Scholarlink

<p>Static curricula frequently fail to accommodate individual differences in learning pace and style, creating a critical gap in technology-enhanced learning environments. This study investigates the impact of a Generative AI-driven adaptive curriculum on student performance and engagement. The study was conducted with 270 students enrolled in an “Innovation and Digital Technology for Learning” course at Khon Kaen University, divided into two groups: one following a static curriculum and the other using the AI-generated dynamic curriculum and randomly assigned via stratified randomization to a static curriculum (control group, <i>n</i> = 135) or a Generative AI-driven adaptive curriculum (experimental group, <i>n</i> = 135). The AI-driven system adapted learning materials in real-time. The experimental group demonstrated significantly superior outcomes: post-test scores improved by 24.2% versus 10.2% in the control group (<i>t</i>(268) = 8.43, <i>p</i> &lt; .001, Cohen’s <i>d</i> = 1.03), with higher engagement intensity (9.1 vs. 5.6&#xa0;h/week), lower cognitive load on the NASA-TLX (<i>M</i> = 38.4 vs. 52.1), and stronger complex problem-solving performance (85.6% vs. 68.2%). The System Usability Scale (SUS) yielded a score of 92.5 (SD = 6.8), indicating excellent usability. These findings suggest that Generative AI-driven adaptive curricula can meaningfully improve student outcomes in technology-enhanced learning contexts, though replication across diverse settings is needed. Future studies should investigate the scalability of AI-driven curricula across disciplines and longer-term educational outcomes.</p>

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

Generative AI for adaptive curriculum design enhances student performance and engagement in technology enhanced learning environments

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月30日

查看全文
  • Charuni Samat,
  • Parama Kwangmuang,
  • Pornsawan Vongtathum,
  • Issara Kanjug,
  • Tharit Wannasen,
  • Pornpisut Duangngern,
  • Nantapoom Gessala,
  • Paritchaya Sarakan,
  • Rapeepan Pitakaso,
  • Ting Ting Wu

摘要

Static curricula frequently fail to accommodate individual differences in learning pace and style, creating a critical gap in technology-enhanced learning environments. This study investigates the impact of a Generative AI-driven adaptive curriculum on student performance and engagement. The study was conducted with 270 students enrolled in an “Innovation and Digital Technology for Learning” course at Khon Kaen University, divided into two groups: one following a static curriculum and the other using the AI-generated dynamic curriculum and randomly assigned via stratified randomization to a static curriculum (control group, n = 135) or a Generative AI-driven adaptive curriculum (experimental group, n = 135). The AI-driven system adapted learning materials in real-time. The experimental group demonstrated significantly superior outcomes: post-test scores improved by 24.2% versus 10.2% in the control group (t(268) = 8.43, p < .001, Cohen’s d = 1.03), with higher engagement intensity (9.1 vs. 5.6 h/week), lower cognitive load on the NASA-TLX (M = 38.4 vs. 52.1), and stronger complex problem-solving performance (85.6% vs. 68.2%). The System Usability Scale (SUS) yielded a score of 92.5 (SD = 6.8), indicating excellent usability. These findings suggest that Generative AI-driven adaptive curricula can meaningfully improve student outcomes in technology-enhanced learning contexts, though replication across diverse settings is needed. Future studies should investigate the scalability of AI-driven curricula across disciplines and longer-term educational outcomes.