arXiv. https://arxiv.org/abs/2607.13370 doi:10.48550/arxiv.2607.13370

q2 · i? · design · r2

Pilot survey Tutor Mode items from the eight-participant CMP511 study. Socratic questioning was endorsed by 7/8, but adaptation when uncertain drew only 3/8 agreement at median 4, which the authors call the most consequential divergence from simulation predictions.

"The Socratic questioning approach was endorsed by 7/8 participants as helpful for deeper thinking, but only 3/8 agreed thatLEAadapted appropriately when they expressed uncertainty (median 4, Neutral)"

Rumble 2026 (2)¶

Rumble, T., Zarrin, J., Lovell, P. G., Falconer, R. (2026). Learning Engagement Assistant (LEA): Cross-Course Scalability and Classroom Evaluation of an Agentic AI Tutoring System. arXiv. https://arxiv.org/abs/2607.13370. doi:10.48550/arxiv.2607.13370

q2 · i? · design · r2

Reflexive thematic analysis of open-text pilot responses identified Tutor Mode scaffolding as rigid under genuine confusion and motivational affirmations as performative for some postgraduate learners, with one participant copy-pasting LEA's own text to progress the dialogue.

"Two participants independently reportedaffirmation fatigue: motivational affirmations felt "disconnected from the conversation flow," with one requesting their removal."

Discussion¶

Related Claims¶

• In an exploratory pilot (n=8), students rated LEA's usability, Chat Mode, Quiz Mode, learning outcomes, and trust mostly positively, with Chat Mode the strongest performer — related
• In simulation, LEA met or exceeded targets in Chat and Tutor modes while Quiz Mode's KC Alignment Score fell below target — reports the opposite
• A question-repetition bug in Quiz Mode, reported by 2/8 pilot participants, was a defect simulation-based evaluation could not identify — related
• Learner engagement was strengthened when the AI tutor adapted to learners' pace and prior knowledge — related
• Pilot students reported the conversational format forced explanation of reasoning, with a majority agreeing it combats rote memorization — related
• Over 80% of pilot students agreed the AI proctor accurately understood their responses and scaffolded them appropriately — reports the opposite
• Students rated peer tutoring with the VoicePod as acceptable and helpful for learning vocabulary — related