Skip to main content

Research-to-Practice AI and Collaborative Learning

For district math and science leaders

AI that helps students think together.

When a group gets stuck, a quick answer can end the discussion. We are exploring how AI could help students work through ideas together—and how teachers can shape that support.

A familiar classroom challenge

One group needs help. The teacher is working with another. What keeps the thinking moving?

Why ask students? The people using AI should help shape it.

Middle school students bring firsthand knowledge of what it feels like to solve problems with classmates. In the design study, students tried an AI peer and then helped specify the help, timing and controls they wanted. Their experience revealed a gap between adult-designed support and how students wanted to collaborate.

Two studies informing the work

Design with students. Examine how they use it.

(These early studies inform the design of AI support. They do not establish improved classroom learning outcomes. The district questions and suggested next steps on this page are partnership directions, rather than demonstrated effects.)

What this means for districts: Teachers can help decide what useful AI support looks like.

The next step is to work with teachers on a specific classroom challenge and examine whether the support fits instruction. The mathematics work is in development; science offers a proposed extension that teachers could help shape from the start.

  • Choose a real challenge

    Identify a moment when groups stall, struggle to explain their reasoning or need help comparing ideas.

  • Shape the response

    Decide what AI should notice, when it should offer a prompt and when it should leave students to keep thinking.

  • Test the classroom fit

    Examine student reasoning, teacher usefulness and the practical demands of implementation together.

Explore Partnership

AI and learning design ♦ Shaping a prototype ♦ Testing fit

Partnership Inquiry Form

A Research Partnership Invitation

  • Bring a question from your math or science classrooms.

    Where do your teachers want better support for students’ group reasoning — and what would they want AI to do?

  • Explore a focused design partnership

    District teachers bring knowledge of students, curriculum and classroom routines. Our team brings expertise in learning design and AI. Together, we could define a manageable question, shape a prototype and agree on what evidence would make it worth using.

Research and Contributors

Full Study Credits

This collaboration connects Meredith Kier and Yixuan (Janice) Zhang at William & Mary with colleagues and students. Each paper’s complete author list appears below.

A single column table with a research citation

Designing AI peers with middle school students

Lyu, W., Wang, Y., Yue, M., Sun, Y., Suh, J., Kier, M., Yao, Z., & Zhang, Y. (2026). Designing AI peers for collaborative mathematical problem solving with middle school students: A participatory design study. In Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems. Association for Computing Machinery. Published paper. Read the open manuscript.
A single column table with a research citation

Student engagement during mathematical modeling

Suh, J. M., Yao, Z., Zhang, Y., Yue, M., Lyu, W., Slutz, E., Kier, M., & McGlone, D. (2026). Middle school students’ engagement with AI peers during collaborative mathematical modeling. The Journal of Educational Research in Mathematics, 36(3), 433–459. Read the study and publication record.
We also recognize the participating students, educators, and research collaborators whose perspectives shape this work.