Abstract：To improve the generalization performance of cross-subject electroencephalogram(EEG) emotion recognition under limited annotation, a brain-region-prior self-supervised pretraining network(BSP-Net) is proposed. Brain-region topology is adopted as an inductive bias. Electrode channels are first modeled in groups by lightweight brain-region-aware channel attention. Then, two self-supervised tasks, masked brain-region modeling and brain-region prototype contrast, are jointly optimized on unlabeled EEG samples. Local inter-region dependencies and cross-subject brain-region representations are learned by the encoder. After pretraining, the encoder is connected to a lightweight capsule head for downstream emotion classification. Experiments on DEAP and DREAMER under a strict trial-level leakage-free leave-one-subject-out(LOSO) protocol show that the performance of cross-subject EEG emotion recognition under limited annotation conditions is improved by BSP-Net.
[1] 权学良,曾志刚,蒋建华,等.基于生理信号的情感计算研究综述.自动化学报, 2021, 47(8): 1769-1784. (QUAN X L, ZENG Z G, JIANG J H, et al. Physiological Signals Based Affective Computing: A Systematic Review. Acta Automatica Sinica, 2021, 47(8): 1769-1784.) [2] 李锦瑶,杜肖兵,朱志亮,等.脑电情绪识别的深度学习研究综述.软件学报, 2023, 34(1): 255-276. (LI J Y, DU X B, ZHU Z L, et al. Deep Learning for EEG-Based Emotion Recognition: A Survey. Journal of Software, 2023, 34(1): 255-276.) [3] 秦天鹏,生慧,岳路,等.脑电信号情绪识别研究综述.计算机工程与应用, 2023, 59(15): 38-54. (QIN T P, SHENG H, YUE L, et al. Review of Research on Emotion Recognition Based on EEG Signals. Computer Engineering and Applications, 2023, 59(15): 38-54.) [4] LAWHERN V J, SOLON A J, WAYTOWICH N R, et al. EEGNet: A Compact Convolutional Neural Network for EEG-Based Brain-Computer Interfaces. Journal of Neural Engineering, 2018, 15(5). DOI: 10.1088/1741-2552/aace8c. [5] SCHIRRMEISTER R T, SPRINGENBERG J T, FIEDERER L D J, et al. Deep Learning with Convolutional Neural Networks for EEG Decoding and Visualization. Human Brain Mapping, 2017, 38(11): 5391-5420. [6] TAO W, LI C, SONG R C, et al. EEG-Based Emotion Recognition via Channel-Wise Attention and Self Attention. IEEE Transactions on Affective Computing, 2023, 14(1): 382-393. [7] SONG T F, ZHENG W M, SONG P, et al. EEG Emotion Recognition Using Dynamical Graph Convolutional Neural Networks. IEEE Transactions on Affective Computing, 2020, 11(3): 532-541. [8] LI J C, PAN W J, HUANG H Y, et al. STGATE: Spatial-Temporal Graph Attention Network with a Transformer Encoder for EEG-Based Emotion Recognition. Frontiers in Human Neuroscience, 2023, 17. DOI: 10.3389/fnhum.2023.1169949. [9] DING Y, TONG C X, ZHANG S L, et al. EmT: A Novel Transformer for Generalized Cross-Subject EEG Emotion Recognition. IEEE Transactions on Neural Networks and Learning Systems, 2025, 36(6): 10381-10393. [10] FAN C H, WANG J Q, HUANG W, et al. Light-Weight Residual Convolution-Based Capsule Network for EEG Emotion Recognition. Advanced Engineering Informatics, 2024, 61. DOI: 10.1016/j.aei.2024.102522. [11] LIU S Q, WANG Z Y, AN Y L, et al. DA-CapsNet: A Multi-branch Capsule Network Based on Adversarial Domain Adaption for Cross-Subject EEG Emotion Recognition. Knowledge-Based Systems, 2024, 283. DOI: 10.1016/j.knosys.2023.111137. [12] WEI Y, LIU Y, LI C, et al. TC-Net: A Transformer Capsule Network for EEG-Based Emotion Recognition. Computers in Biology and Medicine, 2023, 152. DOI: 10.1016/j.compbiomed.2022.106463. [13] GANIN Y, LEMPITSKY V.Unsupervised Domain Adaptation by Backpropagation // Proc of the 32nd International Conference on Machine Learning. San Diego, USA: JMLR, 2015: 1180-1189. [14] LI Y, ZHENG W M, ZONG Y, et al. A Bi-hemisphere Domain Adversarial Neural Network Model for EEG Emotion Recognition. IEEE Transactions on Affective Computing, 2021, 12(2): 494-504. [15] CHEN H, JIN M, LI Z N, et al. MS-MDA: Multisource Marginal Distribution Adaptation for Cross-Subject and Cross-Session EEG Emotion Recognition. Frontiers in Neuroscience, 2021, 15. DOI: 10.3389/fnins.2021.778488. [16] MOHSENVAND M N, IZADI M R, MAES P.Contrastive Representation Learning for Electroencephalogram Classification // Proc of Machine Learning Research. San Diego, USA: JMLR, 2020: 238-253. [17] LI Y, CHEN J, LI F, et al. GMSS: Graph-Based Multi-task Self-Supervised Learning for EEG Emotion Recognition. IEEE Transactions on Affective Computing, 2023, 14(3): 2512-2525. [18] LIU Q L, YE W S, ZHANG L L, et al. EEG-SCMM: Soft Con-trastive Masked Modeling for Cross-Corpus EEG-Based Emotion Recognition // Proc of the 33rd ACM International Conference on Multimedia. New York, USA: ACM, 2025: 5834-5842. [19] KOELSTRA S, MÜHL C, SOLEYMANI M, et al. DEAP: A Database for Emotion Analysis Using Physiological Signals. IEEE Transactions on Affective Computing, 2012, 3(1): 18-31. [20] KATSIGIANNIS S, RAMZAN N.DREAMER: A Database for Emo-tion Recognition through EEG and ECG Signals from Wireless Low-Cost Off-The-Shelf Devices. IEEE Journal of Biomedical and Heal-th Informatics, 2018, 22(1): 98-107. [21] RUSSELL J A.A Circumplex Model of Affect. Journal of Persona-lity and Social Psychology, 1980, 39(6): 1161-1178. [22] HE K M, ZHANG X Y, REN S Q, et al. Deep Residual Learning for Image Recognition // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2016: 770-778. [23] WANG Q L, WU B G, ZHU P F, et al. ECA-Net: Efficient Cha-nnel Attention for Deep Convolutional Neural Networks // Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2020: 11531-11539. [24] HE K M, CHEN X L, XIE S N, et al. Masked Autoencoders Are Scalable Vision Learners // Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2022: 15979-15988. [25] CHEN T, KORNBLITH S, NOROUZI M, et al. A Simple Framework for Contrastive Learning of Visual Representations // Proc of the 37th International Conference on Machine Learning. San Diego, USA: JMLR, 2020: 1597-1607. [26] LI J N, ZHOU P, XIONG C M, et al. Prototypical Contrastive Learning of Unsupervised Representations[C/OL].[2026-04-21]. https://arxiv.org/pdf/2005.04966v4. [27] CARON M, MISRA I, MAIRAL J, et al. Unsupervised Learning of Visual Features by Contrasting Cluster Assignments // Proc of the 34th International Conference on Neural Information Processing Systems. Cambridge, USA: MIT Press, 2020: 9912-9924. [28] SABOUR S, FROSST N, HINTON G E. Dynamic Routing Between Capsules // Proc of the 31st International Conference on Neural Information Processing Systems. Cambridge, USA: MIT Press, 2017: 3859-3869. [29] DAVIDSON R J.Anterior Cerebral Asymmetry and the Nature of Emotion. Brain and Cognition, 1992, 20(1): 125-151. [30] ZHENG W L, LU B L.Investigating Critical Frequency Bands and Channels for EEG-Based Emotion Recognition with Deep Neural Networks. IEEE Transactions on Autonomous Mental Development, 2015, 7(3): 162-175.
2026, Vol. 39 
