Cogn Neurodyn [JOURNAL]
Sun
200 papers
Dynamical analysis of a mean firing rate model in Parkinson's disease.
Xu W, Du Y, Pan X … +2 more , Xu X, Wang Y
Cogn Neurodyn · 2026 Dec · PMID 42028460 · Full text
The generation of pathological oscillations in Parkinson's disease(PD) is closely linked to the synchronous evolution of neuronal populations within the basal ganglia. Given the advantages of average firing rate models i...
The generation of pathological oscillations in Parkinson's disease(PD) is closely linked to the synchronous evolution of neuronal populations within the basal ganglia. Given the advantages of average firing rate models in describing large scale neural dynamics, this paper proposes an extended model that enhances biological plausibility, building upon a previous basal ganglia circuit model. We incorporated the cortex, thalamus and Pedunculopontine Nucleus(PPN) into the basal ganglia model.We calculated the phase locked value and differences in β-band energy proportions across different nuclei, and conducted lesion simulations to validate the biological plausibility of the model. We also introduced two distinct types of dopaminergic parameters into the model to simulate their effects on synapses and, consequently, on network oscillations; the results indicate that relative changes in these parameters may be more likely to induce oscillations than changes in any single value alone. We also investigated the influence of time constants on network activity and found that, whether under normal conditions or in a state of mild dopamine deficiency, the population response rate of the PPN affects the magnitude of oscillation frequencies within the basal ganglia. Furthermore, we conducted a dynamical analysis of synaptic connection delays and weights, discovering that they can induce a transition of the system from a normal state to a pathological oscillatory state. Finally, we performed Morris and Sobol sensitivity analyses to quantitatively assess the influence of various network parameters on oscillatory activity. Through this analysis, we identified the connection strength between the cortex and the thalamic basal nuclei, the bidirectional connection strength of the subthalamic nucleus (STN)-the external segment of the globus pallidus (GPe) loop, and the connection delays in the basal ganglia, the cortico-thalamic system, and the PPN; these parameters play a crucial role in the generation of pathological activity and the regulation of oscillation frequency. These findings provide theoretical guidance for a deeper understanding of the underlying mechanisms of Parkinson's disease and for the alleviation of PD symptoms.
Exploring electromagnetic induction and astrocyte influence in excitatory-inhibitory coupling neuron network.
Gao Z, Feng P
Cogn Neurodyn · 2026 Dec · PMID 42028459 · Full text
This study investigates how electromagnetic induction and astrocytic modulation jointly influence firing dynamics and synchronization in excitatory-inhibitory neuronal networks. Using a computational model that integrate...
This study investigates how electromagnetic induction and astrocytic modulation jointly influence firing dynamics and synchronization in excitatory-inhibitory neuronal networks. Using a computational model that integrates pyramidal neurons, interneurons, and astrocytes with memristor-based electromagnetic feedback and calcium-dependent signaling, we demonstrate that electromagnetic induction generally suppresses neuronal firing, but this effect can be bidirectionally modulated by astrocytic feedback depending on intracellular calcium levels. Key findings reveal the non-linear dependence of neuronal responses on astrocytic calcium thresholds and [Formula: see text] dynamics; the distinct roles of excitatory, inhibitory, and astrocytic coupling in regulating network synchrony and chimera states; and the existence of parameter regimes where astrocytic feedback is overridden under strong electromagnetic induction. These results highlight the critical interplay between electromagnetic and glial mechanisms in shaping network activity, offering insights for models of neural synchronization and potential therapeutic strategies for epilepsy and related disorders.
A MBPAF-memristive Hopfield neural network and its application in image encryption.
Deng S, Jin J, Li Z … +2 more , Chen C, Yu F
Cogn Neurodyn · 2026 Dec · PMID 42007437 · Full text
The complexity of neural dynamics heavily depends on the nonlinear activation functions, and a mixed-bipower activation function (MBPAF) with adjustable parameters is designed for the memristive Hopfield neural network (...
The complexity of neural dynamics heavily depends on the nonlinear activation functions, and a mixed-bipower activation function (MBPAF) with adjustable parameters is designed for the memristive Hopfield neural network (MHNN) to generate complex hyper-chaotic behaviors. Based on the designed MBPAF, a novel MBPAF-memristive Hopfield neural network (MBPAF-MHNN) model is proposed. The complex dynamics of the proposed MBPAF-MHNN model are validated through numerical analyses and further verified via FPGA implementation. Finally, a robust image encryption scheme is designed based on the MBPAF-MHNN model, featuring a plaintext-related "Diffusion-Permutation-Diffusion" architecture with DNA-based operations.
Thermodynamic neural networks and intersection theory: an ontological hypothesis of emergent intelligence.
Kulyk O
Cogn Neurodyn · 2026 Dec · PMID 42007436 · Full text
UNLABELLED: This article proposes a new ontological framework for describing cognitive processes, grounded in and an . At the micro-level, we show that neurons and neuronal populations function as thermodynamic systems...
UNLABELLED: This article proposes a new ontological framework for describing cognitive processes, grounded in and an . At the micro-level, we show that neurons and neuronal populations function as thermodynamic systems that reside in regimes of fluctuations, relaxations, and entropic transitions. At the macro-level, these processes manifest as ordered structures-the topologies of truths and their intersections-that jointly shape the cognitive landscape. We introduce the notion of as the minimal quantum of reality, as well as the (individual or collective) who performs (partitioning), dividing the universal space of truths into known and unknown zones. On this basis, we formulate the , which characterizes the evolving balance between the known and the unknown over time. We further show that synchronization of individual agents via the operator of temporal velocity [Formula: see text] yields a whose entropy is defined as a coherent integration of the entropies of its constituent agents. This construction scales from sensory physiology (receptors as individual agents) to the collective cognition of humanity as a whole. At the concluding level, we introduce the key notion of (splitting)-the multiplicity of individual reconstructions of the same ontological truth-which explains why collective cognition is not a mere sum of private representations but acquires the quality of emergent integration. Taken together, the article demonstrates that cognitive topology is an emergent reflection of thermodynamic dynamics, and that a universal scheme of intersections provides a unified ontological language for processes ranging from the micro-physiology of neurons to macro-level epistemology and the history of civilization. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s11571-026-10438-y.
Glutamate receptors as regulators of synaptic plasticity in neuropsychiatric disorders: pathological insight and translational perspectives.
Srivastava J, Yadav J, Dubey N
Cogn Neurodyn · 2026 Dec · PMID 42007435 · Full text
Glutamate receptors (GluRs) are central regulators of synaptic plasticity, a fundamental physiological process underlying learning, memory, and adaptive behaviour. Synaptic plasticity is a dynamic process in which synaps...
Glutamate receptors (GluRs) are central regulators of synaptic plasticity, a fundamental physiological process underlying learning, memory, and adaptive behaviour. Synaptic plasticity is a dynamic process in which synapses undergo activity-dependent changes between the communicating neurons, forming the cellular basis of cognition. Dysfunctional synaptic plasticity can exacerbate neurological symptoms, thus underlying a broad spectrum of neuropsychiatric disorders (NPDs), including schizophrenia, major depressive disorder and bipolar disorder. Advancements in the field have highlighted that functional impairment in the GluR have heightened excitatory neurotransmission, thus disrupting synaptic transmission, leading to excitotoxicity and neuronal cell defects. These dysfunctions constitute the mechanistic basis of NPDs has prompted researchers to target various GluR families, namely ionotropic glutamate receptors (iGluRs) and metabotropic glutamate receptors (mGluRs). The iGluRs are further subclassified into N-methyl-D-aspartate receptor (NMDAR), α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR), kainate receptor (KAR), and delta glutamate receptor (GluD). Therapeutic modulation of these GluR is aimed at restoring the homeostatic balance between glutamatergic signalling and synaptic functions. This review provides a comprehensive overview of how dysregulation of these GluRs results in maladaptive synaptic plasticity, thus contributing to neuropsychiatric conditions as studied in in vitro, human, and animal models. Additionally, the review features the current therapeutic strategies targeting GluRs modulators in the amelioration of disease phenotype. By integrating the molecular and clinical insights, we advocate that GluRs represents a promising therapeutic target for the modulation of aberrant neural plasticity in NPDs.
Ternary quantitative neural network implemented with tri-valued memristors.
Wang X, Li Z, Li XL … +4 more , Li H, Liu J, Iu HHC, Kang SM
Cogn Neurodyn · 2026 Dec · PMID 41971723 · Full text
As a nano-scale, non-volatile device compatible with CMOS technologies, the memristor is well suited for in-memory computing and provides a new venue for accelerating neural network computing. However, memristor-based ne...
As a nano-scale, non-volatile device compatible with CMOS technologies, the memristor is well suited for in-memory computing and provides a new venue for accelerating neural network computing. However, memristor-based neural networks are typically based on continuous-type memristors or binary memristors. Continuous-type memristors require precise voltage amplitude and duty cycle for setting resistance, which increases the difficulty of practical applications, whereas binary memristors provide limited weight accuracy due to their restricted two memrisitance states. In contrast, tri-valued memristors feature three discrete resistance values, enabling them to represent a broader range of weight states and allowing a more straightforward resistance setting mechanism. This leads to a better balance between the hardware feasibility and the weight representation capacity of the artificial neural network. This paper proposes a ternary neural network design scheme based on tri-valued memristors. The design scheme consists of a fully hardware-based forward computing circuit using a tri-valued memristor crossbar array, a tri-valued memristance-based weight setting method, a weight updating algorithm for the ternary quantized neural network, an activation function circuit, and a winner-take-all (WTA) circuit. This design features the natural synaptic characteristics of memristors and the computational advantages of the quantized neural network, facilitating a hardware platform and a feasible implementation scheme for lightweight all-hardware neural circuit design. The scheme verified with LTSpice circuit simulations enables correct recognition of test characters '', '', and ''. Furthermore, it is extended to the LeNet-5 network on the MemTorch platform, achieving a recognition accuracy of 98.47% in the MNIST benchmark test.
A dual-state generic memristor-neuron coupled system for pain modulation and chaotic dynamics research.
Xu Y, He X, Xu X … +2 more , Banerjee S, Mou J
Cogn Neurodyn · 2026 Dec · PMID 41971722 · Full text
The nonlinear dynamics of neuronal firing play a key role in understanding modulation mechanisms in excitable systems. In this work, a dual-state generic memristor is proposed, in which two internal variables represent a...
The nonlinear dynamics of neuronal firing play a key role in understanding modulation mechanisms in excitable systems. In this work, a dual-state generic memristor is proposed, in which two internal variables represent antagonistic facilitation-like and inhibition-like drives within a phenomenological framework. Coupling this device with the Hindmarsh-Rose neuron yields a four-dimensional DSGM-HR system with bidirectional feedback. Equilibrium analysis, bifurcation diagrams, and Lyapunov exponents reveal transitions among periodic and chaotic firing regimes, as well as multistability under varying coupling strength and external input. A DSP implementation further confirms hardware reproducibility. Pain-related terminology is adopted only as a conceptual narrative to interpret antagonistic modulation, rather than to quantify subjective pain intensity.
JDA-RSDB: a multimodal domain adaptation method for cross-session emotion recognition from EEG and eye movement signals.
Jiménez-Guarneros M, Grande-Barreto J, Fuentes-Pineda G
Cogn Neurodyn · 2026 Dec · PMID 41971721 · Full text
Multimodal emotion recognition has shown growing interest in affective computing, as combining Electroencephalogram (EEG) and eye movement (EM) signals enables the capture of complex emotional processes. However, EEG and...
Multimodal emotion recognition has shown growing interest in affective computing, as combining Electroencephalogram (EEG) and eye movement (EM) signals enables the capture of complex emotional processes. However, EEG and EM signals are exposed to joint distribution differences across different days and recorded sessions, reducing the recognition performance. Currently, domain adaptation has been developed to address such distribution differences. Unfortunately, existing domain adaptation solutions still show suboptimal classification results, since ambiguous and non-discriminative decision boundaries are still learned during distribution matching. This paper presents Joint Distribution Alignment with Refined and Separable Decision Boundaries (JDA-RSDB), a multimodal domain adaptation method for cross-session emotion recognition from EEG and EM signals. Our proposed method assumes that a more discriminative feature representation must be ensured on new sessions during joint distribution matching. For this, JDA-RSDB produces similar marginal and conditional distributions between domains, first aligning feature statistics at modality and domain levels, and then, motivating consistent similarity between fused samples from different domains that produce the same class prediction. Simultaneously, this similarity is enhanced by learning a separable feature space on target data, placing decision boundaries on low-density regions. More importantly, decision boundaries are refined by achieving an agreement between target predictions from a principal classifier and those from an auxiliary classifier. Experiments were conducted on three public datasets, SEED-GER, SEED-IV, and SEED-V, in a cross-session setting. The proposed framework achieves an average accuracy of 83.33%, 80.89%, and 75.17% across the three available sessions on SEED-GER, SEED-IV, and SEED-V, outperforming state-of-the-art solutions.
Domain generalized feature embedded learning for calibration-free event-related potentials recognition.
Luo TJ
Cogn Neurodyn · 2026 Dec · PMID 41971720 · Full text
Event-related potentials (ERPs) play an important role for building EEG-based brain computer interfaces (BCIs). However, due to the complex and varied spatio-temporal characteristics of ERPs across subjects, data distrib...
Event-related potentials (ERPs) play an important role for building EEG-based brain computer interfaces (BCIs). However, due to the complex and varied spatio-temporal characteristics of ERPs across subjects, data distribution across subjects a very important issue to be solved for constructing calibration-free BCIs. To achieve calibration-free ERPs recognition, we propose a omain eneralized eature mbedded earning (DGFEL) method. First, we align the ERPs of each existed subject based on covariance centroids. Then, we enhanced the aligned samples based on xDAWN filter and extract spatio-temporal features. Finally, the spatio-temporal features are further generalized by the decomposed adversarial loss, and we construct a neural network embedding backbone to implement features generalization across subjects. The proposed method has been systematically validated on two benchmark EEG-based ERP datasets, and its classification performance surpasses several state-of-the-art methods as well as deep learning models. Moreover, it effectively captures robust features from existed source subjects, and can be generalized to new subjects without accessing target ERP samples. Our method therefore provides a novel selection to construct calibration-free ERP-BCIs.
Ensemble of transformers for depression emotion classification.
Kasap F, İlhan Omurca S, Ekinci E
Cogn Neurodyn · 2026 Dec · PMID 41971719 · Full text
People act on their emotions even in the most rational decision-making mechanisms in their lives. Emotions are powerful motivators that profoundly influence human behavior and social interactions. Human emotions tend to...
People act on their emotions even in the most rational decision-making mechanisms in their lives. Emotions are powerful motivators that profoundly influence human behavior and social interactions. Human emotions tend to co-occur. Analyzing emotions with this co-occurrence in mind may lead to more accurate insights for addressing various mental health issues. A good example of this is the analysis of depression, which often involves a complex interplay of multiple interrelated emotions rather than a single, isolated feeling. This paper provides a comprehensive analysis of co-occurring emotions in depression by using artificial intelligence methods. We have proposed a transformer-based ensemble model that predicts multiple emotional tendencies associated with depression based on the public DepressionEmo dataset of user posts associated with depression. DistilBERT, RoBERTa, Mental-BERT, Mental-RoBERTa, and DeBERTa are used as pre-trained transformers. The heterogeneous ensemble learning architecture developed using stacking and majority voting methods improves the individual prediction performance of the transformer architectures. Our study is the first to apply the transformer ensemble to the DepressionEmo dataset to identify multiple emotions in an individual's textual psychological posts. Experimental results demonstrate that ensemble-based approaches provide more consistently improved performance compared to individual transformers, particularly in terms of macro-averaged F1 scores under conditions of class imbalance. Among ensemble learning approaches, the highest performance was achieved with Stacking-FFNN, which achieved 0.8121. These ensemble approaches consistently outperformed the strongest individual model, demonstrating the effectiveness of ensemble learning in improving depression emotion classification.
Ethical risks and considerations of brain-controlled and neuromodulation technologies.
Huang K, Yang H, Zhu S … +7 more , Chen Y, Li T, Zhao L, Gong A, Nan W, Xu J, Fu Y
Cogn Neurodyn · 2026 Dec · PMID 41940265 · Full text
Brain-controlled technology (BCT), centered on brain-computer interfaces (BCI), acquires and decodes neural signals to convert subjective intentions into control commands for external devices, establishing an intention o...
Brain-controlled technology (BCT), centered on brain-computer interfaces (BCI), acquires and decodes neural signals to convert subjective intentions into control commands for external devices, establishing an intention output loop. In contrast, neuromodulation technology applies external physical stimuli to the central nervous system to regulate neuronal excitability and brain network states, achieving energy input for functional modulation and therapeutic purposes. The inherent differences in mechanisms and application goals determine that the ethical risk profiles and governance priorities of these two technologies cannot be conflated. Current public communication is characterized by terminology misuse and concept generalization, notably the misinterpretation of neuromodulation as controlling the brain. In response to the resulting ethical anxiety caused by capability extrapolation, this paper first clarifies the functional positioning of both technologies. Subsequently, a three-dimensional assessment model based on reality, reversibility, and technological dependence is constructed to map a stratified ethical risk landscape. The analysis reveals a significant asymmetry in risk distribution: risks of BCT are primarily concentrated on neural privacy leakage and responsibility attribution dilemmas within the intention decoding process, whereas risks of neuromodulation are deeply embedded in the potential erosion of personal identity and subject autonomy induced by external stimuli. To address institutional gaps in the current regulatory system regarding consumer-grade devices and long-term effects, this paper proposes a differentiated tiered governance strategy. It advocates establishing terminology demystification and conceptual rectification as the frontline defense for risk governance. On this basis, the strategy enforces physical defense mechanisms such as hardware fusing and parameter safety windows on the technical side, and strengthens data desensitization and algorithmic accountability on the data side. Ultimately, a multi-subject synergistic governance mechanism covering the full lifecycle from research and development and clinical trials to social application is constructed to provide institutional support for responsible innovation in neurotechnology.
EEG-based brain functional connectivity dynamics in manual and video-based car-following observation among young drivers.
Li P, Qi G, Zhao S … +2 more , Huang A, Guan W
Cogn Neurodyn · 2026 Dec · PMID 41940264 · Full text
UNLABELLED: Understanding the neurophysiological mechanisms underlying driving behavior in young drivers is essential for improving cognitive-aware driver assistance and vehicle-human interaction systems. This study syst...
UNLABELLED: Understanding the neurophysiological mechanisms underlying driving behavior in young drivers is essential for improving cognitive-aware driver assistance and vehicle-human interaction systems. This study systematically examines EEG dynamics and functional brain network reconfigurations across both manual and video-based car-following observation, providing a neurophysiological framework for differentiating driving modes among young adult drivers. EEG characteristics were analyzed under three car-following strategies-aggressive, conservative, and personalized-implemented within a simulated driving environment, to capture the variability of cognitive engagement during distinct control demands. Key findings reveal that power spectral density (PSD) in the θ, β, and γ bands, combined with brain functional connectivity (BFN) measures, effectively characterizes workload-related modulation and attentional resources across driving conditions. A novel computational framework integrating Time-Frequency Common Mutual Information (TFCMI) features with a Parallel Compact Convolutional Neural Network (PCNet) achieved an average classification accuracy of 85.26%, surpassing traditional single-modality approaches. Neurotopographic results further indicate context-dependent functional specialization: frontal regions showed stronger activation and connectivity during manual control, while occipital regions exhibited enhanced synchronization during video-based car-following observation tasks. Collectively, these findings advance the understanding of driving-related cognitive processes in young drivers and provide neuroergonomic insights for designing adaptive human-machine interfaces in future intelligent transportation systems. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s11571-026-10442-2.
Stochastic neural dynamic modeling and analysis under environmental noise for exploring the production of K-complexes.
Wang W, Song J, Zan W … +3 more , Wang B, Li Y, Zhang R
Cogn Neurodyn · 2026 Dec · PMID 41940263 · Full text
UNLABELLED: K-complexes (KCs), sleep-specific neuroprotective waveforms, demonstrate significant modulation by environmental noise (EN). However, the principles governing how EN modulates KCs occurrence remain poorly und...
UNLABELLED: K-complexes (KCs), sleep-specific neuroprotective waveforms, demonstrate significant modulation by environmental noise (EN). However, the principles governing how EN modulates KCs occurrence remain poorly understood. To address this gap, we develop a stochastic neural dynamic model incorporating EN (SNDM-KCs) and explore the modulation effects of EN on KCs from the perspective of stochastic dynamics. The Gaussian colored noise (GCN) is first applied to model EN and introduced into the deterministic Costa neural mass model to build the SNDM-KCs. Next, bifurcation analysis is conducted to demonstrate that the prerequisite for occurrence of KCs corresponds to a large-amplitude departure from a stable equilibrium induced by GCN in the dynamic system. Subsequently, we study the impact of GCN on KCs by integrating SNDM-KCs with defined two metrics to quantitatively measure the elicitation variation of KCs. Numerical simulations suggest that both KCs occurrence probability and rate increase with noise intensity and correlation rate [Formula: see text] of GCN. Meanwhile, building on stochastic escape theory, we establish the relationship between model behaviour and stochastic escape metrics: first escape probability (FEP) and the mean first exit time (MFET), to investigate how EN modulates KCs through the lens of stochastic dynamics. The results demonstrate that as the escape probability of the system rises, the occurrence probability of KC increases accordingly. Meanwhile, a shorter time to escape from the safe domain indicates a faster occurrence rate of KCs. Our work provides a novel dynamical insight for investigating the principles governing how EN modulates KCs occurrence. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s11571-026-10440-4.
Bmssnet: a multi-scale feature and efficient spatial attention fusion model for early recognition of Alzheimer's disease.
Zhou J, Zhang Z, Li X … +4 more , Wan J, Zhang C, Chen M, Liu C
Cogn Neurodyn · 2026 Dec · PMID 41883492 · Full text
Integrating structural magnetic resonance imaging (sMRI) with deep learning techniques is one of the important research directions for automated diagnosis of Alzheimer's disease (AD). Among these, Convolutional Neural Ne...
Integrating structural magnetic resonance imaging (sMRI) with deep learning techniques is one of the important research directions for automated diagnosis of Alzheimer's disease (AD). Among these, Convolutional Neural Networks (CNNs) have been widely adopted as a mainstream approach due to their powerful feature extraction capabilities. However, existing convolutional neural network (CNN)-based voxel models with excellent performance are typically constrained to a single spatial scale. This limitation hinders the effective capture of complex, distributed brain atrophy features of AD and often results in insufficient model interpretability. To address these limitations, we propose BMSSnet, an interpretable AD recognition model based on a multi-scale multi-block attention mechanism. This model adopts a CNN-Transformer hybrid architecture. Specifically, it first captures local anatomical details using a 3D feature extraction network. Subsequently, it utilizes a dual-branch multi-scale attention mechanism to model patches of different sizes, enabling the Transformer to extract global long-range dependencies. Additionally, we devise a lightweight spatial gating unit to facilitate feature spatial interaction while maintaining computational efficiency. For interpretability, the model localizes decision-critical three-dimensional regions of interest (3D ROIs) using attention weights and aligns them with anatomical atlases to verify their pathological relevance. Finally, extensive experiments on the ADNI dataset demonstrate that BMSSnet not only achieves superior diagnostic performance but also accurately localizes AD-associated salient brain regions, offering reliable clinical interpretability.
A new BCI paradigm based on biological brain - digital twin brain dialogue.
Zhang T, Zhang R, Zeng X … +6 more , Zeng M, Xu Y, Xiong Y, Zhang G, Guo D, Yao D
Cogn Neurodyn · 2026 Dec · PMID 41883491 · Full text
Brain-computer interface (BCI) establishes a bidirectional pathway between the brain and external devices. Its applications fall into two main categories: utilizing the brain as a controller (e.g., for prosthetics) or as...
Brain-computer interface (BCI) establishes a bidirectional pathway between the brain and external devices. Its applications fall into two main categories: utilizing the brain as a controller (e.g., for prosthetics) or as a modulation target (e.g., for cognitive regulation). Progress in BCI is constrained by two core bottlenecks: in brain control, limited understanding of neural coding mechanisms restricts improvements in the accuracy and robustness of encoding/decoding algorithms; in brain regulation, one-size-fits-all regulatory strategies struggle to address significant individual variability, resulting in heterogeneous therapeutic responses. Inspired by neuroscience advances, this perspective proposes a new biological brain - digital twin brain based BCI (BDBCI) paradigm. Here, the biological brain acts as an empirical anchor and ultimate validation platform, while a high-fidelity digital twin brain (DTB) serves as a theoretical inference engine and virtual testbed. Specifically, experimental induction is applied to the biological brain to distill preliminary conclusions, such as brain-behavior mappings and brain-stimulation causal relationships, which are then used to construct and calibrate the DTB model. Subsequently, on the DTB platform, large-scale model deduction is conducted to validate and deepen these preliminary insights mechanistically, thereby optimizing control/regulation parameters or informing the parameter ranges for the next round of experimental induction and model deduction. Through this BDBCI paradigm, we aim to advance BCI research from empirical trial-and-error toward a new era of model-driven, predictable, and explainable precision science.
Modeling spectral EEG interactions using graph-structured variational representation learning.
Chodvadiya S, Suchithra MS
Cogn Neurodyn · 2026 Dec · PMID 41883490 · Full text
Emotion recognition from electroencephalogram (EEG) signals remains a challenging problem due to the high dimensionality, nonlinearity, and complex spectral dependencies inherent in neural activity. Conventional deep lea...
Emotion recognition from electroencephalogram (EEG) signals remains a challenging problem due to the high dimensionality, nonlinearity, and complex spectral dependencies inherent in neural activity. Conventional deep learning approaches often treat EEG features independently, thereby limiting their ability to capture structured spectral relationships. In this work, we propose a graph-based representation learning framework that models frequency-domain EEG features as nodes within a structured graph and leverages a Graph Neural Network-Variational Autoencoder (GNN-VAE) to learn compact latent representations. Spectral adjacency is defined using k-ring neighborhood connectivity, enabling localized message passing across contiguous frequency bands. The learned latent embeddings are subsequently classified using recurrent and attention-based temporal models to capture sequential dependencies across spectral segments. Experiments conducted on an EEG emotion dataset comprising three affective states demonstrate that the proposed approach consistently outperforms traditional machine learning baselines and non-graph deep learning models, achieving an accuracy of [Formula: see text] 91% and F1-score of 0.903. Ablation analyses further confirm the contribution of graph-based encoding and variational regularization to improved generalization. While the current study focuses on fixed spectral connectivity and subject-dependent evaluation, the results highlight the potential of graph-structured latent modeling for EEG-based emotion recognition and provide a foundation for future extensions incorporating adaptive graph learning and explainable representations.
Research on the perception of Huizhou traditional street nightscapes: a lab experiment using EEG.
Li Z, Li X, Fang M … +2 more , Sun X, Zhang W
Cogn Neurodyn · 2026 Dec · PMID 41868455 · Full text
With the continuous development of the nighttime economy in recent years, urban nocturnal illumination has received widespread attention. The evaluation of night lighting in traditional commercial streets, as a common el...
With the continuous development of the nighttime economy in recent years, urban nocturnal illumination has received widespread attention. The evaluation of night lighting in traditional commercial streets, as a common element of urban history and commerce, is of great importance. In this study, we aimed to conduct a comprehensive investigation of the nighttime illumination of traditional urban streets, exemplified by Tunxi Old Street and Liyang IN Alley, Huangshan City, China, using methods such as electroencephalography(EEG) and the semantic differential technique. Two main results were generated. 1) In the night lighting of traditional commercial streets, reasonable illuminance must be achieved to avoid an incongruous nocturnal atmosphere that substantially affects street quality. 2) Regarding lighting selection, floodlighting produces the best effects, followed by compound lighting, whereas linear lighting yielded the poorest results.
GCNFormNet: branched graph-transformer architecture for EEG-based emotion recognition.
Raghav A, Indolia S
Cogn Neurodyn · 2026 Dec · PMID 41868454 · Full text
Existing EEG-based emotion recognition pipelines rely on complex preprocessing techniques, making it difficult to assess the true capability of the underlying architecture. To address this, we propose GCNFormNet, a hybri...
Existing EEG-based emotion recognition pipelines rely on complex preprocessing techniques, making it difficult to assess the true capability of the underlying architecture. To address this, we propose GCNFormNet, a hybrid architecture that models dynamic spatial and temporal dependencies in EEG signals. Our primary contribution is the dual-branched design, where GCN layers model graph-structured spatial relationships using a dynamically generated adjacency matrix, while Transformer blocks capture complex temporal dynamics using a Performer-based self-attention mechanism. We also replaced the traditional layer normalization of Transformers with DynamicTanh (DyT), an element-wise activation function that mimics the S-shaped mappings produced by normalization layers. We evaluate GCNFormNet using the unified EEGain framework on four benchmark datasets: SEED, SEED-IV, DEAP, and DREAMER. Our results demonstrate competitive performance, with the highest accuracy among compared methods on SEED-IV (0.46), supported by statistical significance tests. An interpretability analysis of the learned adjacency matrices revealed neurophysiologically meaningful connectivity patterns, including hemispheric asymmetry and prefrontal dominance discovered purely from data without anatomical priors. Finally, ablation and sensitivity analyses confirmed the synergistic contribution of GCN and Transformer components on three datasets, while revealing a dataset-specific dependency on DEAP where the full architecture was not optimal.
Emerging roles of astrocyte for treatment of focal epilepsy and mechanisms underlying lesion development.
Ji Q, Zhang Y, Yang Z
Cogn Neurodyn · 2026 Dec · PMID 41868453 · Full text
This study aims to elucidate astrocyte-mediated regulation of focal epileptic seizures and mechanisms underlying the development of epilepsy. To address this, we propose an improved cortical layer model to investigate th...
This study aims to elucidate astrocyte-mediated regulation of focal epileptic seizures and mechanisms underlying the development of epilepsy. To address this, we propose an improved cortical layer model to investigate the function and therapeutic value of astrocyte in neurological disorders. By analyzing the spatio-temporal characteristics of focal epilepsy seizures and their propagation, we find that high frequency inhibitory stimulation appeared to effectively delay or prevent seizures. In addition, the results suggest that different frequencies of Ca oscillation and levels of coupling strengths have substantial effects on focal epilepsy. Based on experimental and clinical research findings, we develop a potential clinical application process for epilepsy development and delineate its implications for the possibility of postoperative epilepsy recurrence.
Coexistence of infinitely many attractors in cosine-type memristor-driven hopfield neural networks and its application to image encryption.
Yin X, Zhao G, Chen C … +3 more , You Y, Zhou C, Zhang Y
Cogn Neurodyn · 2026 Dec · PMID 41868452 · Full text
This paper designs two improved passive cosine-type ideal memristors and incorporates them into the Hopfield neural network, thereby proposing a novel cosine-type memristor-driven Hopfield neural network (CMDHNN). The mo...
This paper designs two improved passive cosine-type ideal memristors and incorporates them into the Hopfield neural network, thereby proposing a novel cosine-type memristor-driven Hopfield neural network (CMDHNN). The model exhibits a planar equilibrium set and demonstrates extreme multistability, characterized by the coexistence of infinitely many attractors. The boundedness of the system is rigorously proven using the Lyapunov method. Nonlinear dynamics analysis tools, including bifurcation diagrams, Lyapunov exponent spectra, phase portraits, and time series plots, are employed to thoroughly investigate the model's complex chaotic dynamics. Leveraging the chaotic system of the proposed CMDHNN, an image encryption scheme is developed, in which chaotic sequences are utilized to generate diffusion and permutation key streams for encrypting the plaintext image. The results indicate that the encryption scheme based on this model exhibits excellent robustness and can effectively resist various common attacks.
About
- Frequency
- Sun
- Papers found
- 200
- RSS feed
- Subscribe