Towards a Psychologically Grounded Framework for Ethical, Inclusive, and AI-Enhanced Education
TL;DR
A conceptual framework offering AI designers, teacher educators, and policymakers a psychologically grounded, empirically mapped, and internationally contextualised basis for developing AI-supported educational systems that advance equity, learner engagement, and meaningful educational transformation is proposed.
Abstract
Background: The rapid integration of artificial intelligence (AI) into educational systems worldwide presents significant opportunities for personalised learning, inclusive pedagogy, and data-informed instructional practice, yet AI systems designed without an understanding of how learners acquire knowledge, sustain motivation, regulate emotions, and respond to diverse social and cultural contexts may unintentionally reinforce existing educational inequalities. Objective: This paper proposes a conceptual framework that positions educational psychology as the primary foundation, rather than a secondary consideration, for the design and evaluation of AI-supported educational systems. Methods: The framework was developed through a systematic narrative synthesis of peer-reviewed literature, international policy documents, and recent empirical evidence, drawing on searches of PsycINFO, ERIC, Scopus, and Web of Science (2011–2025), supplemented by forward and backward citation tracking and review of international policy reports. The search yielded approximately 620 potentially relevant sources, of which approximately 65 were incorporated following title/abstract screening and full-text review against explicit inclusion criteria. Results: The resulting framework comprises three interconnected components — a psychologically informed understanding of learner diversity, AI-enabled inclusive and innovative pedagogical practices, and reflective teaching within intelligent learning environments — mapped, through a reference table, against specific AI applications, illustrative empirical evidence, and inclusion implications. A comparative analysis across East Asia, Europe, Sub-Saharan Africa, and South Asia further identifies region-specific challenges, policy contexts, and persistent gaps in cross-cultural validation, particularly regarding generative AI and adaptive learning tools. Conclusion: The framework offers AI designers, teacher educators, and policymakers a psychologically grounded, empirically mapped, and internationally contextualised basis for developing AI-supported educational systems that advance equity, learner engagement, and meaningful educational transformation, provided that access, teacher readiness, cultural validity, and ethical safeguards are adequately addressed.