Multimodal Learning Analytics as A Feedback Analytics Tool: AI-Driven Real-Time Feedback Through Eye Tracking and Facial Expression Analysis
Advances in Analytics for Learning and Teaching, Springer International Publishing Ag, ss.103-155, 2026
- Yayın Türü: Kitapta Bölüm / Araştırma Kitabı
- Basım Tarihi: 2026
- Doi Numarası: 10.1007/978-3-032-31540-3_5
- Yayınevi: Springer International Publishing Ag
- Sayfa Sayıları: ss.103-155
- Anahtar Kelimeler: Adaptive feedback, Cognitive and affective state inference, Eye tracking, Facial expression analysis, Feedback analytics, Multimodal learning analytics, Real-time feedback
- Atatürk Üniversitesi Adresli: Evet
Özet
Feedback analytics the systematic examination of how feedback is produced, interpreted, and enacted in educational contexts has emerged as a critical extension of both feedback research and learning analytics. This chapter positions Multimodal Learning Analytics (MMLA) as a powerful analytical instrument within the feedback analytics framework, demonstrating how the integration of cognitive, affective, and behavioral data streams can enhance the generation, timing, and optimization of feedback processes. Drawing on eye-tracking data, facial expression analysis, physiological signals, and digital interaction logs, MMLA enables fine-grained, real-time detection of learners’ moment-to-moment states, thereby supporting adaptive and personalized feedback delivery. The chapter presents the theoretical foundations linking MMLA to feedback analytics, discusses multimodal data collection and processing strategies, examines the architectural components of real-time feedback systems, and addresses ethical considerations surrounding affective and cognitive data use. Application scenarios across face-to-face, online, blended, and simulation-based learning environments are explored with particular attention to how MMLA-based feedback can foster learner autonomy, self-regulation, and instructional adaptation. The chapter concludes by identifying future research directions for feedback analytics, including explainable AI architectures, culturally responsive modeling, and scalable feedback system design.