Understanding influence of students’ use of generative AI (GenAI) on their learning outcomes is one of the critical issues in the current education sphere, whereas impacts on group-level learning outcomes in creative collaborative problem solving contexts are unexplored. We present one of the first studies to preliminarily investigate how university students’ GenAI use in collaborative learning settings related to their knowledge creation quality as group-level learning outcomes, where students worked toward designing future classroom. We found that 1) students used GenAI much more to visually embody their ideas than to find more innovative ideas beyond their own ideas, 2) frequent use of GenAI were linked with middle-level learning outcomes rather than high-level learning outcomes. These findings suggests that while GenAI can support collaborative problem solving moderately, higher learning outcomes require effective integration of human creativity with AI-generated content. Further study is required to identify how high-outcome groups collaboratively developed their ideas using GenAI and what support is effective for students’ more constructive use of GenAI for knowledge creation.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Exploring Students’ Generative AI Usage Patterns and Knowledge Creation in Collaborative Problem Solving

  • Shotaro Naganuma,
  • Tsubasa Minematsu,
  • Sachika Shibukawa,
  • Aikana Ohno,
  • Yukinori Wakihama

摘要

Understanding influence of students’ use of generative AI (GenAI) on their learning outcomes is one of the critical issues in the current education sphere, whereas impacts on group-level learning outcomes in creative collaborative problem solving contexts are unexplored. We present one of the first studies to preliminarily investigate how university students’ GenAI use in collaborative learning settings related to their knowledge creation quality as group-level learning outcomes, where students worked toward designing future classroom. We found that 1) students used GenAI much more to visually embody their ideas than to find more innovative ideas beyond their own ideas, 2) frequent use of GenAI were linked with middle-level learning outcomes rather than high-level learning outcomes. These findings suggests that while GenAI can support collaborative problem solving moderately, higher learning outcomes require effective integration of human creativity with AI-generated content. Further study is required to identify how high-outcome groups collaboratively developed their ideas using GenAI and what support is effective for students’ more constructive use of GenAI for knowledge creation.