Labrado Antolín, María Isabel2025-12-052025-12-052026-01-01Labrado, M. (2026). Talking to machines: How communication style shapes student engagement with AI tutors. American Journal of STEM Education, 19, 37-58. https://doi.org/10.32674/t2qnzc903069-007210.32674/t2qnzc90https://hdl.handle.net/20.500.14352/128499As artificial intelligence (AI) chatbots become integral to higher education, this qualitative study explores how undergraduate students interact with them during business strategy tasks. Grounded in the Value-Based Adoption Model and utilizing ATLAS.ti for content and co-occurrence analysis, this study analyzes emotional tone and cognitive strategies in 15 student–AI conversations. Students who used a relational tone and followed up with questions demonstrated deeper critical thinking, whereas those who employed neutral tones and passive inquiries showed lower engagement. Co-occurrence analysis highlighted key patterns, such as neutral tone and simple inquiries. Findings suggest that socio-affective alignment in human–AI interaction fosters higher-order thinking, providing pedagogical insights into how AI integration can enhance both cognitive depth and emotional engagement in learning environments.engjournal article3065-1190https://doi.org/10.32674/t2qnzc90https://mail.ojed.org/STEM/article/view/9124open access378.4004.837.01I21I23AI literacyHuman-AI interactionQualitative content analysisAtlas.tiHigher educationVirtual tutorSTEM in EducationEnseñanza de las Ciencias SocialesInteligencia artificial (Informática)Enseñanza universitaria1203.04 Inteligencia Artificial