Culture FUSION-AI: Integrating Adaptive Artificial Intelligence and Cultural Pedagogy for Inclusive Elementary Science Education

Abstract

This study aimed to examine the effectiveness of the Culture FUSION-AI model in improving elementary students’ science learning processes and learning achievement on the topic of the human reproductive system. The study was motivated by the need for innovative, inclusive, culturally responsive, and technology-enhanced learning models capable of accommodating diverse student characteristics while maintaining local cultural values. A pre-experimental one-group pretest-posttest design was employed involving 30 fifth-grade students. The Culture FUSION-AI model integrates Acehnese cultural and Islamic values, student-centered inquiry, adaptive Artificial Intelligence (AI), collaborative learning, and scientific character development. Data were collected through learning achievement tests, observation sheets, questionnaires, interviews, and documentation. Quantitative data were analyzed using descriptive statistics, paired-samples t-tests, and normalized gain (N-Gain) analysis, while qualitative data were analyzed thematically. The results showed that students’ mean scores improves, with a gain score of 20.7 and an N-Gain value of 0.64, indicating moderate-to-high learning improvement. Observation results also demonstrated high levels of active participation, learning motivation, collaboration skills, critical attitude, and scientific attitude during implementation. Statistical analysis revealed a significant improvement in learning outcomes. These findings suggest that the Culture FUSION-AI model has the potential to support meaningful, inclusive, and differentiated science learning by integrating adaptive AI technology with local cultural values. The model may serve as a promising approach for enhancing elementary science education while maintaining cultural relevance and supporting diverse learner needs.