Unpacking the relationships among key factors in the effectiveness of inquiry-based learning: What changes does ChatGPT bring?

Previous studies on inquiry-based learning have seldom examined the structural relationships among curiosity, engagement, inquiry skills, and cognitive outcomes in generative AI environments. To address this gap, this study utilized ANCOVA and PLS-SEM with 462 Taiwanese high school students to compare traditional and ChatGPT-assisted IBL. Results indicated that while ChatGPT-assisted IBL significantly enhanced problem-solving skills, its impact on critical thinking and creativity was limited, as automated responses reduced opportunities for skepticism and independent verification. Furthermore, mediation analysis revealed that curiosity did not directly drive knowledge construction; its effects were fully mediated by learning engagement, underscoring engagement’s critical bridging role. Ultimately, this study clarifies how generative AI reshapes students’ learning pathways across distinct instructional settings, offering valuable empirical benchmarks for designing effective AI-supported learning scaffolds.