DEVELOPMENT OF A PSYCHOPHYSIOLOGICAL STRESS MODEL FOR INTELLIGENT ENVIRONMENTAL CONTROL SYSTEMS: A HUMAN-CENTRIC AFFECTIVE COMPUTING APPROACH

Abstract Objective This study aims to develop an intelligent environmental control system based on psychophysiological stress modeling and affective computing to optimize human mental well-being in confined spaces. By integrating psychological stress theory and neural response mechanisms, this research seeks to establish a real-time stress assessment framework that dynamically adjusts environmental parameters (e.g., lighting, temperature, airflow) to mitigate anxiety, emotional dysregulation, and stress-related symptoms. Subjects and Methods The study involved 60 human participants exposed to a controlled simulated confined environment, divided into experimental (n = 30, adaptive system) and control (n = 30, static environment) groups. Psychophysiological indicators—heart rate variability (HRV), electrodermal activity, and behavioral cues (e.g., posture shifts, facial expressions)—were collected via wearable sensors and computer vision. A machine learning model (random forest and LSTM networks) classified stress levels using multimodal data. The intelligent system dynamically adjusted environmental parameters based on real-time stress feedback. Psychological scales (Perceived Stress Scale, State–Trait Anxiety Inventory) and task performance metrics were evaluated over four weeks. Results The stress classification model achieved 89% accuracy, with HRV and electrodermal activity as key predictors. The experimental group demonstrated significantly reduced stress levels (p<.01), showing a 30% decrease in self-reported anxiety and a 25% improvement in emotional regulation compared to controls. Task performance efficiency increased by 15%, and physiological markers indicated enhanced recovery from stress-induced arousal. The adaptive system effectively modulated environmental stimuli, such as softening lighting during high-stress periods. Conclusions Affective computing-driven environmental control systems can significantly improve human mental health by dynamically mitigating psychophysiological stress. This approach highlights the importance of real-time, data-informed interventions for anxiety reduction and emotional stability. Future research should explore applications in clinical settings for individuals with diagnosed anxiety disorders and integrate neurophysiological biomarkers for precision mental health support. Acknowledgment Funded by the “Tianchi Talent” Introduction Program of Xinjiang (Letter from the Xinjiang Human Resources and Social Security Department [2024] No. 57) (2023TCLJ03); Scientific Research Project for High-level Talents of Xinjiang Institute of Technology (XJLG2023G007); Autonomous Region-level College Students’ Innovation and Training Program Project of Xinjiang Institute of Technology in 2024 (S202413558012).