Effectiveness Effectiveness of Intelligent Tutoring Systems, Adaptive Learning, and Artificial Intelligence-Based Technologies in Mathematics Learning: An Umbrella Review of Systematic Reviews and Meta-Analyses

Abstract

This umbrella review synthesizes evidence from systematic reviews and meta-analyses on the effectiveness of intelligent tutoring systems, adaptive learning platforms, and artificial intelligence-based technologies in mathematics learning. A semi-automated search was conducted on 18 December 2025 using an indexed academic search engine, complemented by forward-backward citation chasing. Eligible reviews were screened and appraised using a proxy version of AMSTAR 2 focused on critical domains. Across the included reviews, technology-enhanced and adaptive interventions generally showed small-to-moderate positive effects on mathematics achievement, with more consistent benefits when feedback was immediate, personalization was meaningful, and implementation was supported by teachers and well-aligned tasks. Evidence for non-cognitive outcomes (e.g., motivation, self-efficacy, engagement) was positive but less consistently reported. Heterogeneity across contexts and substantial overlap among primary studies remain major challenges. The review highlights implementation conditions, recurring moderators, and priority gaps to strengthen future research and reporting in mathematics education.