Psychophysiological measurements of mental workload for military pilots – results from a simulator study

Kurzfassung

INTRODUCTION: As military aviation becomes increasingly complex, the demands on HumanMachine Interaction are growing. For future fighter jets in particular, which are planned to fly in conjunction with remote carriers, this will lead to a further increase in pilot workload and responsibility. Adaptive automation systems, based on psychophysiological measurements of pilots, could assist the pilot and improve flight safety. For this approach, however, it is necessary to be able to reliably record the different workload states of the pilot. MATERIAL AND METHODS: Pilots completed a 60-minute simulator flight in a full-cockpit simulator, during which they had to perform course corrections, as well as monitor the aircraft’s altitude. In eight 5-minute blocks, subjects had to make course corrections either directly (0-back task), or with a time delay as a 1-, 2-, or 3-back task. The difficulty levels were randomized and each was performed twice. During the entire test, the pilots were monitored using EEG, cortical oxygenation (fNIRS), ECG, eye tracking, blood pressure, skin conductance, body temperature, and respiratory rate measurements using the LiveAmp-32 device, NIRSport2 device, and mobPhysioLab. They were also asked about their subjective mental workload level (MWL). The aim of the study was to find out if the psychophysiological results can be used to differentiate between the performance and MWL of the pilot. RESULTS: Fourteen active male pilots (ages 27 to 49 years) were included in the study. Frontal theta activity (EEG), changes in fNIRS, heart rate, and average NN-intervals are usable to discriminate MWL (p < 0.05). However, other heart rate variability parameters, breathing rate, eye tracking, skin conductivity, temperature, and blood pressure show no significant differences between the different difficulty levels. Subjective MWL ratings increased in higher difficulty levels. CONCLUSIONS: The present study shows that the psychophysiological reactions occurring from increasing MWL are detectable, and can be used to differentiate between the modulated task difficulties. The results of our study can now be used in a development of a model to identify the minimally needed set of psychophysiological parameters. These measures can be used as input for adaptive automation systems that relieve pilots in stressful situations and allow them to focus ontheir primary tasks, especially in the context of increasing Human-Machine Teaming in military aviation.