AI teaching agents made a university classroom feel more present and connected to students, but did not produce a statistically significant increase in learning engagement.
The distinction emerges from a study published November 2 in the Journal of Computer Assisted Learning. Researchers followed 36 first-year educational-technology students through a ten-week course, refining AI teaching agents over three successive design stages.
The first added gestures to presentations. The second introduced real-time questions and answers mediated by the teacher. The third offered individualized text-based question answering.
Questionnaires, observations, student feedback and discussion records tracked the classroom experience. Students’ reported sense of presence increased significantly across the iterations. Engagement scores edged upward descriptively, but the change did not meet the study’s statistical significance threshold.
The findings add nuance to discussion of AI assistants in university study and the skills teachers need to use AI effectively.
There was no control group, and the designs were introduced sequentially. Consequently, the study cannot isolate which feature caused the observed changes.
For educators, the practical lesson is to evaluate classroom experience and learning engagement separately. An assistant can feel more responsive without demonstrably making students more engaged.