The quasi-experimental design of the study

The current study aimed to compare the impact of using the Canva platform and generative AI tools (ChatGPT → DALL·E/Sora) on developing history student-teachers' ability to simplify historical concepts and the quality of their visual outputs, while also exploring the role of digital engagement as an explanatory variable. A quasi-experimental pre-post design was used with two parallel groups (n = 10 each), relying on a standardized test to measure concept simplification, a visual output quality checklist, and a digital engagement scale, in addition to semi-structured interviews. The results showed that the Canva group outperformed in visual symbol production and overall simplification ability, though not statistically significant. In contrast, generative tools significantly excelled in design quality, creativity, coherence, total score, and digital engagement. No direct relationship was found between simplification ability and output quality, and digital engagement did not moderate the outcomes. The study recommends integrating interactive design and generative tools into university history curricula, alongside developing digital assessment tools.

本研究旨在对比Canva平台与生成式AI (Generative AI)工具（ChatGPT→DALL·E/Sora）对历史专业师范生简化历史概念能力及视觉产出质量的影响，同时探究数字投入作为解释变量的作用。本研究采用准实验前后测设计，设置两个平行组（每组n=10），除半结构化访谈外，还采用标准化概念简化能力测试、视觉产出质量检查表与数字投入量表开展数据采集。结果显示，Canva组在视觉符号创作与整体简化能力上表现更优，但未达到统计学显著性差异。相较而言，生成式AI工具在设计质量、创意性、连贯性、总分及数字投入维度上均显著更优。研究未发现简化能力与产出质量间存在直接关联，且数字投入未对实验结果产生调节作用。本研究建议将交互式设计工具与生成式AI工具融入高校历史课程体系，并同步研发数字化评估工具。