The advancement of brain–computer interfaces (BCI) and body sensors in wearable devices has opened new possibilities for the design of personalized educational experiences. This study presents a proposal for integrating neurophysiological data—such as heart rate, heart rate variability, and electrodermal activity—with artificial intelligence algorithms to improve attention, motivation, and academic performance in primary school students in a semi-urban context in Mexico. The research was conducted with a sixth-grade group (21 students) in a public school, using a quasi-experimental intervention design. Each student used low-cost smartbands to record physiological variables during school sessions. These data were analyzed using machine learning models and supported by an AI system (ChatGPT, in its free version), which generated feedback for the teacher and adaptive recommendations regarding pace, type of activity, and level of cognitive load. The results show that feedback based on biometric data contributes to increased on-task behavior, reduced cognitive fatigue, decreased stress levels, and improvements in reading comprehension and mathematical problem solving. In addition, both students and the teacher reported positive perceptions of the intervention, highlighting the motivating and humanized nature of didactic personalization. This work provides evidence of the pedagogical, technical, and ethical feasibility of integrating accessible neurotechnologies with artificial intelligence in school environments with limited resources and proposes a replicable model for similar educational contexts.

Authors:
José Luis Beltrán Sánchez, CAM Michoacán, Mexico

About the Presenter(s)
José Luis Beltrán Sánchez is a public elementary school teacher in Michoacán, Mexico. He is also a professor in the Bachelor’s Degree in Basic Education program at the Universities for Wellbeing Benito Juárez, and he is currently pursuing a second do

See this presentation on the full schedule – Sunday Schedule

Bookmark