Scholarlink

<p>The advent of generative artificial intelligence (Gen-AI) has introduced transformative possibilities across diverse domains, prompting growing interest in its implications for education and cognitive development. Some researchers have begun examining the impact of Gen-AI, particularly on higher-order thinking (HOT), including critical thinking, creativity, problem-solving, and computational thinking. Concerns exist in current discourse that Gen-AI may harm HOT. However, empirical research on the specific effects of Gen-AI on HOT remains scattered, and no consensus has been reached. This study synthesized 19 experimental and quasi-experimental studies conducted in different contexts (<i>k</i> = 68), involving 2,347 participants. A three-level meta-analysis was employed to account for within- and between-study variability, assessing the impact of Gen-AI on HOT and exploring the effects of moderators. The results revealed that Gen-AI had a significant positive effect on enhancing students’ HOT (Hedges’s <i>g</i> = 0.851, <i>p</i> &lt; .001, 95% CI [0.452, 1.250]). Moderator analyses were conducted based on impact target, Gen-AI elements, study contexts, and methodological characteristics. When the sample size was less than 80, the promotion effect was more significant. Both short-term interventions (less than 4&#xa0;weeks) and long-term interventions (more than 8&#xa0;weeks) have the potential to produce significant positive effects. The meta-analysis results support the view of Gen-AI as a powerful tool for promoting students’ HOT development, providing data-driven evidence for future educational practices and policymaking.</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

The impact of generative artificial intelligence on students’ higher order thinking: Evidence from a three-level meta-analysis

  • 期刊论文

  • 发表日期：

    2025年9月20日

查看全文
  • Xinxiao Nie,
  • Yuan Tian,
  • Mengjie Liu,
  • Di Wu,
  • Yunxiao Guo

摘要

The advent of generative artificial intelligence (Gen-AI) has introduced transformative possibilities across diverse domains, prompting growing interest in its implications for education and cognitive development. Some researchers have begun examining the impact of Gen-AI, particularly on higher-order thinking (HOT), including critical thinking, creativity, problem-solving, and computational thinking. Concerns exist in current discourse that Gen-AI may harm HOT. However, empirical research on the specific effects of Gen-AI on HOT remains scattered, and no consensus has been reached. This study synthesized 19 experimental and quasi-experimental studies conducted in different contexts (k = 68), involving 2,347 participants. A three-level meta-analysis was employed to account for within- and between-study variability, assessing the impact of Gen-AI on HOT and exploring the effects of moderators. The results revealed that Gen-AI had a significant positive effect on enhancing students’ HOT (Hedges’s g = 0.851, p < .001, 95% CI [0.452, 1.250]). Moderator analyses were conducted based on impact target, Gen-AI elements, study contexts, and methodological characteristics. When the sample size was less than 80, the promotion effect was more significant. Both short-term interventions (less than 4 weeks) and long-term interventions (more than 8 weeks) have the potential to produce significant positive effects. The meta-analysis results support the view of Gen-AI as a powerful tool for promoting students’ HOT development, providing data-driven evidence for future educational practices and policymaking.