Multimodal Cueing In Precision Hovering Tasks: Predicting Pilot Cognitive Workload Via Physiological Measurements

Abstract

This study investigates the relationship between mental workload (MWL) and full-body haptic haptic feedback dur- ing an ADS-33 MTE-like precision hovering task involving fourteen participants. The task was conducted under four conditions combining good and degraded visual environments (GVE/DVE) with or without haptic cues. Per- ceived MWL was measured using the Bedford Workload Questionnaire (BWQ), while physiological signals, including cardio-respiratory activity, brain activity (fNIRS), skin temperature, and electrodermal activity, were used as objective indicators of cognitive load. Data were analyzed using univariate statistical tests and multivariate Generalized Linear Mixed Models (GLMM). Results showed that haptic feedback significantly reduced perceived MWL, particularly in degraded visual conditions. Physiological signals demonstrated sensitivity to workload variations, and the GLMM explained up to 88% of the variance. Prediction performance peaked under the most demanding condition, suggesting a stronger physiological response to increased cognitive load. These findings underscore the potential of haptic feed- back to mitigate MWL in safety-critical operations and highlight physiological monitoring as a reliable approach for future adaptive flight systems.