PUW! Cognitive strength in context: An innovative, multi-modal framework for real-time analysis of stress and...

Student well-being is a multidimensional construct encompassing physical and mental health. However, its assessment in educational contexts remains fragmented, limiting understanding of the diverse factors that shape students' experiences. This gap is particularly pronounced for cognitively strong students (CSS), whose unique intellectual, emotional, and social challenges are frequently overlooked by traditional approaches.

Current work introduces a methodology to develop a holistic understanding of student well-being by integrating multimodal data, uncovering key patterns that influence well-being and academic performance, and offering actionable insights for targeted interventions.

The framework, developed through an extensive literature review and refined using the Delphi method, integrates three core components: physiological measures (e.g., heart rate variability and sleep data) to detect stress; self-reporting tools for qualitative insights; and surveys to capture psychological stressors and socio-economic influences. While broadly applicable, it emphasizes CSS-specific stressors.

Qualitative and quantitative data are analyzed using large language models for sentiment analyses and classical machine learning techniques analyze physiological and survey data integration. This approach aims to identify patterns in stress and coping mechanisms, offering insights into broader dimensions of well-being.

This research provides a scalable, ethically grounded methodology for AI-supported assessment of student well-being, supporting interventions that promote healthy development, particularly for CSS.