A study on the synergistic effects of teacher–student interaction patterns and an AI-based emotional feedback system in secondary school chemistry laboratory classes: an intervention experiment grounded in inquiry-based instruction
Year: 2026
Type: article
Abstract: While prior research has examined how teacher-student interaction influences student learning in conventional chemistry laboratory instruction, few studies have systematically disentangled the synergistic mechanisms that connect dynamic patterns of teacher-student interaction with AI-driven emotional feedback systems-and, critically, how these interactions jointly shape educational outcomes in AI-integrated chemistry laboratory classrooms. This gap constrains the evidence-informed design of emotion-responsive, intelligent chemistry pedagogy. This study employed a mixed-methods design integrating a quasi-experimental intervention with qualitative thematic analysis. A semester-long intervention was implemented with two cohorts of ninth-grade students: an experimental group ( n = 40) receiving inquiry-based chemistry laboratory instruction augmented by an AI-powered emotional feedback system, and a control group ( n = 38) receiving equivalent inquiry-based instruction supported solely by conventional teacher-student interaction. The study comparatively analyzed students' emotional experiences across the two instructional conditions. Distinct emotional trajectories emerged between the groups. The AI-supported experimental group exhibited a statistically significant increase in chemistry laboratory anxiety over the semester, with no meaningful improvement in subjective achievement. In contrast, the control group maintained stable anxiety levels and demonstrated consistent growth in perceived achievement. Moreover, the control group outperformed the experimental group in procedural compliance and the rigor of hands-on laboratory operations. No statistically significant intergroup differences were observed for other affective indicators-including sense of accomplishment and willingness to collaborate. These findings advance the theoretical understanding of human-machine collaborative pedagogy within intelligent learning environments. Empirically grounded, they yield actionable insights for the design of chemistry laboratory instruction in AI-integrated settings-particularly in developing learner-centered, emotion-aware frameworks for human-AI co-teaching.
Source: Frontiers in Education
Authors: Hong-xia Liu, Jia Li, Yuhan Zhang
Institution: Anshan Normal University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,353
OpenAlex ID: W7225808874