Deep Learning as a New Paradigm in Islamic Religious Education in Madrasahs: A Literature Review

Abstract

This study addresses the persistent challenge in Islamic Religious Education (PAI) in madrasahs, which often emphasizes rote memorization rather than meaningful understanding and value internalization. The study aims to examine deep learning as an emerging pedagogical paradigm and explore its implications for PAI in madrasahs. This research employed a qualitative systematic literature review approach by analyzing 32 relevant national and international scholarly publications on deep learning and Islamic education. The literature review and selection process were conducted using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) approach, including identification, screening, eligibility, and inclusion stages, followed by thematic and conceptual synthesis. The findings indicate that deep learning, characterized by mindful, meaningful, and joyful learning, is strongly aligned with the principles of Islamic education, particularly in promoting reflective thinking (tafakkur), deep understanding (tadabbur), and character formation. The study also highlights important implications for curriculum design, teaching strategies, teacher roles, and authentic assessment practices in madrasahs. This article concludes that deep learning provides a transformative framework for strengthening the cognitive, affective, and spiritual dimensions of PAI learning. The novelty of this study lies in its conceptual integration of deep learning theory with Islamic educational principles as a foundation for transforming PAI learning in madrasah contexts.