Multimodal fusion of Electroencephalography (EEG) and functional Near-Infrared Spectroscopy (fNIRS) has shown great promise in Brain-Computer Interface (BCI) tasks. However, due to differences in physical mechanisms, temporal dynamics, and semantic representations between the two modalities, the fusion process faces significant challenges such as heterogeneity and temporal misalignment. To address this, we propose a cross-modal decoupled deformable distillation (CMD$^{3}$) method, which aims to achieve flexible, efficient, and interpretable EEG-fNIRS fusion learning. CMD$^{3}$first decouples the feature representations of each modality into modality-independent and modality-specific spaces to separately model commonality and complementary information. A deformable feature extraction network is then designed to process shared and specific features individually, enabling cross-modal temporal alignment via predicted dynamic offsets, thereby mitigating response delays between modalities. Furthermore, to facilitate inter-modal knowledge transfer, we construct a dual-space graph distillation module to explicitly migrate semantic information across modalities, with learnable edge weights used to adaptively regulate the distillation strength. CMD$^{3}$is systematically evaluated on public datasets covering emotion recognition and motor imagery tasks. Experimental results demonstrate that CMD$^{3}$consistently outperforms existing fusion approaches in classification performance. Offset visualization further reveals physiologically meaningful temporal attention patterns learned by the model, validating the effectiveness and explainability of the proposed method.

Electroencephalogram (EEG) has been widely used for emotion recognition due to its portability and high temporal resolution. It makes success in subject-dependent scenario but faces significant challenges in cross-subject emotion recognition because of non-stationarity of EEG and individual differences. Most previous studies treat all individuals as a single source domain for transferring emotional knowledge, which may introduce irrelevant information and lead to negative transfer. Besides, there is a potential risk that some important information of common emotional features might be ignored. To deal with the issues, we propose a framework called hybrid source selection fusion domain-invariant attention (HSSFDA) for cross-subject emotion recognition. First, source domains are selected by leveraging local and global similarity for knowledge transfer. Then, a specialized attention mechanism is employed to focus on important emotional information extracted from the domain-invariant features. Finally, domain-invariant and domain-specific features are fused to enhance emotion recognition performance. To evaluate the proposed method, experiments are conducted on several public datasets including SEED, SEED_IV, DREAMER and DEAP. The results demonstrate that HSSFDA achieves accuracies of 85.07 %, 72.11 %, 62.36 %, 77.17 %, 58.51 %, and 63.55 % on SEED, SEED_IV, valence and arousal of DREAMER, and valence and arousal of DEAP datasets, respectively, demonstrating competitive performance compared to popular and state-of-the-art methods. Furthermore, we apply the HSSFDA to a self-recorded dataset collected by self-developed three-channel device and validate its effectiveness in practical applications. In conclusion, HSSFDA is a feasible method for cross-subject emotion recognition and has the potential to broaden the application of EEG in the field of affective computing.

Event-driven neuromorphic computing involves sparse and asynchronous signal activity, which leads to irregular computation patterns and fine-grained concurrency. As a result, processing architectures need to support both high parallelism and energy efficiency. Among existing architectural solutions, superscalar designs exhibit significant potential for addressing high parallelism demands. However, conventional superscalar processors, which rely on synchronous circuits, maintain high-frequency clocking at all times, leading to substantial power inefficiency in sparse computation scenarios. To address this issue, we propose an asynchronous superscalar architecture that replaces global clocking with fully local handshake-based control, implemented using a bundled-data asynchronous protocol. The design supports decoding of up to 64 scalar instructions per cycle and implements the RISC-V RV32IMC instruction set. A prototype was fabricated using a 110 nm complementary metal oxide semiconductor (CMOS) process and was evaluated through post-layout simulation. Operating at 1.2 V, the processor delivers a peak INT8 throughput of 669.4 GOPS, with a static power consumption of 421 mW.

Cross-subject emotion recognition remains a challenge due to inter-subject variability, which limits the generalized ability of models to unseen subjects. Existing studies commonly rely on tasking-state EEG data from the target subject for adaptation, which requires additional emotion-elicitation experiments and limits practical deployment. Motivated by findings that resting-state EEG can reflect individual-specific neural characteristics, this study proposes a Discriminative Knowledge Fuzzy Transfer Learning Guided by Resting-state EEG (DKFTL-R) for cross-subject emotion recognition without requiring tasking-state EEG data from the target subject. First, resting-state EEG is leveraged to characterize subject-specific neural signatures, by which source-domain selection is informed. Second, an emotion-style projection alignment module is introduced, in which discriminative emotional knowledge and domain-specific style are integrated via adaptive weighting so that a more transferable representation is obtained. Finally, a Takagi-Sugeno-Kang fuzzy classifier is employed to perform fuzzy inference on the transferable representation. Experiments are conducted on DEAP and DENS datasets, where accuracies of 58.79%, 55.89%, 62.91%, and 60.42% are achieved, respectively, demonstrating competitive performance compared with popular and recent baseline methods. To evaluate practical applicability and deployability, the proposed method is conducted on a self-constructed emotion EEG dataset (BHE-EMO), and it achieves 67.00% accuracy for two-class classification and 44.92% for three-class classification tasks, further demonstrating its effectiveness and engineering potential in real-world settings. In conclusion, we propose a new perspective on cross-subject emotion recognition by integrating resting-state EEG information with fuzzy modeling. This study also introduces a new calibration paradigm for affective brain-computer interface systems.

Anxiety disorders (AD) are prevalent psychiatric conditions that profoundly impact adolescent neural development. Abnormal delta-beta cross-frequency coupling (CFC) has been identified as a key electrophysiological marker of altered neural dynamics in individuals with AD. However, most existing studies focus on static analysis within restricted brain regions and predefined frequency bands, which limits the understanding of large-scale dynamic neural communication. Therefore, we propose a novel cross-frequency coupling directed brain network (CFCDBN) framework, which integrates personalized CFC estimation and causal information flow modeling to capture the dynamic interactions of the brain network in AD. Personalized CFC significantly improves the precise representation of AD-related neural dynamics by adaptive frequency band division and individualized oscillation feature extraction, overcoming the limitations of traditional CFC methods. The analysis reveals significant delta-beta coupling abnormalities in the left hemisphere of AD, accompanied by disrupted directional pathways involving the thalamus, precuneus, and insula. These findings suggest impaired emotional and cognitive communication from the subcortical to cortical regions. To validate the efficacy of CFCDBN in distinguishing AD patients from healthy individuals, we developed a direction-aware graph neural network (DA-GNN) model that uses CFCDBN representations as input to capture dynamic neural patterns in causal brain connectivity. Experimental results show that the model consistently outperforms traditional machine learning methods and undirected GNN baselines in automatic AD identification, achieving a classification accuracy of 77.8%, and confirming the value of CFCDBN as a robust biomarker for AD-related network dysfunction. These findings not only deepen our understanding of the neural dynamics underlying AD, but also lay the foundation for personalized and mechanism-driven neuromodulation strategies. The core implementation of the CFCDBN framework is available on GitHub: https://github.com/wdxcjnb6/CFCDBN.

In noninvasive brain-computer interfaces (BCIs), EEG analysis plays a critical role, with neural networks serving as a cornerstone for signal decoding. Existing neural network approaches for EEG signal recognition require extensive manual design and hyperparameter tuning, leading to inefficiencies and making them impractical for embedded devices due to their large model size. To address these limitations, we propose mixed-level differentiable and hardware-aware neural architecture search (MDH-NAS), a framework that automatically generates lightweight neural networks tailored for EEG classification. Unlike traditional DARTS methods, MDH-NAS employs a hybrid optimization strategy that balances global and local search spaces, thereby accelerating and refining architecture discovery. It introduces explicit size constraints during the search process to ensure deployability on embedded devices. MDH-NAS demonstrates autonomous generation of architectures for tasks such as motor imagery (MI) and depression recognition, achieving 87.80% accuracy on the BCI-IV dataset and 90.09% on the MODMA dataset. When deployed on the EAIDK-610 board across heterogeneous tasks, it attains 85.37% accuracy on the EEG Motor Movement/Imagery dataset. This method reduces architecture discovery time by 89% and enhances prediction accuracy by 8.70% compared to baseline methods, highlighting its potential for scalable EEG analysis and real-world embedded deployment.

Emotions regulation play an important role in human behavior, but exhibit considerable heterogeneity among individuals, which attenuates the generalization ability of emotion models. In this work, we aim to achieve robust emotion prediction through efficient disentanglement of affective semantic representations. In detail, the data generation mechanism behind observations from different perspectives is causally set, where latent variables that relate to emotion are explicitly separate into three parts: the intrinsic-related part, the extrinsic-related part, and the spurious-related part. Affective semantic features consist of the first two parts, with the understanding that spurious latent variables generate the inherent biases in the data. Furthermore, a variational autoencoder with a reformulated objective function is proposed to learn such disentangled latent variables, and only adopts semantic representations to perform the final classification task, avoiding the interference of spurious variables. In addition, for electroencephalography (EEG) data used in this article, a space-frequency mapping method is introduced to improve information utilization. Comprehensive experiments on popular emotion datasets show that the proposed method can achieve competitive intersubject generalization performance. Our results highlight the potential of efficient latent representation disentanglement in addressing the complexity challenges of emotion recognition.

Emotion inhibitory control is a key executive function of the human brain, which regulates behavior by suppressing inappropriate responses. It plays an integral part in alleviating negative emotions, improving mood, and preventing depression. Transcutaneous auricular vagus nerve stimulation (taVNS) has been proved to enhance behavioral control, potentially suppressing negative emotions or facilitating their reduction in healthy individuals. However, the neurocomputational mechanisms underlying taVNS-induced neuroenhancement remain unclear. In this work, a portable electroencephalography (EEG) acquisition and stimulation device is designed to collect eight-channel EEG signals and deliver taVNS to both sides of ears. Then, we design a protocol that successfully induced negative emotions in healthy subjects. Next, we conduct a sham-controlled experiment, involving 28 healthy subjects, to explore the changes in EEG of negative emotions under taVNS. Finally, we primarily analyze the power spectrum density (PSD) of EEG signals and the functional connectivity network of the brain, based on the phase locking value (PLV), to assess the effect of taVNS on neural activity induced by negative emotions. The results of the experiment reveal that taVNS is a promising method for enhancing emotional inhibitory control by reducing PSD in the alpha band and enhancing PLV within prefrontal inhibitory control networks. In addition, differences in graph theory parameters between the Sham and taVNS conditions indicate that taVNS helps regulate negative emotions. In conclusion, this study demonstrates that taVNS enhances inhibitory control and reveals its neurocomputational mechanisms of EEG in healthy individuals during the development of negative emotions. And results indicate that taVNS could serve as a promising neuromodulation therapy for psychiatric disorders and individuals with depression or emotional distress.

The development of affective computing and medical electronic technologies has led to the emergence of Artificial Intelligence (AI)-based methods for the early detection of depression. However, previous studies have often overlooked the necessity for the AI-assisted diagnosis system to be wearable and accessible in practical scenarios for depression recognition. In this work, we present an on-board executable multi-feature transfer-enhanced fusion model for our custom-designed wearable three-lead Electroencephalogram (EEG) sensor, based on EEG data collected from 73 depressed patients and 108 healthy controls. Experimental results show that the proposed model exhibits low-computational complexity (65.0 K parameters), promising Floating-Point Operations (FLOPs) performance (25.6 M), real-time processing (1.5 s/execution), and low power consumption (320.8 mW). Furthermore, it requires only 202.0 KB of Random Access Memory (RAM) and 279.6 KB of Read-Only Memory (ROM) when deployed on the EEG sensor. Despite its low computational and spatial complexity, the model achieves a notable classification accuracy of 95.2%, specificity of 94.0%, and sensitivity of 96.9% under independent test conditions. These results underscore the potential of deploying the model on the wearable three-lead EEG sensor for assisting in the diagnosis of depression.

Phonocardiogram classification methods using deep neural networks have been widely applied to the early detection of cardiovascular diseases recently.Despite their excellent recognition rate, the sizeable computational complexity limits their further development.Nowadays, knowledge distillation (KD) is an established paradigm for model compression.While current research on multi-teacher KD has shown potential to impart more comprehensive knowledge to the student than single-teacher KD, this approach is not suitable for all scenarios.This paper proposes a novel KD strategy to realise an adaptive multi-teacher instruction mechanism.We design a teacher selection strategy called voting network to tell the contribution of different teachers on each distillation points, so that the student can choose the useful information and renounce the redundant one.An evaluation demonstrates that our method reaches excellent accuracy (92.8 %) while maintaining a low computational complexity (0.7 M).

With great pride and enthusiasm, we present the inaugural edition of IEEE Transactions on Computational Social Systems (TCSS) for 2024. Reflecting on the year gone by, 2023 stands as a hallmark of academic excellence and prolific output, wherein our journal has successfully disseminated a substantial volume of scholarly work—301 articles encompassing approximately 3600 pages, distributed across six distinct issues.

Currently, the integration of artificial intelligence (AI) techniques with multimodal physiological signals represents a pivotal approach to detect affective disorders (ADs). With the increasing complexity and diversity of physiological signal modalities, researchers have introduced various AI methods using multimodal physiological signals to improve model classification performance and explainability to increase trust and facilitate clinical adoption. Among these methods, spiking neural networks (SNNs) stand out as a promising avenue due to their alignment with the operating principles of the human brain, robust biological explainability, and adeptness in processing spatial–temporal information in an efficient event-driven manner with low power consumption. Furthermore, the emergence of neuromorphic computing (NC) chips based on SNNs has greatly bolstered the field of NC, enabling effective support for objective, pervasive, and wearable AI-assisted medical diagnostic devices for ADs and other diseases. This article presents a review of recent achievements in multimodal AD detection and points out the associated challenges in utilizing multimodal physiological signals and NC based on SNNs for AD detection. Building upon this foundation, we give perspectives on future work. The intended readership for this review consists of researchers in the fields of cognitive computing, computational psychophysiology, affective computing, NC, and brain-inspired computing. We hope that this survey not only garners increased attention from the scientific community but also serves as a valuable guide for future studies in this field.

Welcome to the second issue of IEEE Transactions on Computational Social Systems (TCSS) in 2022. In this issue, we are going to present 25 regular articles. After the “scanning the issue,” I would like to share some of my opinions and perspectives on the fundamentals of computational psychophysiology: theory and methodology.