Human Mental Workload: A Survey and a Novel Inclusive Definition

Frontiers in Psychology, Journal Year: 2022, Volume and Issue: 13

Published: June 2, 2022

Human mental workload is arguably the most invoked multidimensional construct in Factors and Ergonomics, getting momentum also Neuroscience Neuroergonomics. Uncertainties exist its characterization, motivating design development of computational models, thus recently actively receiving support from discipline Computer Science. However, role human performance prediction assured. This work aimed at providing a synthesis current state art assessment through considerations, definitions, measurement techniques as well applications, Findings suggest that, despite an increasing number associated research works, single, reliable generally applicable framework for does not yet appear fully established. One reason this gap existence wide swath operational built upon different theoretical assumptions which are rarely examined collectively. A second that three main classes measures, self-report, task performance, physiological indices, have been used isolation or pairs, but more conjunction all together. Multiple definitions complement each another we propose novel inclusive definition to next generation empirical-based research. Similarly, by comprehensively employing physiological, task-performance, self-report robust assessments can be achieved.

Language: Английский

---

Using machine learning methods and EEG to discriminate aircraft pilot cognitive workload during flight

Scientific Reports, Journal Year: 2023, Volume and Issue: 13(1)

Published: Feb. 13, 2023

Pilots of aircraft face varying degrees cognitive workload even during normal flight operations. Periods low may be followed by periods high and vice versa. During such changing demands, there exists potential for increased error on behalf the pilots due to boredom or excessive task demand. To further understand in aviation, present study involved collection electroencephalogram (EEG) data from ten (10) collegiate aviation students a live-flight environment single-engine aircraft. Each pilot possessed Federal Aviation Administration (FAA) commercial certificate either FAA class I II medical certificate. flew standardized profile representing an average instrument training sequence. For analysis, we used four main sub-bands recorded EEG signals: delta, theta, alpha, beta. Power spectral density (PSD) log energy entropy each sub-band across 20 electrodes were computed subjected two feature selection algorithms (recursive elimination (RFE) lasso cross-validation (LassoCV), stacking ensemble machine learning algorithm composed support vector machine, random forest, logistic regression. Also, hyperparameter optimization tenfold improve model performance, reliability, generalization. The step resulted 15 features that can considered indicator pilots' states. Then these applied algorithm, highest results achieved using selected RFE with accuracy 91.67% (± 0.11), precision 93.89% 0.09), recall F-score 91.22% 0.12), mean ROC-AUC 0.93 0.06). indicated combination PSD entropy, along well-designed algorithms, suggest use discriminate low, medium, augment system design, including automation safety.

Language: Английский

---

An Evaluation of the EEG Alpha-to-Theta and Theta-to-Alpha Band Ratios as Indexes of Mental Workload

Frontiers in Neuroinformatics, Journal Year: 2022, Volume and Issue: 16

Published: May 16, 2022

Many research works indicate that EEG bands, specifically the alpha and theta have been potentially helpful cognitive load indicators. However, minimal exists to validate this claim. This study aims assess analyze impact of alpha-to-theta theta-to-alpha band ratios on supporting creation models capable discriminating self-reported perceptions mental workload. A dataset raw data was utilized in which 48 subjects performed a resting activity an induced task demanding exercise form multitasking SIMKAP test. Band were devised from frontal parietal electrode clusters. Building model testing done with high-level independent features frequency temporal domains extracted computed over time. Target for training subjective ratings collected after demand activities. Models built by employing Logistic Regression, Support Vector Machines Decision Trees evaluated performance measures including accuracy, recall, precision f1-score. The results high classification accuracy those trained ratios. Preliminary also show logistic regression support vector machines can accurately classify contributes body knowledge demonstrating richness information temporal, spectral statistical discrimination

Language: Английский

---

Assessing the effect of construction noise frequency on mental workload of construction workers with varying task difficulty using EEG data

Applied Acoustics, Journal Year: 2025, Volume and Issue: 232, P. 110571 - 110571

Published: Feb. 1, 2025

Language: Английский

---

Cognitive workload assessment during VR forklift training

International Journal of Industrial Ergonomics, Journal Year: 2025, Volume and Issue: 107, P. 103718 - 103718

Published: March 30, 2025

Language: Английский

---

Driving-Induced Neurological Biomarkers in an Advanced Driver-Assistance System

Sensors, Journal Year: 2021, Volume and Issue: 21(21), P. 6985 - 6985

Published: Oct. 21, 2021

Physiological signals are immediate and sensitive to neurological changes resulting from the mental workload induced by various driving environments considered a quantifying tool for understanding association between outcomes cognitive workloads. Neurological assessment, outside of highly-equipped clinical setting, requires an ambulatory electroencephalography (EEG) headset. This study aimed quantify biomarkers during resting state two different scenarios states in virtual environment. We investigated responses seventeen healthy male drivers. EEG data were measured initial state, city-roadways expressway using portable headset simulator. During experiment, participants drove while experiencing workloads due environments, such as road traffic conditions, lane surrounding vehicles, speed limit, etc. The power beta gamma bands decreased, delta waves, theta, frontal theta asymmetry increased relative state. Delta-alpha ratio (DAR) delta-theta (DTR) showed strong correlation with Binary machine-learning (ML) classification models near-perfect accuracy Moderate performances observed multi-class classification. An EEG-based prediction approach may be utilized advanced driver-assistance system (ADAS).

Language: Английский

---

Pretrial Theta Band Activity Affects Context-dependent Modulation of Response Inhibition

Journal of Cognitive Neuroscience, Journal Year: 2022, Volume and Issue: 34(4), P. 605 - 617

Published: Jan. 21, 2022

Abstract The ability to inhibit a prepotent response is crucial prerequisite of goal-directed behavior. So far, research on inhibition has mainly examined these processes when there little no cognitive control during the decision respond. We manipulated “context” in which be exerted (i.e., controlled or an automated context) by combining Simon task with go/no-go and focused theta band activity. To investigate role inhibition, we also how far activity pretrial period modulates context-dependent variations inhibition. This was done EEG study applying beamforming methods. Here, n = 43 individuals. show that context, as opposed compromises performance increases need for control. related modulations superior frontal middle regions. Of note, results showed period, associated right inferior cortex, substantially correlated direction obtained correlation provides insights into functional relevance data suggest reflects some form attentional sampling inform possible upcoming signaling

Language: Английский

---

Detecting Mental Workload in Surgical Teams Using a Wearable Single-Channel Electroencephalographic Device

Journal of surgical education, Journal Year: 2019, Volume and Issue: 76(4), P. 1107 - 1115

Published: Jan. 27, 2019

Language: Английский

---

Automated detection of driver fatigue from electroencephalography through wavelet-based connectivity

Journal of Applied Biomedicine, Journal Year: 2020, Volume and Issue: 41(1), P. 316 - 332

Published: Sept. 6, 2020

Language: Английский

---

Optimized electroencephalogram and functional near-infrared spectroscopy-based mental workload detection method for practical applications

BioMedical Engineering OnLine, Journal Year: 2022, Volume and Issue: 21(1)

Published: Feb. 2, 2022

Mental workload is a critical consideration in complex man-machine systems design. Among various mental detection techniques, multimodal techniques integrating electroencephalogram (EEG) and functional near-infrared spectroscopy (fNIRS) signals have attracted considerable attention. However, existing EEG-fNIRS-based methods certain defects, such as signal acquisition channels low accuracy, which restrict their practical application.The configuration was optimized by analyzing the feature importance recognition model more accurate convenient method constructed. A classical Multi-Task Attribute Battery (MATB) task conducted with 20 participating volunteers. Subjective scale data, 64-channel EEG two-channel fNIRS data were collected.A higher number of correspond to accuracy. there no obvious improvement accuracy once reaches 26, four-level 76.25 ± 5.21%. Partial results physiological analysis verify previous studies, that θ power concentration O2Hb prefrontal region increase while HHb decreases difficulty. It further observed, for first time, energy each band significantly different occipital lobe region, [Formula: see text] bands increased The changing range mean amplitude high-difficulty tasks compared those low-difficulty tasks.The channel 26 two frontal channels. 5.21% obtained, than previously reported results. proposed can promote application technology military, driving, other human-computer interaction systems.

Language: Английский

---