A Comparative Study on Classification of Working Memory Tasks Using EEG Signals

Assessing a mental workload level using electroencephalography (EEG) signals represents an active research area. The development of low-cost wireless EEG headsets drew the attention of researchers in the field of critical human-machine collaboration systems. In this paper, some classification methods are used to discriminate the working memory load levels using EEG raw data records. The brain waves were acquired during several tasks performed according to the n-back paradigm. Selecting the right features and assessing the workload memory level remains a challenging task for a broad range of practical applications. The preprocessed EEG signals are classified with different algorithms suitable for non-stationary signals. The proposed experiments guide an empirical feature selection and correlate EEG signals classification based on n-back memory tests using different acquisition devices. A professional Brain Products helmet and a low-cost wireless Emotiv Epoc+ device were used for comparison. The results prove the impact of two multiclass classifiers, Random Forests (RF) and Support Vector Machine (SVM) and feature selection, on EEG signals classification for n-back working memory tasks and highlight some disadvantages of wireless EEG acquisition tools. The results obtained suggest that both RF and SVM classifiers performed well on EEG data. Workload levels can be precisely identified using RF and a limited set of frontally located electrodes for data acquisition on both EEG devices.

[1]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[2]  Nasour Bagheri,et al.  Multiple classifier system for EEG signal classification with application to brain–computer interfaces , 2012, Neural Computing and Applications.

[3]  Chia-Yen Yang,et al.  Utilizing the extent of theta–gamma synchronization to estimate visuospatial memory ability , 2014, Australasian Physical & Engineering Sciences in Medicine.

[4]  John A. Groeger,et al.  Twenty years of load theory—Where are we now, and where should we go next? , 2016, Psychonomic bulletin & review.

[5]  Musa Peker,et al.  A Comparative Study on Classification of Sleep Stage Based on EEG Signals Using Feature Selection and Classification Algorithms , 2014, Journal of Medical Systems.

[6]  Christopher James,et al.  Neuronal oscillations in the EEG under varying cognitive load: A comparative study between slow waves and faster oscillations , 2013, Clinical Neurophysiology.

[7]  Jianhua Zhang,et al.  Recognition of Cognitive Task Load levels using single channel EEG and Stacked Denoising Autoencoder , 2016, CCC 2016.

[8]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[9]  Debatri Chatterjee,et al.  Cognitive load measurement - A methodology to compare low cost commercial EEG devices , 2014, 2014 International Conference on Advances in Computing, Communications and Informatics (ICACCI).

[10]  Xun Chen,et al.  Classification of EEG Signals Using a Multiple Kernel Learning Support Vector Machine , 2014, Sensors.

[11]  Nian-Shing Chen,et al.  Yet Another Objective Approach for Measuring Cognitive Load Using EEG-Based Workload , 2016, 2016 IEEE 16th International Conference on Advanced Learning Technologies (ICALT).

[12]  Debatri Chatterjee,et al.  Unsupervised approach for measurement of cognitive load using EEG signals , 2013, 13th IEEE International Conference on BioInformatics and BioEngineering.

[13]  Tao Zhang,et al.  AR based quadratic feature extraction in the VMD domain for the automated seizure detection of EEG using random forest classifier , 2017, Biomed. Signal Process. Control..

[14]  M Congedo,et al.  A review of classification algorithms for EEG-based brain–computer interfaces , 2007, Journal of neural engineering.

[15]  Simone Spagnol,et al.  MODEL-BASED OBSTACLE SONIFICATION FOR THE NAVIGATION OF VISUALLY IMPAIRED PERSONS , 2016 .

[16]  Markus Kivikangas,et al.  Brain oscillatory 4–35Hz EEG responses during an n-back task with complex visual stimuli , 2012, Neuroscience Letters.

[17]  Magdalena Borys,et al.  An analysis of EEG signal combined with pupillary response in the dynamics of human cognitive processing , 2016, 2016 9th International Conference on Human System Interactions (HSI).

[18]  Majid Fallahi,et al.  Analysis of the mental workload of city traffic control operators while monitoring traffic density: A field study , 2016 .

[19]  Jacek Gwizdka,et al.  Using Wireless EEG Signals to Assess Memory Workload in the $n$-Back Task , 2016, IEEE Transactions on Human-Machine Systems.

[20]  B. Kotchoubey,et al.  EEG correlates of working memory performance in females , 2017, BMC Neuroscience.

[21]  J. Epps,et al.  Evaluation of Working Memory Load using EEG Signals , 2010 .

[22]  Aamir Saeed Malik,et al.  A comparison of brain regions based on EEG during multimedia learning cognitive activity , 2015, 2015 IEEE Student Symposium in Biomedical Engineering & Sciences (ISSBES).

[23]  Fang Chen,et al.  Characterizing working memory load using EEG delta activity , 2011, 2011 19th European Signal Processing Conference.

[24]  Debatri Chatterjee,et al.  Validation of stimulus for EEG signal based cognitive load analysis , 2015, 2015 Fifth National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics (NCVPRIPG).

[25]  Mohd Fauzi Othman,et al.  Eye blinks removal in single-channel eeg using Savitzky-Golay referenced adaptive filtering: A comparison with independent component analysis method , 2015 .

[26]  Nigel H. Lovell,et al.  Characterization of memory load in an arithmetic task using non-linear analysis of EEG signals , 2012, 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[27]  K Yu,et al.  Cognitive workload modulation through degraded visual stimuli: a single-trial EEG study , 2015, Journal of neural engineering.

[28]  Fang Chen,et al.  Spectral EEG featuresfor evaluating cognitive load , 2011, 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[29]  Yoav Kessler,et al.  Decomposing the n-back task: An individual differences study using the reference-back paradigm , 2016, Neuropsychologia.

[30]  Guido Sanguinetti,et al.  Single-trial classification of EEG in a visual object task using ICA and machine learning , 2014, Journal of Neuroscience Methods.

[31]  Cornelis J. Stam,et al.  APPLYING RANDOM FOREST MACHINE LEARNING TO DIAGNOSE ALZHEIMER’S DISEASE AND DEMENTIA WITH LEWY BODIES: A COMBINATION OF ELECTROENCEPHALOGRAPHY (EEG), CLINICAL PARAMETERS AND BIOMARKERS , 2016, Alzheimer's & Dementia.

[32]  B. Kotchoubey,et al.  EEG correlates of working memory performance in females , 2017 .

[33]  Silvia Villa,et al.  An extension of Mercer theorem to matrix-valued measurable kernels , 2013 .

[34]  Pavlo D. Antonenko,et al.  Using Electroencephalography to Measure Cognitive Load , 2010 .

[35]  Kyriaki Kalimeri,et al.  Exploring multimodal biosignal features for stress detection during indoor mobility , 2016, ICMI.

[36]  Zhang Jianhua,et al.  Recognition of Cognitive Task Load levels using single channel EEG and Stacked Denoising Autoencoder , 2016, 2016 35th Chinese Control Conference (CCC).

[37]  P. Tibbetts :Cognitive Neuroscience: The Biology of the Mind , 2009 .