• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利：CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法，用中文像聊天一样搜索，搜遍4000万医学文献。AI智能推荐，让科研检索更轻松。

立即免费搜索

文件翻译

保留排版，准确专业，支持PDF/Word/PPT等文件格式，支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述，25分钟生成高质量综述，智能提取关键信息，辅助科研写作。

立即免费体验

基于堆叠图注意力卷积网络的脑电图独立于任务的认知工作量判别

Task-Independent Cognitive Workload Discrimination Based on EEG with Stacked Graph Attention Convolutional Networks.

作者信息

Wei Chenyu, Zhao Xuewen, Song Yu, Liu Yi

机构信息

Tianjin Key Laboratory for Control Theory and Applications in Complicated Systems, School of Electrical Engineering and Automation, Tianjin University of Technology, Tianjin 300384, China.

College of Mechanical and Electrical Engineering, Harbin Engineering University, Harbin 150001, China.

出版信息

Sensors (Basel). 2025 Apr 9;25(8):2390. doi: 10.3390/s25082390.

DOI:10.3390/s25082390
PMID:40285080
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC12031105/
Abstract

In the field of neuroeconomics, the assessment of cognitive workload is a crucial issue with significant implications for real-world applications. Previous research has made progress in task-based germane cognitive load classification, but decentralized studies focusing on task-independent assessment have often produced less than optimal results. In this study, we present a stacked graph attention convolutional networks (SGATCNs) model to tackle the challenges related to task-independent cognitive workload assessment using EEG spatial information. The model employs the differential entropy (DE) and power spectral density (PSD) features of each EEG channel across four frequency bands (delta, theta, alpha, and beta) as node information. For the construction of the network structure, phase-locked values (PLVs), phase-lag indices (PLIs), Pearson correlation coefficients (PCCs), and mutual information (MI) are utilized and evaluated to generate a functional brain network. Specifically, the model aggregates spatial information on the dynamic map by stacking the graph attention layers and utilizes the convolution module to extract the frequency domain information from between the networks under each frequency band. We conducted a cognitive workload experiment with 15 subjects and selected three representative psychological experimental task paradigms (N-back, mental arithmetic, and Sternberg) to induce different levels of cognitive workload (low, medium, and high). Our framework achieved an average accuracy of 65.11% in recognizing the task-independent cognitive workload across the three scenarios.

摘要

在神经经济学领域，认知工作量的评估是一个关键问题，对实际应用具有重要意义。以往的研究在基于任务的相关认知负荷分类方面取得了进展，但专注于独立于任务的评估的分散研究往往产生的结果不尽如人意。在本研究中，我们提出了一种堆叠图注意力卷积网络（SGATCNs）模型，以应对使用脑电图（EEG）空间信息进行独立于任务的认知工作量评估所面临的挑战。该模型将四个频段（δ、θ、α和β）上每个EEG通道的微分熵（DE）和功率谱密度（PSD）特征作为节点信息。对于网络结构的构建，利用并评估锁相值（PLV）、相位滞后指数（PLI）、皮尔逊相关系数（PCC）和互信息（MI）来生成功能性脑网络。具体而言，该模型通过堆叠图注意力层在动态地图上聚合空间信息，并利用卷积模块从每个频段下的网络之间提取频域信息。我们对15名受试者进行了认知工作量实验，并选择了三种具有代表性的心理实验任务范式（N-back、心算和斯特恩伯格任务）来诱导不同水平的认知工作量（低水平、中等水平和高水平）。我们的框架在识别三种场景下独立于任务的认知工作量时，平均准确率达到了65.11%。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/86e7818a9a35/sensors-25-02390-g018.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/6edc1bb60eaf/sensors-25-02390-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/b4032d3a72d9/sensors-25-02390-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/0db198c46af0/sensors-25-02390-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/a5787f8c927b/sensors-25-02390-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/13c29a4c66d8/sensors-25-02390-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/7e8d0b54dd44/sensors-25-02390-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/579711631a6d/sensors-25-02390-g007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/82a351c8d950/sensors-25-02390-g008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/5ed5bfbe3ec4/sensors-25-02390-g009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/3bd71b84f2db/sensors-25-02390-g010.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/413a751520cf/sensors-25-02390-g011.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/a889db395f46/sensors-25-02390-g012.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/60ffcaa36e3e/sensors-25-02390-g013.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/7e8c998cbf2b/sensors-25-02390-g014.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/4b5c22ef9b99/sensors-25-02390-g015.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/207454acfb97/sensors-25-02390-g016.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/a4a72237d6b5/sensors-25-02390-g017.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/86e7818a9a35/sensors-25-02390-g018.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/6edc1bb60eaf/sensors-25-02390-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/b4032d3a72d9/sensors-25-02390-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/0db198c46af0/sensors-25-02390-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/a5787f8c927b/sensors-25-02390-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/13c29a4c66d8/sensors-25-02390-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/7e8d0b54dd44/sensors-25-02390-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/579711631a6d/sensors-25-02390-g007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/82a351c8d950/sensors-25-02390-g008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/5ed5bfbe3ec4/sensors-25-02390-g009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/3bd71b84f2db/sensors-25-02390-g010.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/413a751520cf/sensors-25-02390-g011.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/a889db395f46/sensors-25-02390-g012.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/60ffcaa36e3e/sensors-25-02390-g013.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/7e8c998cbf2b/sensors-25-02390-g014.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/4b5c22ef9b99/sensors-25-02390-g015.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/207454acfb97/sensors-25-02390-g016.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/a4a72237d6b5/sensors-25-02390-g017.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/c719/12031105/86e7818a9a35/sensors-25-02390-g018.jpg

相似文献

1
Task-Independent Cognitive Workload Discrimination Based on EEG with Stacked Graph Attention Convolutional Networks.基于堆叠图注意力卷积网络的脑电图独立于任务的认知工作量判别
Sensors (Basel). 2025 Apr 9;25(8):2390. doi: 10.3390/s25082390.
2
Learning Spatial-Spectral-Temporal EEG Features With Recurrent 3D Convolutional Neural Networks for Cross-Task Mental Workload Assessment.基于循环三维卷积神经网络学习空间-谱-时 EEG 特征进行跨任务精神工作负荷评估。
IEEE Trans Neural Syst Rehabil Eng. 2019 Jan;27(1):31-42. doi: 10.1109/TNSRE.2018.2884641. Epub 2018 Dec 3.
3
Subject-Specific Cognitive Workload Classification Using EEG-Based Functional Connectivity and Deep Learning.基于 EEG 功能连接和深度学习的特定主题认知工作负荷分类。
Sensors (Basel). 2021 Oct 9;21(20):6710. doi: 10.3390/s21206710.
4
Cross-subject emotion recognition in brain-computer interface based on frequency band attention graph convolutional adversarial neural networks.基于频带注意力图卷积对抗神经网络的脑机接口跨主体情绪识别。
J Neurosci Methods. 2024 Nov;411:110276. doi: 10.1016/j.jneumeth.2024.110276. Epub 2024 Sep 3.
5
Cross-task cognitive workload recognition using a dynamic residual network with attention mechanism based on neurophysiological signals.基于神经生理信号的注意力机制动态残差网络的跨任务认知负荷识别。
Comput Methods Programs Biomed. 2023 Mar;230:107352. doi: 10.1016/j.cmpb.2023.107352. Epub 2023 Jan 13.
6
An Efficient Graph Learning System for Emotion Recognition Inspired by the Cognitive Prior Graph of EEG Brain Network.一种受脑电图脑网络认知先验图启发的高效情感识别图学习系统。
IEEE Trans Neural Netw Learn Syst. 2025 Apr;36(4):7130-7144. doi: 10.1109/TNNLS.2024.3405663. Epub 2025 Apr 4.
7
Task-State EEG Signal Classification for Spatial Cognitive Evaluation Based on Multiscale High-Density Convolutional Neural Network.基于多尺度高密度卷积神经网络的用于空间认知评估的任务状态脑电信号分类
IEEE Trans Neural Syst Rehabil Eng. 2022;30:1041-1051. doi: 10.1109/TNSRE.2022.3166224. Epub 2022 Apr 25.
8
Emotion Classification from Multi-Band Electroencephalogram Data Using Dynamic Simplifying Graph Convolutional Network and Channel Style Recalibration Module.基于动态简化图卷积网络和通道风格重校准模块的多波段脑电图数据情绪分类。
Sensors (Basel). 2023 Feb 8;23(4):1917. doi: 10.3390/s23041917.
9
Emotion recognition using multi-scale EEG features through graph convolutional attention network.通过图卷积注意力网络利用多尺度脑电图特征进行情绪识别。
Neural Netw. 2025 Apr;184:107060. doi: 10.1016/j.neunet.2024.107060. Epub 2024 Dec 27.
10
Spectral and Temporal Feature Learning With Two-Stream Neural Networks for Mental Workload Assessment.基于双流神经网络的心理工作负荷评估的光谱和时间特征学习。
IEEE Trans Neural Syst Rehabil Eng. 2019 Jun;27(6):1149-1159. doi: 10.1109/TNSRE.2019.2913400. Epub 2019 Apr 26.

本文引用的文献

1
Graph convolutional networks: a comprehensive review.图卷积网络：全面综述。
Comput Soc Netw. 2019;6(1):11. doi: 10.1186/s40649-019-0069-y. Epub 2019 Nov 10.
2
Cross-Task Mental Workload Recognition Based on EEG Tensor Representation and Transfer Learning.基于 EEG 张量表示和迁移学习的跨任务精神工作负荷识别。
IEEE Trans Neural Syst Rehabil Eng. 2023;31:2632-2639. doi: 10.1109/TNSRE.2023.3277867. Epub 2023 Jun 14.
3
Everything is connected: Graph neural networks.万物皆相连：图神经网络。
Curr Opin Struct Biol. 2023 Apr;79:102538. doi: 10.1016/j.sbi.2023.102538. Epub 2023 Feb 9.
4
Cross-task cognitive workload recognition using a dynamic residual network with attention mechanism based on neurophysiological signals.基于神经生理信号的注意力机制动态残差网络的跨任务认知负荷识别。
Comput Methods Programs Biomed. 2023 Mar;230:107352. doi: 10.1016/j.cmpb.2023.107352. Epub 2023 Jan 13.
5
Classification of Drivers' Mental Workload Levels: Comparison of Machine Learning Methods Based on ECG and Infrared Thermal Signals.驾驶员心理负荷水平分类：基于 ECG 和红外热信号的机器学习方法比较。
Sensors (Basel). 2022 Sep 26;22(19):7300. doi: 10.3390/s22197300.
6
An Evaluation of the EEG Alpha-to-Theta and Theta-to-Alpha Band Ratios as Indexes of Mental Workload.脑电图α波与θ波以及θ波与α波频段比率作为心理负荷指标的评估
Front Neuroinform. 2022 May 16;16:861967. doi: 10.3389/fninf.2022.861967. eCollection 2022.
7
Classification of Contrasting Discrete Emotional States Indicated by EEG Based Graph Theoretical Network Measures.基于 EEG 的图论网络测量的对比离散情绪状态分类。
Neuroinformatics. 2022 Oct;20(4):863-877. doi: 10.1007/s12021-022-09579-2. Epub 2022 Mar 14.
8
EEG Based Dynamic Functional Connectivity Analysis in Mental Workload Tasks With Different Types of Information.基于脑电图的不同类型信息下脑力工作负荷任务的动态功能连接分析。
IEEE Trans Neural Syst Rehabil Eng. 2022;30:632-642. doi: 10.1109/TNSRE.2022.3156546. Epub 2022 Mar 21.
9
Cross-Task Cognitive Workload Recognition Based on EEG and Domain Adaptation.基于 EEG 和领域自适应的跨任务认知工作负荷识别。
IEEE Trans Neural Syst Rehabil Eng. 2022;30:50-60. doi: 10.1109/TNSRE.2022.3140456. Epub 2022 Jan 28.
10
Single feature spatio-temporal architecture for EEG Based cognitive load assessment.基于 EEG 的认知负荷评估的单特征时空架构。
Annu Int Conf IEEE Eng Med Biol Soc. 2021 Nov;2021:3717-3720. doi: 10.1109/EMBC46164.2021.9630107.