Enhancing Interdisciplinary Learning and Innovative Practice in Students through Mixed Reality: A Deep Learning Approach

Abstract

In the rapidly evolving landscape of information technology, mixed reality (MR) technology has been increasingly recognized as a crucial instrument for educational innovation. This technology, which amalgamates elements of virtual reality (VR) and augmented reality (AR), fosters an environment characterized by enhanced interactivity and profound immersion. Such an environment has been instrumental in improving both the efficiency and quality of learning, especially in cultivating students’ innovative practice capabilities and facilitating interdisciplinary learning. Despite significant advancements in the application of MR in educational contexts, challenges persist in the precise extraction of task performance and behavioral characteristics of students within MR environments. Furthermore, the integration of interdisciplinary information for effective prediction of learning outcomes remains a complex undertaking. This study conducts a systematic analysis of MR technology’s current applications in education, with a focus on strategies that leverage MR technology to support student innovation practices and interdisciplinary learning. Shortcomings in existing research related to the extraction of task performance and behavioral characteristics are identified, and the limitations of conventional predictive models in managing the integration of interdisciplinary information are discussed. To address these challenges, a model based on deep learning for data fusion is proposed. This model is complemented by an end-to-end training approach for task-behavior correlation prediction, with the aim of enhancing both the accuracy of predictions and their practical applicability in educational settings.