Scholarlink

<p>This study addresses persistent pedagogical challenges in inorganic chemistry—such as abstract concept visualization, imprecise learning diagnosis, and underdeveloped critical thinking—by proposing a triadic teaching model, “Perception-Diagnosis-Critical Thinking,” powered by generative artificial intelligence (GAI). The model employs GAI-driven interactive simulations to visualize abstract concepts, utilizes dynamic knowledge graphs for real-time learning diagnosis and personalized pathways, and designs comparative and critical tasks to foster higher-order thinking in authentic engineering contexts. A quasi-experimental study involving 75 first-year students (experimental group: <i>n</i> = 37; control group: <i>n</i> = 38) demonstrated that the model significantly improved students’ knowledge mastery (*<i>t</i>* = 6.70, *<i>p</i>* &lt; 0.0001), enhanced specific dimensions of complex problem-solving (*<i>t</i>* = 2.055, *<i>p</i>* = 0.044), and elicited more positive learning experiences (*<i>t</i>* = 2.121, *<i>p</i>* = 0.038). This research offers a transferable paradigm for the deep integration of AI in STEM education, positioning GAI not merely as a tool but as an architectural component of the instructional cycle.</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

From Tool to Architect: Constructing a Generative AI-Empowered “Perception–Diagnosis–Critical Thinking” Triadic Teaching Model in Inorganic Chemistry

  • 期刊论文

  • 发表日期：

    2026年5月9日

查看全文
  • Ling Li

摘要

This study addresses persistent pedagogical challenges in inorganic chemistry—such as abstract concept visualization, imprecise learning diagnosis, and underdeveloped critical thinking—by proposing a triadic teaching model, “Perception-Diagnosis-Critical Thinking,” powered by generative artificial intelligence (GAI). The model employs GAI-driven interactive simulations to visualize abstract concepts, utilizes dynamic knowledge graphs for real-time learning diagnosis and personalized pathways, and designs comparative and critical tasks to foster higher-order thinking in authentic engineering contexts. A quasi-experimental study involving 75 first-year students (experimental group: n = 37; control group: n = 38) demonstrated that the model significantly improved students’ knowledge mastery (*t* = 6.70, *p* < 0.0001), enhanced specific dimensions of complex problem-solving (*t* = 2.055, *p* = 0.044), and elicited more positive learning experiences (*t* = 2.121, *p* = 0.038). This research offers a transferable paradigm for the deep integration of AI in STEM education, positioning GAI not merely as a tool but as an architectural component of the instructional cycle.