A EEG-Based Model for Identifying Human Errors Under Different Loads
Poster No:
47
Submission Type:
Abstract Submission
Authors:
Mengran Wan1, Zhongrui Li1, Tong Li1, Ying Zeng2, Bin Yan2
Institutions:
1Information Engineering University, Zhengzhou, Henan, 2Information Engineering Univer, Zhengzhou, Henan
First Author:
Co-Author(s):
Introduction:
In high-risk and complex industries, such as aviation, healthcare, and nuclear engineering, human errors can lead to incalculable losses. However, the inducing factors of human error are diverse (Islam et al. 2018), and how to build an effective human error identification model remains a challenge in this field.Previous studies have used Event-Related Potentials (ERPs) to identify and analyze human errors(H. Si-Mohammed et al. 2020).However this paper constructs our own dataset of human errors under different cognitive loads, extracting EEG signal features from a period prior to the occurrence of errors . Therefore,this study constructs a novel human error identification model using EEG signals, which can reflect real-time changes in brain activity, to achieve the identification of human errors across various states.
Methods:
Studying human errors under different workloads helps to comprehensively understand the causes of errors and improve safety and efficiency.We will divide the local EEG dataset constructed in our previously study (Li et al., 2024) into human error datasets under four workload levels. Which includes low, medium, high, and extremely high workload levels, with each workload level corresponding to a fixed difficulty level of spatial configuration tasks. The dataset contains 3,781 correct trials and 3,816 error trials (Figure 1). In this study, we proposed a human error recognition model based on EEG signals. To extract EEG data prior to the occurrence of human errors and analyze the data before human errors occurred, we take the EEG data from the 3 seconds before image judgment as a single trial sample. We first perform preprocessing operations (Peng 2019) such as filtering, artifact removal, and re-referencing to eliminate artifacts and noise from the EEG signals. Then, we extract commonly used differential entropy features (Chen et al. 2016) through Short-Time Fourier Transform. Finally, we employed five different classifiers-KNN, RF, SVM, DAC, and Bayes-for human error identification.
·Data Processing for Simulated Spatial Configuration Tasks
Results:
Figure 2 shows the recognition accuracy of the proposed method for human errors under different cognitive load levels. The results indicate that SVM outperforms other classifiers in recognizing human errors across different cognitive load levels. Furthermore, the classification accuracy of SVM is highest at the low cognitive load level, reaching 85.91%. Additionally, we observed that the recognition accuracy of human errors decreases as the cognitive load level increases. This may be due to the fact that as the cognitive load increases, brain activity becomes more complex and diverse, making the brain activity patterns during human errors more difficult to distinguish and recognize.
·The results of human error recognition of the proposed method under different workload levels.
Conclusions:
Overall, this study constructed a human error recognition model based on brain signals, demonstrating that by analyzing brain signals before a human error occurs, it is possible to identify the error state. The study also found that cognitive load has a significant impact on the recognition of human errors. This study confirms the potential of brain signals in identifying human errors under different loads, providing technical support for future human error warning systems.
Brain Stimulation:
Non-Invasive Stimulation Methods Other 1
Emotion, Motivation and Social Neuroscience:
Social Cognition 2
Higher Cognitive Functions:
Executive Function, Cognitive Control and Decision Making
Motor Behavior:
Brain Machine Interface
Novel Imaging Acquisition Methods:
EEG
Keywords:
Other - EEG, human error, recognition, SVM
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.
Chen S, Luo Z, Gan H. 2016. An entropy fusion method for feature extraction of EEG. Neural Computing & Applications.
Islam R, Khan F, Abbassi R, Garaniya V. 2018. Human error assessment during maintenance operations of marine systems – What are the effective environmental factors? Safety Science. 107:85–98.
Zhong L, Li T, Ying Z, Yuan G, Dian G, Kai Y, Yi H, Bin Y. 2024. A novel method of cognitive overload assessment based on a fusion feature selection using EEG signals, J Neural Eng. https://doi.org/10.1088/1741-2552/ad9cc0
Peng, W. (2019). EEG Preprocessing and Denoising. In: Hu, L., Zhang, Z. (eds) EEG Signal Processing and Feature Extraction. Springer, Singapore. https://doi.org/10.1007/978-981-13-9113-2_5