A Three-Class Classification of Cognitive Workload Based on EEG Spectral Data
Source abstract
Abstract
Evaluation of cognitive workload finds its application in many areas, from educational program assessment through professional driver health examination to monitoring the mental state of people carrying out jobs of high responsibility, such as pilots or airline traffic dispatchers. Estimation of multilevel cognitive workload is a task usually realized in a subject-dependent way, while the present research is focused on developing the procedure of subject-independent evaluation of cognitive workload level. The aim of the paper is to estimate cognitive workload level in accordance with subject-independent approach, applying classical machine learning methods combined with feature selection techniques. The procedure of data acquisition was based on registering the EEG signal of the person performing arithmetical tasks divided into six intervals of advancement. The analysis included the stages of preprocessing, feature extraction, and selection, while the final step covered multiclass classification performed with several models. The results discussed show high maximal accuracies achieved: ~91% for both the validation dataset and for the cross-validation approach for kNN model.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: EEG and Brain-Computer Interfaces · Human-Automation Interaction and Safety · Non-Invasive Vital Sign Monitoring
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Institute of Computer Science, Faculty of Electrical Engineering and Computer Science, Lublin University of Technology, 20618 Lublin, Poland
Department of Experimental Psychology, The John Paul II Catholic University of Lublin, 20950 Lublin, Poland
References (68)
Show source-provided references
1. High-resolution EEG mapping of cortical activation related to working memory: Effects of task difficulty, type of processing, and practice10.1093/cercor/7.4.374 ↗
2. Binet, A., and Henri, V. (1898). La Fatigue Intellectuelle, Librairie C. Reinwald, Schleicher Frères.
3. Mental fatigue impairs physical performance in humans10.1152/japplphysiol.91324.2008 ↗
4. Measuring cognitive load in the presence of educational video: Towards a multimodal methodology10.14742/ajet.3084 ↗
5. Mental workload dynamics in adaptive interface design10.1109/21.17382 ↗
6. Boff, K.R., Kaufman, L., and Thomas, J.P. (1986). Workload assessment methodology. Handbook of Perception and Human Performance, 2: Cognitive Processes and Performance, Sage.
7. NASA-Task Load Index (NASA-TLX); 20 years later10.1177/154193120605000909 ↗
8. Continuous mental effort evaluation during 3D object manipulation tasks based on brain and physiological signals10.1007/978-3-319-22701-6_35 ↗
9. IV The psychometrics of mental workload: Multiple measures are sensitive but divergent10.1177/0018720814539505 ↗
10. Offline and on-line stress detection through processing of the pupil diameter signal10.1007/s10439-013-0880-9 ↗
11. Measurement of cognitive load in HCI systems using EEG power spectrum: An experimental study10.1016/j.procs.2016.04.068 ↗
12. Estimating workload using EEG spectral power and ERPs in the n-back task10.1088/1741-2560/9/4/045008 ↗
13. EEG alpha and theta oscillations reflect cognitive and memory performance: A review and analysis10.1016/S0165-0173(98)00056-3 ↗
14. Andreassi, J.L. (1995). Psychophysiology: Human Behavior and Physiological Response, Lawrence Erlbaum Associates, Inc.
15. Estimating brain load from the EEG10.1100/tsw.2009.83 ↗
16. Putze, F., Jarvis, J.-P., and Schultz, T. (2010, January 23–26). Multimodal recognition of cognitive workload for multitasking in the car. Proceedings of the 2010 20th Conference on Pattern Recognition, Istanbul, Turkey.10.1109/ICPR.2010.913 ↗
17. Combined assessment of attentional reserve and cognitive-motor effort under various levels of challenge with a dry EEG system10.1111/psyp.13059 ↗
18. Measurement of attentional reserve and mental effort for cognitive workload assessment under various task demands during dual-task walking10.1016/j.biopsycho.2018.01.009 ↗
19. Brain-computer interface for workload estimation: Assessment of mental efforts in learning processes10.1016/j.eswa.2018.06.027 ↗
20. Changes in mental workload and motor performance throughout multiple practice sessions under various levels of task difficulty10.1016/j.neuroscience.2018.09.019 ↗
21. Efficient mental work- load estimation using task-independent EEG features10.1088/1741-2560/13/2/026019 ↗
22. Classifying visuomotor workload in a driving simulator using subject specific spatial brain patterns10.3389/fnins.2013.00149 ↗
23. EEG-based mental workload neurometric to evaluate the impact of different traffic and road conditions in real driving settings10.3389/fnhum.2018.00509 ↗
24. Operator functional state classification using multiple psychophysiological features in an air traffic control task10.1518/hfes.45.3.381.27252 ↗
25. Real-time assessment of mental workload using psychophysiological measures and artificial neural networks10.1518/hfes.45.4.635.27088 ↗
26. Adaptive training using an artificial neural network and EEG metrics for within- and cross-task workload classification10.1016/j.neuroimage.2011.07.047 ↗
27. Novel classification system for classifying cognitive workload levels under vague visual stimulation10.1109/JSEN.2017.2727539 ↗
28. Brainshield: HREEG study of perceived pilot mental workload
29. Random forests10.1023/A:1010933404324 ↗
30. Estimating cognitive workload using wavelet entropy-based features during an arithmetic task10.1016/j.compbiomed.2013.08.021 ↗
31. Grimes, D., Tan, D.S., Hudson, S.E., Shenoy, P., and Rao, R.P.N. (2008, January 5–10). Feasibility and pragmatics of classifying working memory load with an electroencephalograph. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, Florence, Italy.10.1145/1357054.1357187 ↗
32. Cognitive workload modulation through degraded visual stimuli: A single-trial EEG study10.1088/1741-2560/12/4/046020 ↗
33. Cross-subject workload classification with a hierarchical Bayes model10.1016/j.neuroimage.2011.07.094 ↗
34. Exploring the fatigue affecting electroencephalography based functional brain networks during real driving in young males10.1016/j.neuropsychologia.2019.04.004 ↗
35. Support vector machine (SVM) classification of cognitive tasks based on electroencephalography (EEG) engagement index10.1080/2326263X.2017.1338012 ↗
36. Estimation of workload using EEG data and classification using linear classifiers
37. Müller-Putz, G., Bauernfeind, G., Brunner, C., Steyrl, D., Wriessnegger, S., and Scherer, R. (2014). Towards cross-subject workload prediction. Proceedings of the 6th International Brain-Computer Interface Conference 2014, Graz University of Technology Publishing House.
38. Detecting driving stress in physiological signals based on multimodal feature analysis and kernel classifiers10.1016/j.eswa.2017.01.040 ↗
39. How does day-to-day variability in psychophysiological data affect classifier accuracy?10.1177/154193121005400317 ↗
40. Driver drowsiness classification using fuzzy wavelet-packet-based feature-extraction algorithm10.1109/TBME.2010.2077291 ↗
41. Magnusdottir, E.H., Johannsdottir, K.R., Bean, C., Olafsson, B., and Gudnason, J. (2017, January 11–14). Cognitive workload classification using cardiovascular measures and dynamic features. Proceedings of the 2017 8th IEEE International Conference on Cognitive Infocommunications (CogInfoCom), Debrecen, Hungary.10.1109/CogInfoCom.2017.8268269 ↗
42. A feature selection method for multilevel mental fatigue EEG classification10.1109/TBME.2007.890733 ↗
43. Real-time mental arithmetic task recognition from EEG signals10.1109/TNSRE.2012.2236576 ↗
44. Emotion recognition from EEG using higher order crossings10.1109/TITB.2009.2034649 ↗
45. Selection of input features across subjects for classifying crewmember workload using artificial neural networks10.1109/TSMCA.2002.807036 ↗
46. EEG-based prediction of cognitive workload induced by arithmetic: A step towards online adaptation in numerical learning10.1007/s11858-015-0754-8 ↗
47. Classification of verbal and quantitative mental tasks using phase locking values between EEG signals
48. EEG-based mental fatigue measurement using multi-class support vector machines with confidence estimate10.1016/j.clinph.2008.03.012 ↗
49. Plechawska-Wójcik, M., Borys, M., Tokovarov, M., Kaczorowska, M., Wesołowska, K., and Wawrzyk, M. (2018, January 4–6). Classifying cognitive workload based on brain waves signal in the arithmetic tasks’ study. Proceedings of the 2018 11th International Conference on Human System Interaction (HSI), Gdansk, Poland.10.1109/HSI.2018.8431105 ↗
50. LU triangularization extreme learning machine in EEG cognitive task classification10.1007/s00521-017-3142-1 ↗
51. Cognitive workload assessment based on the tensorial treatment of EEG estimates of cross-frequency phase interactions10.1007/s10439-014-1143-0 ↗
52. EEG spectral analysis of human cognitive workload study
53. A novel method for classifying driver cognitive distraction under naturalistic conditions with information from near-infrared spectroscopy10.3389/fnhum.2018.00431 ↗
54. Identification of driver cognitive workload using support vector machines with driving performance, physiology and eye movement in a driving simulator10.1007/s12541-013-0179-7 ↗
55. Beyond subjective self-rating: EEG signal classification of cognitive workload10.1109/TAMD.2015.2441960 ↗
56. Classification of motor imagery using combination of feature extraction and reduction methods for brain-computer interface10.5755/j01.itc.48.2.23091 ↗
57. Borys, M., Tokovarov, M., Wawrzyk, M., Wesołowska, K., Plechawska-Wójcik, M., Dmytruk, R., and Kaczorowska, M. (2017, January 23–25). An analysis of eye-tracking and electroencephalography data for cognitive load measurement during arithmetic tasks. Proceedings of the 2017 10th International Symposium on Advanced Topics in Electrical Engineering (ATEE), Bucharest, Romania.10.1109/ATEE.2017.7905130 ↗
58. The PREP pipeline: Standardized preprocessing for large-scale EEG analysis
59. On the influence of high-pass filtering on ICA-based artifact reduction in EEG-ERP
60. The Harvard Automated Processing Pipeline for Electroencephalography (HAPPE): Standardized processing software for developmental and high-artifact data10.3389/fnins.2018.00097 ↗
61. Roffo, G., and Melzi, S. (2016, January 19). Features selection via eigenvector centrality. Proceedings of the International Workshop on New Frontiers in Mining Complex Patterns (NFMCP 2016), Riva del Garda, Italy.
62. Automatic classification of artifactual ICA-components for artifact removal in EEG signals10.1186/1744-9081-7-30 ↗
63. ICLabel: An automated electroencephalographic independent component classifier, dataset, and website10.1016/j.neuroimage.2019.05.026 ↗
64. Application of convolutional neural networks to four-class motor imagery classification problem
65. Sun, Y., and Li, J. (2006, January 25–29). Iterative RELIEF for feature weighting. Proceedings of the 23rd International Conference on Machine Learning, Pittsburgh, PA, USA.10.1145/1143844.1143959 ↗
66. Feature selection based on mutual information criteria of max-dependency max-relevance and min-redundancy10.1109/TPAMI.2005.159 ↗
67. Gu, Q., Li, Z., and Han, J. (2012). Generalized fisher score for feature selection. arXiv.
68. Comparative analysis of machine learning techniques in assessing cognitive workload