“I Am Confused! How to Differentiate Between…?” Adaptive Follow-Up Questions Facilitate Tutor Learning with Effective Time-On-Task

It also aims to develop students' ability to interpret, discriminate, and critically select among feedback from AI, teachers, and peers.

“(2) develop their ability to interpret, discriminate, and critically select among feedback from AI, teachers, and peers [128-130];”

AI often acts as a dialogic coach or teachable agent that organizes a sequence of demonstration, explanation, questioning, and feedback.

“Beyond answering, AI often acts as a dialogic coach or teachable agent that organizes a sequence of demonstration, explanation, questioning, and feedback, as illustrated by the APLUS mathematics environment and curiosity question training agents [6,102,128].”

As a learning partner, AI provides high-quality concept maps, instructional analogies, and personalized feedback, creating deep learning experiences and prompting reflection on socio-ethical implications of technology, thus seeding epistemic insight.

“As a learning partner, AI provides high-quality concept maps [146], instructional analogies [94], and personalized feedback [128,130], creating deep learning experiences [95] and prompting reflection on socio-ethical implications of technology-thus seeding epistemic insight [117,132,135,138,151].”

These gains tend to depend on structured pedagogical design such as adaptive questioning, error analysis, or project-based learning.

“Notably, these gains tend to depend on structured pedagogical design such as adaptive questioning [127,128], error analysis [77,103], or project-based learning [108].”