Intelligent Instructional Navigation: Development of a Wearable Sensor-Based Adaptive Learning Model to Improve Self-Efficacy and Aquatic Skills

Abstract

The digital transformation in higher education necessitates a shift from subjective, manual pedagogical evaluations toward objective, data-driven strategies, particularly in complex physical domains like aquatic skills. This study aims to develop and analyze an intelligent instructional navigation model using wearable sensor-based artificial intelligence (AI) to provide real-time biometric feedback and accelerate swimming competencies and student self-efficacy. Employing a quasi-experimental design, the research integrated smartwatches and AI algorithms to monitor biometric metrics and intensity zones (Z1–Z5), providing immediate haptic scaffolding during learning sessions. The results indicate that the AI-driven model significantly enhances evaluation objectivity and motor adaptation speed compared to conventional methods, with the experimental group achieving substantially higher psychomotor scores (8.8 vs. 7.1; p < 0.01). Notably, the findings reveal that training duration without precise intensity zone management does not significantly improve performance, highlighting biometric-based scaffolding as the critical variable for instructional success. This study concludes that transforming instructional frameworks through AI-driven navigation acts as a vital catalyst for achieving Sustainable Development Goals (SDGs), specifically Quality Education and Good Health, by ensuring a measurable, safe, and adaptive learning ecosystem. This model offers a scalable pedagogical framework for modernizing sports education through the scholarship of teaching and learning (SoTL).