随着具身认知理论理论的深入发展，其研究逐渐聚焦于具身交互和情境学习。在科学技术飞速发展的背景下，儿童教育环境不断更新完善，作为教育体系关键一环的识字教育也在不断创新升级，以顺应时代的需求。尽管已有诸多研究探讨了基于具身认知理论的语言学习策略，但将算法技术引入其中的设计实践仍较为稀缺。由此，本研究旨在探索如何将识字教育与算法技术相结合，并通过多模态教学方法提升儿童的识字兴趣。
本研究采用文献分析与定量研究方法，系统梳理了具身认知相关理论，并对相关研究热点进行可视化分析。为解决儿童在真实情境中学习汉字的技术难题，本研究基于增强现实技术将现实与虚拟信息叠加，设计出符合儿童认知特点的汉字具身教学系统。为鼓励儿童在情境中学习汉字，该系统整合了80种生活中常见物体的数据集，设置了交通、动物、食物和日常生活等四种识字情境。基于具身认知理论的指导，确保儿童在学习汉字过程中能够充分调动感官和肢体，提升学习效果。在实践中，还通过YOLOv5算法实现增强现实效果，为该系统构建了图片识别、视频识别和摄像头识别三种汉字识别模式。
在设计实践完成后，研究邀请了27名儿童参与该系统的可用性测试，使用SUS量表对上述三种识别方式赋分，以评估系统的易用性和可用性，并通过单因素方差分析检验识别功能对系统可用性的影响。实验结果显示，汉字具身教学系统具有良好的易用性和可用性，且儿童在使用过程中积极参与。
本研究提出，儿童汉字教学系统的设计应遵循具身性、直观性、符合儿童认知特征和趣味性等原则。基于增强现实、具身体验和真实情境参与的思路构建的汉字具身教学系统具有一定的理论创新价值。在实践层面上，本研究丰富了儿童具身学习汉字的交互设计方法，展现了较高的实践创新意义。研究结果为当代儿童汉字互动教学提供了技术参考与经验总结，并对该领域的创新设计具有一定的启发性。

With the development of the embodied cognition theory, research has increasingly focused on embodied interactions and contextual learning. Against the backdrop of rapid technological advancement, children’s educational environments are constantly being updated, with literacy education—a critical component of the educational system—also evolving to meet contemporary demands. Although numerous studies have explored language learning strategies grounded in embodied cognition theory, design practices that integrate algorithmic technology remain scarce. This study aims to explore how literacy education can be combined with algorithmic technology to enhance children’s interest in literacy through multimodal teaching methods.
This research employs literature analysis and quantitative research methods to systematically review relevant theories of embodied cognition and visualize key research topics. To address the technical challenges of children learning Chinese characters in real contexts, this study leverages augmented reality (AR) technology to overlay real and virtual information, developing an embodied teaching system for Chinese characters that aligns with children's cognitive characteristics. To encourage children to learn characters within meaningful contexts, the system incorporates a dataset of 80 commonly encountered objects and creates four distinct literacy scenarios: transportation, animals, food, and daily life. Under the guidance of embodied cognition theory, the system enables children to engage their senses and physical movements in learning Chinese characters, thereby enhancing the learning experience. Practically, the study uses the YOLOv5 algorithm to implement AR effects, establishing three character recognition modes—image, video, and camera-based.
Upon completion of the design, 27 children participated in usability testing of the system. Scores for each recognition mode were collected using the SUS scale to evaluate ease of use and overall usability, followed by a one-way ANOVA to examine the effect of recognition functionality on system usability. The findings indicate that the embodied Chinese character teaching system demonstrates high usability, with children showing active engagement throughout.
This study proposes that children’s Chinese character teaching systems should adhere to principles of embodiment, intuitiveness, cognitive alignment, and enjoyment. Constructed on the principles of augmented reality, embodied experience, and real-world contextual involvement, this system provides theoretical and practical innovation. The findings enrich the interactive design methods for embodied learning in children’s literacy education and offer technological references and insights for contemporary interactive Chinese character teaching.