🤝
 Collaborators: Daniel M. Oppenheimer; Allison E. Connell PenskyÂ
Overview
Since their inception, educators have pondered what Large Language Model (LLM) chatbots - such as Chat GPT and Gemini - mean for student learning. On one hand, they provide an incredibly hard-to-detect tool that students can use for cheating and plagiarism. But on the other hand, LLMs present endless opportunities to enhance student learning in previously unthinkable ways. In this project, we test a novel intervention in which LLMs are used to enhance a classic classroom technique - collaborative learning - with the goal of developing students' argumentative writing skills. More specifically, this intervention challenges students to write an argumentative essay, debate the merit of their essay with an LLM, then re-write their essay using the feedback provided by the LLM. By testing this intervention, we seek to answer questions such as:
- Can students think of LLMs as intelligent collaborators, rather than just tools?
- Does student-LLM collaboration have the same learning benefits as student-student collaboration?
- Do students apply the concepts they learn from collaborating with LLMs, even in future tasks that do not involve LLMs?
- Does collaborating with LLMs improve students' knowledge of how to use LLMs to augment their future learning and productivity?
Publications
Oppenheimer, D. M., Cash, T. N., & Connell Pensky, A. E. (Under Review). You’ve Got AI Friend in Me: LLMs as Collaborative Learning Partners. Preprint: https://doi.org/10.31219/osf.io/8q67uÂ
