Design and Evaluation of an AI-Supported Global Shared Learning Strategy in Engineering Science Courses

Abstract

This paper analyses a Global Shared Learning Classroom (GSL) experience developed between undergraduate students from the Dominican Republic and Puebla, Mexico, focused on collaborative research into the relevance of different vitamins for human health. The study was implemented as a pilot quasi-experimental design, integrating international collaborative learning, digital tools, and regulated artificial intelligence support as part of a structured pedagogical architecture. The intervention was articulated through the Dynamic Table for Research and Dissemination (TABDIM), used to organize, systematize, and synthesize the scientific information generated by binational teams. Students analyzed various vitamins considering physicochemical properties, biochemical functions, health implications, food sources, and practical applications. To support the research process, the specialized chatbot Albiomim was integrated as a cognitive scaffolding tool for consultation and conceptual clarification. The outcomes of the collaborative work were presented through a dynamic digital poster within an immersive virtual environment accessible via avatars, complemented by a reflective phase supported by the Padlet platform. Quantitative analysis of the results revealed a significant increase in conceptual learning, with mean scores rising from 65.68 to 78.18, along with a strong positive correlation between pre- and post-intervention results (r = 0.8006). Overall, the findings suggest that the intentional and structured integration of international collaboration, experiential learning, digital tools, and artificial intelligence contributes positively to conceptual learning, student engagement, and the overall learning experience in science courses for engineering education, providing relevant evidence for the design of innovative learning environments in higher education.