Scholarlink

This study explores the neural mechanisms underpinning collaborative creative thinking in engineering design, focusing on comparisons between human-human and human-artificial intelligence (AI) interactions. Fourteen engineering students performed divergent thinking tasks under three experimental conditions: individual effort, collaboration with another human, and collaboration with a generative AI (GPT-4). The results demonstrated that collaboration, regardless of the partner type, significantly enhanced the usefulness of generated ideas compared to individual efforts, though originality scores remained consistent across conditions. Electroencephalography (EEG) analysis revealed distinct patterns of brain activation associated with collaborative processes. Collaboration with a human partner activated memory-related regions such as the temporal lobe and hippocampus, as well as the right parietotemporal junction, a region linked to cooperative behaviors. Notably, collaboration with AI further engaged the right dorsolateral prefrontal cortex, highlighting its role in cognitive flexibility. These findings suggest that AI collaboration fosters unique neural pathways that may facilitate adaptive thinking and problem-solving.

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

Neural Mechanisms of Collaborative Creative Thinking in Engineering Design: Human-Human Versus Human-AI Interaction

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Kazutaka Ueda,
  • Fukashi Mikami,
  • Kosei Sato,
  • Koji Koizumi,
  • Keisuke Nagato,
  • Yukihisa Yokoyama

摘要

This study explores the neural mechanisms underpinning collaborative creative thinking in engineering design, focusing on comparisons between human-human and human-artificial intelligence (AI) interactions. Fourteen engineering students performed divergent thinking tasks under three experimental conditions: individual effort, collaboration with another human, and collaboration with a generative AI (GPT-4). The results demonstrated that collaboration, regardless of the partner type, significantly enhanced the usefulness of generated ideas compared to individual efforts, though originality scores remained consistent across conditions. Electroencephalography (EEG) analysis revealed distinct patterns of brain activation associated with collaborative processes. Collaboration with a human partner activated memory-related regions such as the temporal lobe and hippocampus, as well as the right parietotemporal junction, a region linked to cooperative behaviors. Notably, collaboration with AI further engaged the right dorsolateral prefrontal cortex, highlighting its role in cognitive flexibility. These findings suggest that AI collaboration fosters unique neural pathways that may facilitate adaptive thinking and problem-solving.