AI-Supported Emotional Well-Being in Learning: A Review of Empirical Studies

• In Event: Virtual Poster Session
  In Virtual iPresentation Only: SIG-Advanced Technologies for Learning Virtual Poster Session

Abstract

As AI becomes more integrated into education, many emotional well-being challenges emerged. Most research emphasizes cognitive outcomes, with limited research and theoretical grounding to explain how AI shapes learners’ emotional experiences.This scoping review addresses the gap by examining 26 empirical studies on AI tools designed to support emotional well-being. Guided by an integrative framework, we map AI features to three emotional domains: achievement emotions, social emotions, and emotion regulation. Following the PRISMA guidelines, we systematically searched ERIC, Scopus, and Web of Science. Findings show AI can reduce anxiety and stress, boost motivation, but offer limited support for emotion regulation, while social-emotional development remains understudied. The review calls for emotionally responsive AI design and broader affective research in education.

Authors

• Run Wen, Xi'an Jiaotong-Liverpool University

• Na Li, Xi'an Jiaotong-Liverpool University

• Manolis Mavrikis, University College London