;Traditional virtual reality (VR) and augmented reality (AR) learning systems, while helpful in creating contextualized learning environments and transforming abstract theoretical knowledge into practical applications, are mostly pre-designed and controlled by teachers or system developers. This makes students feel disconnected from the learning process, as they can only passively experience and receive knowledge without opportunities for active exploration. This contradicts Papert′s constructivist learning theory. In contrast, personalized learning environments emphasize that learning content should reflect students′ interests and preferences and adapt to their life experiences, thereby promoting the achievement of desired learning goals. However, most current applications of digital reality in education fail to achieve this, especially regarding digital self-representation. The creation of digital avatars usually follows fixed templates and limited options, preventing students from fully expressing themselves and establishing psychological connections in virtual environments. This reduces their sense of self-efficacy and engagement in the learning process. Therefore, this study designed a user-customized digital contextual learning system that allows students to flexibly expand learning content and freely arrange scenes in the digital reality based on their own life experiences and interests. Students can use highly personalized digital avatars to learn in the digital reality they design, showcasing their learning performance to peers and teachers. The experimental results show that this learning mode effectively enhances students′ psychological ownership of the learning environment and positively impacts their learning motivation and outcomes.
