Artificial intelligence and empathetic communication to enhance learning engagement, motivation, and effectiveness

Abstract With advances in artificial intelligence (AI), textual sentiment analysis has become increasingly popular for extracting word relationships and uncovering hidden emotions. This study applied sentiment analysis to explore students’ emotions during learning because emotions significantly impact motivation and outcomes. Effective teacher-student communication fosters understanding and enhances learning but requires strong empathetic skills to recognize and address students’ emotions. To support this, we developed an AI-assisted learning system integrating KeyMoji, ChatGPT, and Unity that analyzes emotions in real time and provides empathetic responses. The results showed that learning emotions were associated with teacher-student interactions and self-reported academic performance. Students who reported receiving empathetic responses also reported higher engagement and perceived learning outcomes, although these relationships cannot be interpreted as causal given the study design. This study provides preliminary evidence that AI sentiment analysis and empathetic communication are statistically associated with students’ learning emotions, motivation, engagement, and perceived learning outcomes. These findings suggest that AI-generated empathetic feedback supports effective motivational learning processes. The post-test findings indicate that AI-generated empathetic feedback is statistically associated with perceived learning outcomes indirectly through learning emotions and motivation, although these relationships should be interpreted as associational rather than causal, given the post-test single-group research design. Consequently, this study provides a transferable blueprint that advances both affective computing theory and practical digital pedagogy.