Online cognitive overload assessment method based on few-channel EEG Signals
Poster No:
1324
Submission Type:
Abstract Submission
Authors:
Zhongrui Li1, Ying Zeng1, Li Tong1, Bin Yan1
Institutions:
1Information Engineering University, Zhengzhou, Henan
First Author:
Co-Author(s):
Introduction:
Cognitive overload, as an extreme load state, poses significant negative impacts on individual work efficiency and mental and physical health (Biondi et al. 2021). The high temporal resolution of electroencephalogram (EEG) signals provides a new perspective for the online monitoring of cognitive overload states (Kosachenko et al. 2023). However, while the multichannel nature of EEG signals offers rich information, it also introduces challenges of higher computational resources and longer preparation times. Therefore, this study designs an online cognitive overload assessment method based on a small number of EEG channels, aiming to improve the efficiency of the model by reducing the number of channels while ensuring the accuracy of the assessment.
Methods:
To improve the efficiency of the model, this paper proposed a cognitive overload assessment method based on a few-channel EEG signals, integrating the advantages of the feature selection methods of Mutual Information (MI) (Piho and Tjahjadi 2020) and Fisher Score (Luo et al. 2024) to comprehensively consider the relevance and separability of channel features, thereby enhancing the effectiveness of the selected features. The overall framework of the method is shown in Figure 1, which mainly includes the construction of the offline model and pseudo-online testing. In the construction of the offline model, noise and artifacts are removed through preprocessing operations and the short-time Fourier transforms (Shen et al. 2024) are used to extract frequency-domain features from each channel, such as the spectrum, power spectral density (PSD), and differential entropy (DE). Then, the channels are ranked and selected based on the MI and Fisher Score feature selection methods, and the Support Vector Machine (SVM) (Han and Song 2022) is used as classifiers. During pseudo-online testing, the data processing uses the channel signals selected during the offline model creation process. The performance of the proposed method is verified based on an EEG dataset constructed from our previous experiments (Li et al. 2024), which includes cognitive overload and non-cognitive overload states. The EEG data were collected using a g.HIamp system with 62 channels, and the dataset size is 32×2×50 (number of subjects × cognitive state × trials). For each state of all subjects, the first 48 trials are used for offline model construction, and the last 2 trials are used for pseudo-online analysis.
Fig. 1. The overall framework of the proposed method.
Fig. 1. The overall framework of the proposed method.
Results:
Figure 2A shows the incremental evaluation performance of the sorted subset of channels. The results indicate that the fusion feature selection method outperforms single feature selection methods, and the top four channels of the sorted channels subset can achieve performance convergence, with a classification performance as high as 81.58%. Figure 2B shows the offline and online performance when only four channels are used, with the online performance reaching 77.25%. Figure 2C displays the electrode position distribution of the channels selected by the fusion feature selection method, which are located at the frontal lobe (FP1 and FP2) and the parietal lobe (FCZ and FC2). These brain regions are closely related to advanced cognitive functions (Otero and Barker 2014; Humphreys and Lambon Ralph 2015), reflecting that changes in frontal and parietal lobe activities can capture variations in brain activity patterns during cognitive overload states.
Fig. 2. The result analysis chart of the proposed method.
Fig. 2. The result analysis chart of the proposed method.
Conclusions:
In summary, this study proposed a cognitive overload assessment method based on few-channel EEG signals, which improves the overall efficiency of the model while ensuring assessment performance. This method provides an effective solution for online monitoring and assessment of cognitive overload states. In the future, this method is expected to find further applications and developments in human-machine interaction systems.
Higher Cognitive Functions:
Executive Function, Cognitive Control and Decision Making 2
Space, Time and Number Coding
Higher Cognitive Functions Other
Modeling and Analysis Methods:
EEG/MEG Modeling and Analysis 1
Novel Imaging Acquisition Methods:
EEG
Keywords:
Other - EEG, cognitive overload assessment, channel selection, online.
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
Provide references using APA citation style.
Han L, Song X. 2022. Epileptic Seizure Prediction Using Effective Brain Network and SVM on EEG. Proceedings of the 14th International Conference on Bioinformatics and Biomedical Technology.
Humphreys GF, Lambon Ralph MA. 2015. Fusion and Fission of Cognitive Functions in the Human Parietal Cortex. Cereb Cortex. 25:3547–3560.
Kosachenko AI, Kasanov D, Kotyusov AI, Pavlov YG. 2023. EEG and pupillometric signatures of working memory overload. Psychophysiology. 60:e14275.
Li Z, Tong L, Zeng Y, Gao Y, Gong D, Yang K, Hu Y, Yan B. 2024. A novel method of cognitive overload assessment based on a fusion feature selection using EEG signals. Journal of Neural Engineering.
Luo Y, Mu W, Wang L, Wang J, Wang P, Gan Z, Zhang L, Kang X. 2024. An EEG channel selection method for motor imagery based on Fisher score and local optimization. J Neural Eng. 21:036030.
Otero TM, Barker LA. 2014. The Frontal Lobes and Executive Functioning. In: Goldstein S,, Naglieri JA, editors. Handbook of Executive Functioning. New York, NY: Springer New York. p. 29–44.
Piho L, Tjahjadi T. 2020. A Mutual Information Based Adaptive Windowing of Informative EEG for Emotion Recognition. IEEE Trans Affective Comput. 11:722–735.
Shen M, Yang F, Wen P, Song B, Li Y. 2024. A real-time epilepsy seizure detection approach based on EEG using short-time Fourier transform and Google-Net convolutional neural network. Heliyon. 10:e31827.