Artificial Intelligence in Digital Learning Design for Personalized and Adaptive Learning: A PRISMA-Based Systematic Literature Review

Abstract

Artificial intelligence (AI) now shapes how learning experiences are designed, but the research linking digital and instructional learning design with AI, learning analytics, and personalized or adaptive learning is scattered across separate fields. This paper reports a systematic literature review (SLR) carried out under the PRISMA 2020 protocol. We examined open-access, English-language documents indexed in Scopus that together address learning design, AI or analytics, personalization or adaptivity, and an explicit framework, model, approach, or strategy. Of 86 records identified, one duplicate was removed, twelve were excluded during title and abstract screening, and ten during full-text assessment, which left 63 studies for synthesis. Thematic coding grouped these into six clusters: generative AI and large language models for instructional design (C1); adaptive and personalized learning systems and frameworks (C2); learning analytics for learning design and feedback (C3); cognitive and self-regulated learning with AI (C4); immersive and emerging technologies (C5); and reviews and meta-research (C6). The review answers two questions: which AI-based frameworks, models, approaches, and strategies are used to design digital learning (RQ1), and how those designs deliver personalization and adaptivity (RQ2). We consolidate the evidence into a conceptual AI-enhanced Digital Learning Design (AI-DLD) framework and discuss the open problems of equity, ethics, data privacy, and human oversight.