Evaluation of drivers' mental workload based on multi-modal physiological signals DOI Creative Commons

Qiliang ZHANG,

Kunhua YANG,

Xingda Qu

et al.

JOURNAL OF SHENZHEN UNIVERSITY SCIENCE AND ENGINEERING, Journal Year: 2022, Volume and Issue: 39(3), P. 278 - 286

Published: May 1, 2022

Accurately assessing the driver's mental workload is of great significance to reduce traffic accidents caused by overload. This study aims evaluate drivers' in simulated typical driving scenarios, with N-back cognitive tasks used manipulate varied levels task difficulty. We collect data on multi-modal physiological signals including electroencephalogram (EEG), electrocardiogram (ECG), and electrodermal activity (EDA) signals, subjective load National Aeronautics Space Administration index (NASA_TLX) during completion process driver experiment, propose a series classification models based feature analysis pattern recognition signals. These are verified machine learning algorithms random forest, decision tree k-nearest neighbor models. The results show that accuracy varies different modalities EEG-based yield highest among single-modal models, followed EDA-based ECG-based Multi-modal-based generally perform better than forest algorithm three-modal EEG, ECG EDA has accuracy.

Language: Английский

Effects of Mental Workload Manipulation on Electroencephalography Spectrum Oscillation and Microstates in Multitasking Environments DOI Creative Commons
Wenbin Li, Shan Cheng, Jing Dai

et al.

Brain and Behavior, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 1, 2025

ABSTRACT Introduction Multitasking during flights leads to a high mental workload, which is detrimental for maintaining task performance. Electroencephalography (EEG) power spectral analysis based on frequency‐band oscillations and microstate global brain network activation can be used evaluate workload. This study explored the effects of workload simulated flight multitasking EEG parameters. Methods Thirty‐six participants performed with low workloads after 4 consecutive days training. Two levels were set by varying number subtasks. signals acquired task. Power analyses EEG. The indices four frequency bands (delta, theta, alpha, beta) classes (A–D) calculated, changes in parameters under different compared, relationships between two types analyzed. Results theta‐, alpha‐, beta‐band powers higher than condition. Compared condition, condition had lower explained variance time B but D. Less frequent transitions microstates A more C D observed conditions. positively correlated delta‐, powers, whereas duration was negatively power. Conclusion detect not completely isolated multitasking.

Language: Английский

Citations

0
EEG-Based Evaluation of Mental Workload in a Simulated Industrial Human-Robot Interaction Task DOI Open Access
Babak Fazli, Seyed Saman Sajadi, Amir Homayoun Jafari‬

et al.

Health Scope, Journal Year: 2025, Volume and Issue: 14(1)

Published: Feb. 19, 2025

Background: The rapid advancement of robotics and artificial intelligence is poised to revolutionize industrial settings through widespread automation. This study investigates the impact robotic assistance on human operator mental workload (MWL) within a simulated environment. Utilizing electroencephalography (EEG) measure changes in alpha theta band power, we aim identify cognitive challenges associated with human-robot collaboration (HRC) inform design safer more efficient collaborative systems. Objectives: main objective current was assess MWL interaction (HRI) task. Methods: EEG data were collected from 17 participants (aged 25 - 35 years) using 64-channel system while they engaged an ecologically valid task that induced three distinct levels load: Low, medium, high. Subsequent analysis focused power frequency bands, employing repeated-measures ANOVA load brain activity. Results: A revealed significant across different difficulty levels. bands F3, F4, Fz, as well alpha, beta, gamma P3, P4, Pz, emerged promising indicators for differentiating between varying tasks. Conclusions: Electroencephalography spectral particularly reliable indicator MWL. These exhibit dynamic response fluctuating demands, especially human-robotic

Language: Английский

Citations

0
An Explainable Machine Learning Framework for Predicting Driving States Using Electroencephalogram DOI
Iqram Hussain, Se Jin Park, AKM Azad

et al.

Medical Engineering & Physics, Journal Year: 2025, Volume and Issue: unknown, P. 104355 - 104355

Published: May 1, 2025

Language: Английский

Citations

0
The effect of mental schema evolution on mental workload measurement: an EEG study with simulated quadrotor UAV operation DOI
Heng Gu, Qunli Yao, He Chen

et al.

Journal of Neural Engineering, Journal Year: 2022, Volume and Issue: 19(2), P. 026058 - 026058

Published: April 1, 2022

Abstract Objective . Mental workload is the result of interactions between demands an operation task, environment in which task performed, and skills, behavior perception performer. Working under a high mental can significantly affect operator’s ability to choose optimal decisions, judgments motor actions while operating unmanned aerial vehicle (UAV). However, effect schema, reflects level expertise operator, on remains unclear. Here, we propose theoretical framework for describing how evolution schema affects from perspective cognitive processing. Approach We recruited 51 students participate 10-day simulated quadrotor UAV flight training exercise. The EEG power spectral density (PSD)-based metrics were used investigate changes neural responses caused by variations at different stages evolution. Main results It was found that influenced direction change trends frontal theta PSD, parietal alpha central beta are indicators workload. Initially, before formed, only PSD increased with increasing difficulty; when initially being developed, decreased difficulty, difficulty. Finally, as gradually matured, trend three did not differences became more pronounced across difficulty levels, narrowed. Significance Our describe relationship operators evolved. This suggests activity be identify experienced performing provide accurate measurements but also insights into development skill level.

Language: Английский

Citations

16
ARFN: An Attention-Based Recurrent Fuzzy Network for EEG Mental Workload Assessment DOI
Z Wang, Yu Ouyang, Hong Zeng

et al.

IEEE Transactions on Instrumentation and Measurement, Journal Year: 2024, Volume and Issue: 73, P. 1 - 14

Published: Jan. 1, 2024

Assessing mental workload using electroencephalogram (EEG) signals is a significant research avenue within the brain-computer interface domain. However, due to low signal-to-noise ratio in EEG and inter-individual variability data acquisition, achieving high accuracy generalization feature extraction classification for assessment still challenging. We propose novel deep learning framework named attention-based recurrent fuzzy network (ARFN) assessment. In ARFN, we adopt recursive module which employs attention mechanism rule mechanism, respectively, flexibly extract features related workload. The former can frequency domain of signals, while latter used represent membership degrees distribution features, so as find effective rules classification. Subsequently, output directed into long short-term memory (LSTM) further temporal EEG, followed by fully connected layer Softmax function experimental results on three public datasets show that ARFN outperforms other state-of-the-art models

Language: Английский

Citations

3
Synchronization levels in EEG connectivity during cognitive workloads while driving DOI
Nafise Naseri, Fatemeh Parastesh, Farnaz Ghassemi

et al.

Nonlinear Dynamics, Journal Year: 2024, Volume and Issue: unknown

Published: Dec. 11, 2024

Language: Английский

Citations

3
Monitoring army drivers’ workload during off-road missions: An experimental controlled field study DOI
Carolina Díaz-Piedra, Héctor Rieiro, Leandro L. Di Stasi

et al.

Safety Science, Journal Year: 2020, Volume and Issue: 134, P. 105092 - 105092

Published: Nov. 21, 2020

Language: Английский

Citations

19
A Methodological Review on Prediction of Multi-Stage Hypovigilance Detection Systems Using Multimodal Features DOI Creative Commons
Qaisar Abbas, Abdullah Alsheddy

IEEE Access, Journal Year: 2021, Volume and Issue: 9, P. 47530 - 47564

Published: Jan. 1, 2021

Several hypovigilance detection systems (HDx) were developed to avoid road-side accidents due driver fatigue. They have suffered from several limitations. Notably many of these are focused on center-head position define an area interest (often referred as PERCLOS (percentage eye closure)) without considering the face occlusion problem, light illumination, and suffer poor response time. These HDx mostly depend image processing, vision-based, multisensor-based features. To address problems, author utilized vision, sensors, environmental, vehicular-based features that integrated together by fusion predict multistage HDx. Lately, few studies combination multimodal deep learning (DL) architectures. Those multimodal-based (M-HDx) feasible stages fatigue (multi-stage). However, there is a need critically measure performance M-HDx carrying out comparative analysis recognize multi-stage in terms hardware-based benchmarks. Moreover, it important evaluate using different features-set with respect traditional advanced machine techniques. Therefore, primary aim this work algorithm feature modeling, then compare advantages differences other work. In paper, study conducted state-of-the-art survey articles statistically measuring performance. After experiments systems, paper concludes still research gap real-time development systems. end, summarizes directions, challenges, applications assist researchers for further research.

Language: Английский

Citations

15
This Is Your Brain on Autopilot 2.0: The Influence of Practice on Driver Workload and Engagement During On-Road, Partially Automated Driving DOI Creative Commons
Amy S. McDonnell,

Kaedyn W. Crabtree,

Joel M. Cooper

et al.

Human Factors The Journal of the Human Factors and Ergonomics Society, Journal Year: 2023, Volume and Issue: 66(8), P. 2025 - 2040

Published: Sept. 26, 2023

Objective This on-road study employed behavioral and neurophysiological measurement techniques to assess the influence of six weeks practice driving a Level 2 partially automated vehicle on driver workload engagement. Background partial automation requires maintain supervisory control detect “edge cases” that is not equipped handle. There mixed evidence regarding whether drivers can do so effectively. also an open question how familiarity with cognitive states over time. Method Behavioral measures visual engagement were recorded from 30 participants at two testing sessions—with six-week familiarization period in-between. At both sessions, drove engaged (Level 2) 0) interstate highways while reaction times detection response task (DRT) (EEG) metrics frontal theta parietal alpha recorded. Results DRT results demonstrated placed more load than manual decreased workload—though only when environment was relatively simple. EEG showed null effects automation. Conclusion Driver influenced by level automation, specific highway characteristics, time, but neural level. Application These findings expand our understanding under

Language: Английский

Citations

6
SSVEP-assisted RSVP brain–computer interface paradigm for multi-target classification DOI
Li‐Wei Ko,

D Sandeep Vara Sankar,

Yufei Huang

et al.

Journal of Neural Engineering, Journal Year: 2020, Volume and Issue: 18(1), P. 016021 - 016021

Published: Dec. 9, 2020

Abstract Brain–computer Interface (BCI) is actively involved in optimizing the communication medium between human brain and external devices. Objective. Rapid serial visual presentation (RSVP) a robust highly efficient BCI technique recognizing target objects but suffers from limited selections. Hybrid systems that combine steady-state evoked potential (SSVEP) RSVP can mitigate this limitation allow users to operate on multiple targets. Approach. This study proposes novel hybrid SSVEP-RSVP improve performance of classifying target/non-target multi-target scenario. In paradigm, SSVEP stimulation helps identifying user’s focus location stimuli elicit event-related potentials differentiate non-target objects. Main results. The proposed model achieved an offline accuracy 81.59% by using 12 electroencephalography (EEG) channels online (real-time) 78.10% when only four EEG are considered. Further, biomarkers physiological states analyzed assess cognitive (mental fatigue user attention) participants based resting theta alpha band powers. results indicate inverse relationship power, validating subjects’ affected for long-term use BCI. Significance. Our findings demonstrate combination improves further enhances possibility performing command tasks, which inevitable real-world applications. Additionally, state discussed imply need attractive experimental paradigm reduces disparities provide enhanced performance.

Language: Английский

Citations

16