Date
AI Overview
Why it matters
Online learners often lack personalized peer interactions that support cognitive development and engagement. Existing large language model companions only mimic conversational styles without adapting to students' individual cognitive and psychological states.
Study at a glance
- Question: Can AI peer companions that infer learners' mental states improve students' cognitive engagement during online discussions?
- Setting and data: Ten-week undergraduate course in digital circuit design at a top-50 university in China, including 87 students and 4 teaching assistants.
- Method: Randomized controlled trial comparing AI-mediated discussions to human teaching-assistant discussions, supplemented by automated simulations, expert evaluations, and student surveys.
- AI tools: OnlineMate, Deepseek-V3.2
Key findings
- OnlineMate raised students' average cognitive level to 5.20 compared to 4.10 for standard multi-agent setups in simulations, while also boosting emotional engagement scores.
- In a real university course, students collaborating with OnlineMate agents scored higher in classroom expression, listening, coordination, evaluation, and collaboration than those working with human teaching assistants.
- Student cognitive engagement peaked after five discussion rounds and plateaued beyond four AI agents, suggesting that excessive agents or prolonged discussions risk cognitive overload.
Summary drafted with AI from the full paper and may contain errors. See the full paper for details.
In online learning environments, students often lack personalized peer
interactions, which play a crucial role in supporting cognitive development and
learning engagement. Although previous studies have utilized large language
models (LLMs) to simulate interactive dynamic learning environments for
students, these interactions remain limited to conversational exchanges,
lacking insights and adaptations to the learners' individualized learning and
cognitive states. As a result, students' interest in discussions with AI
learning companions is low, and they struggle to gain inspiration from such
interactions. To address this challenge, we propose OnlineMate, a multi-agent
learning companion system driven by LLMs that integrates the Theory of Mind
(ToM). OnlineMate is capable of simulating peer-like agent roles, adapting to
learners' cognitive states during collaborative discussions, and inferring
their psychological states, such as misunderstandings, confusion, or
motivation. By incorporating Theory of Mind capabilities, the system can
dynamically adjust its interaction strategies to support the development of
higher-order thinking and cognition. Experimental results in simulated learning
scenarios demonstrate that OnlineMate effectively fosters deep learning and
discussions while enhancing cognitive engagement in online educational
settings.
Research Categories
What is the application?
Who is the user?
Who age?
Study design
AI tools
Publication Source
Publisher
arXiv
