Slide
Centre Interdisciplinaire
de Recherche et d’Innovation
en Cybersécurité et Société
de Recherche et d’Innovation
en Cybersécurité et Société
1.
Tiwari, A.; Arrabito, R.; Davoust, A.; Falk, T. H.
Quantifying Biological Sex Leakage in Electroencephalography-Based Mental Workload Measurement and Its Impact on Model Performance Article d'actes
Dans: IEEE Int. Conf. Hum.-Mach. Syst., ICHMS, p. 392–397, Institute of Electrical and Electronics Engineers Inc., 2026, ISBN: 979-833154511-6 (ISBN), (Journal Abbreviation: IEEE Int. Conf. Hum.-Mach. Syst., ICHMS).
Résumé | Liens | BibTeX | Étiquettes: Biomedical signal processing, Demographic information, Designing systems, Economic and social effects, Electroencephalography, Electrophysiology, Human operator, Mental workload, Modeling performance, Optimal performance, Performance, Population statistics, Real- time, Well being, Workload measurements
@inproceedings{tiwariQuantifyingBiologicalSex2026,
title = {Quantifying Biological Sex Leakage in Electroencephalography-Based Mental Workload Measurement and Its Impact on Model Performance},
author = {A. Tiwari and R. Arrabito and A. Davoust and T. H. Falk},
url = {https://www.scopus.com/pages/publications/105045575255?origin=resultslist},
doi = {10.1109/ICHMS69701.2026.11602273},
isbn = {979-833154511-6 (ISBN)},
year = {2026},
date = {2026-01-01},
booktitle = {IEEE Int. Conf. Hum.-Mach. Syst., ICHMS},
pages = {392–397},
publisher = {Institute of Electrical and Electronics Engineers Inc.},
abstract = {Detection of mental workload is essential for designing systems that can monitor and adapt to the needs of human operators, thereby ensuring their optimal performance and well-being. Features derived from neurophysiological signals, particularly electroencephalograms (EEG), provide a realtime, unobtrusive, and objective indicator of mental workload. However, EEG signals are often correlated with demographic variables, including biological sex and age, which can introduce systematic performance differences across demographic groups, while unintentionally encoding demographic information into models trained on EEG data. In this work, we investigate and quantify the influence of user demographics on mental workload detection models using fairness analysis. Furthermore, we propose domain-adversarial training, in which an adversary penalizes the model for retaining sex-related information, to mitigate leakage of private demographic information (biological sex). The proposed approach simultaneously increases the odds ratio for sex from 0.83 to 0.90 (values closer to 1 indicate reduced demographic disparity) while reducing statistical significance. This approach, however, also leads to a reduction in mental workload detection performance, with a 12.7% decrease in balanced accuracy (BACC), suggesting an inherent trade-off between performance and privacy protection. © 2026 IEEE.},
note = {Journal Abbreviation: IEEE Int. Conf. Hum.-Mach. Syst., ICHMS},
keywords = {Biomedical signal processing, Demographic information, Designing systems, Economic and social effects, Electroencephalography, Electrophysiology, Human operator, Mental workload, Modeling performance, Optimal performance, Performance, Population statistics, Real- time, Well being, Workload measurements},
pubstate = {published},
tppubtype = {inproceedings}
}
Detection of mental workload is essential for designing systems that can monitor and adapt to the needs of human operators, thereby ensuring their optimal performance and well-being. Features derived from neurophysiological signals, particularly electroencephalograms (EEG), provide a realtime, unobtrusive, and objective indicator of mental workload. However, EEG signals are often correlated with demographic variables, including biological sex and age, which can introduce systematic performance differences across demographic groups, while unintentionally encoding demographic information into models trained on EEG data. In this work, we investigate and quantify the influence of user demographics on mental workload detection models using fairness analysis. Furthermore, we propose domain-adversarial training, in which an adversary penalizes the model for retaining sex-related information, to mitigate leakage of private demographic information (biological sex). The proposed approach simultaneously increases the odds ratio for sex from 0.83 to 0.90 (values closer to 1 indicate reduced demographic disparity) while reducing statistical significance. This approach, however, also leads to a reduction in mental workload detection performance, with a 12.7% decrease in balanced accuracy (BACC), suggesting an inherent trade-off between performance and privacy protection. © 2026 IEEE.
2.
Moinnereau, M. -A.; Tiwari, A.; Benesch, D.; Bolt, N.; Krätzig, G. P.; Paré, S.; Falk, T. H.
Subjective and Objective Assessment of the Impact of Stress and Mental Workload on Cybersickness During Virtual Reality Training Article de journal
Dans: Applied Human Factors and Ergonomics International, vol. 178, p. 87–98, 2025, ISSN: 27710718 (ISSN).
Résumé | Liens | BibTeX | Étiquettes: Biosensors, cybersickness, Electroencephalography, Mental workload, Stress, virtual reality, VR training
@article{moinnereauSubjectiveObjectiveAssessment2025,
title = {Subjective and Objective Assessment of the Impact of Stress and Mental Workload on Cybersickness During Virtual Reality Training},
author = {M. -A. Moinnereau and A. Tiwari and D. Benesch and N. Bolt and G. P. Krätzig and S. Paré and T. H. Falk},
url = {https://www.scopus.com/pages/publications/105031119708?origin=resultslist},
doi = {10.54941/ahfe1006348},
issn = {27710718 (ISSN)},
year = {2025},
date = {2025-01-01},
journal = {Applied Human Factors and Ergonomics International},
volume = {178},
pages = {87–98},
publisher = {AHFE International},
abstract = {Cybersickness is an issue in immersive virtual reality (VR), akin to motion and simulator sickness, resulting in symptoms such as nausea, dizziness, and eye strain. Cybersickness has been shown to affect a significant portion of VR users. In training scenarios involving demanding tasks (e.g., for first responders’ training), however, reports of cybersickness symptoms are higher than those for the average user. It is hypothesized that the stress and mental workload generated by these scenarios may be the cause for this increased propensity for cybersickness. In this study, we investigate the impact of stress alone, mental workload alone, and their combined impact on cybersickness levels. The levels of stress and mental workload are manipulated while participants perform a driving simulator task. In the high mental workload condition, the driver has to keep an eye on the road while driving from one location to another, as well as monitor and count the number of pedestrians wearing a certain colour shirt. In the high stress condition, traffic conditions become heavy, background noise increases, and sudden breaks are needed to avoid accidents (e.g., from a ball rolling into the road to a car suddenly changing lanes). Lastly, the combined condition contains all the elements of the previous two conditions. In all cases, a baseline driving period is present (without stress or workload) and is used for comparisons within each subject. Both self-report and neurophysiological measurements are used to gauge the impact of these three conditions on cybersickness. Self-report questionnaires are used to assess stress (DASS-21), mental workload (NASA-TLX), and cybersickness symptoms (SSQ) at several instances during the experiment. In turn, an instrumented Meta Quest 3 VR headset is used equipped with 16 electroencephalography (EEG) and electro-oculography (EOG) sensors, while wearable devices are used to monitor photoplethysmography (PPG), electrocardiography (ECG), and respiration signals. These neurophysiological signals are used to continuously extract measures of mental workload, stress, and other cognitive/affective states almost in real-time. In this paper, we describe the experimental setup, the instrumented headset, the EEG and biosignal metrics that are computed, and provide preliminary subjective and objective findings based on the first 12 participants (four per condition). The study is ongoing and aims to collect data from 60 participants (20 per condition). It is hoped that these preliminary insights will help the research community refine VR training protocols, making them more comfortable and effective for students. © 2025. Published by AHFE Open Access. All rights reserved.},
keywords = {Biosensors, cybersickness, Electroencephalography, Mental workload, Stress, virtual reality, VR training},
pubstate = {published},
tppubtype = {article}
}
Cybersickness is an issue in immersive virtual reality (VR), akin to motion and simulator sickness, resulting in symptoms such as nausea, dizziness, and eye strain. Cybersickness has been shown to affect a significant portion of VR users. In training scenarios involving demanding tasks (e.g., for first responders’ training), however, reports of cybersickness symptoms are higher than those for the average user. It is hypothesized that the stress and mental workload generated by these scenarios may be the cause for this increased propensity for cybersickness. In this study, we investigate the impact of stress alone, mental workload alone, and their combined impact on cybersickness levels. The levels of stress and mental workload are manipulated while participants perform a driving simulator task. In the high mental workload condition, the driver has to keep an eye on the road while driving from one location to another, as well as monitor and count the number of pedestrians wearing a certain colour shirt. In the high stress condition, traffic conditions become heavy, background noise increases, and sudden breaks are needed to avoid accidents (e.g., from a ball rolling into the road to a car suddenly changing lanes). Lastly, the combined condition contains all the elements of the previous two conditions. In all cases, a baseline driving period is present (without stress or workload) and is used for comparisons within each subject. Both self-report and neurophysiological measurements are used to gauge the impact of these three conditions on cybersickness. Self-report questionnaires are used to assess stress (DASS-21), mental workload (NASA-TLX), and cybersickness symptoms (SSQ) at several instances during the experiment. In turn, an instrumented Meta Quest 3 VR headset is used equipped with 16 electroencephalography (EEG) and electro-oculography (EOG) sensors, while wearable devices are used to monitor photoplethysmography (PPG), electrocardiography (ECG), and respiration signals. These neurophysiological signals are used to continuously extract measures of mental workload, stress, and other cognitive/affective states almost in real-time. In this paper, we describe the experimental setup, the instrumented headset, the EEG and biosignal metrics that are computed, and provide preliminary subjective and objective findings based on the first 12 participants (four per condition). The study is ongoing and aims to collect data from 60 participants (20 per condition). It is hoped that these preliminary insights will help the research community refine VR training protocols, making them more comfortable and effective for students. © 2025. Published by AHFE Open Access. All rights reserved.