Exploring Gender Differences in Cognitive Load and Stress Responses: A Machine Learning Approach using Physiological Data in Learning Environments

Abstract The rapid integration of technology in contemporary education presents new challenges, such as technology-related anxiety and test-induced stress that can impair learning outcomes. This study examines how stress affects cognitive load during a learning task and explores gender differences in physiological responses. Sixty participants (30 females and 30 males) were randomly assigned to the control and stress conditions while engaging in a learning task. Wearable sensors were continuously monitoring physiological indicators, such as heart rate and skin conductance, to capture real-time stress and cognitive-based data. A comprehensive data processing pipeline, coupled with machine learning techniques and feature reduction using hierarchical clustering, was applied to classify stress states and levels of cognitive load. Our findings indicate that physiological measures, particularly those derived from skin conductance, effectively differentiate between stressed and non-stressed states. Notably, models tailored to female participants achieved classification accuracies of up to 90%, suggesting more consistent stress responses compared to their male counterparts, who required a broader range of features to reach similar performance. While distinguishing cognitive load levels proved more challenging, the insights gained pave the way for developing adaptive, real-time monitoring systems that could enhance stress management and optimise personalised learning strategies.