Abstract:

In modern STEM education, there is a growing need to introduce innovative and objective approaches to assess cognitive load, engagement, and well-being of students in the learning process. Traditional methods based on self-assessment, surveys, and grades often do not reflect the dynamic changes in the state of learners and limit the possibilities for timely pedagogical adaptation. The present study proposes an interdisciplinary educational-technological approach to assess stress in students in a STEM environment by using wearable IoT devices and monitoring the load of the autonomic nervous system, based on the analysis of heart rate variability (HRV). This approach aligns with the principles of STEAM education, where interdisciplinary integration of science, technology, engineering, arts, and mathematics fosters creativity, critical thinking, and problem-solving skills. By incorporating physiological feedback into the learning process, the proposed framework supports more engaging, reflective, and student-centered learning experiences. The method integrates real-time physiological data collection via a wearable system, followed by the extraction of HRV indicators to study cognitive load and stress. The data is analyzed in the context of various educational activities in STEAM education, including lectures and laboratory exercises, in order to identify variations in the levels of engagement and difficulty of the learning content. Local analysis in short time windows provides the opportunity to track momentary responses to specific learning tasks, which allows for a more precise assessment of teaching effectiveness. The experimental setup includes observation of students during real-life classes, in which physiological signals are synchronized with the time markers of the educational process, which allows for correlation analysis between HRV parameters and specific learning activities, such as solving tasks, learning new material, and working in a laboratory environment. The proposed framework contributes to the field of learning analytics by providing a continuous and objective indicator of the state of students, which can assist teachers in adapting learning content and pedagogical strategies. The results demonstrate the potential of IoT-based technologies to identify the most challenging elements in STEM learning, which creates prerequisites for improving instructional design and increasing learning efficiency. From a well-being perspective, the proposed system allows for early detection of excessive load and the risk of burnout, which is key for sustainable learning and development of students. The main contribution of the study lies in the development of an integrated educational-technological framework that uses wearable IoT devices and HRV-based physiological analysis to objectively assess cognitive load and to identify challenging elements in STEM learning, in order to support adaptive and personalized pedagogical strategies. In conclusion, the integration of wearable IoT technologies and physiological analysis in educational practice represents a promising tool for the development of adaptive, personalized, and learner-centered STEAM educational environments.

Keywords:

STEM education, learning analytics, wearable IoT devices, student stress assessment, student well-being.