PREDICTION OF DESIGN FATIGUE USING PHYSIOLOGICAL AND PSYCHOLOGICAL INDICATORS

Abstract

In design contexts that require intensive cognitive processing, prolonged mental engagement can lead to fatigue, which hinders creativity, effective decision-making, and productivity. This study outlines a predictive model for design fatigue based on the integration of physiological data—electroencephalography (EEG), heart rate variability (HRV), and eye-tracking— and self-reported data on emotional state and cognitive load. With 20 participants, data was collected while they completed a series of design tasks along with the recording of their bio-signals and subjective feedback. Analysis of the multimodal data revealed strong relationships with fatigue and specific neurophysiological biomarkers of fatigue including heightened frontal theta activity, reduced heart rate variability, and responded with narrowed conjugate gaze. There was also a clear classification of fatigue states within the machine learning models, which permitted the real-time estimation of cognitive load. These results demonstrate the possibility of creating systems that intelligently adapt to and manage cognitive workloads to facilitate optimal performance during engineering and design activities. Privacy, cognitive bias, and consent within the context of monitoring mental functioning are also discussed to promote the ethical use of cognitive monitoring systems.