Cogn Neurodyn [JOURNAL]
Sun
200 papers
Dynamics and energy encoding of a star-like neuron network composed of the Wang-Zhang model induced by compressing a sphere into a fingertip.
Shi X, Xu W, Zhuang L … +3 more , Meng F, Jiang H, Wang Z
Cogn Neurodyn · 2026 Dec · PMID 41868451 · Full text
The information transmission of the tactile system is closely related to neuron network dynamics and energy metabolism. However, the correlation mechanism between neuron encoding and energy consumption for fingertip unde...
The information transmission of the tactile system is closely related to neuron network dynamics and energy metabolism. However, the correlation mechanism between neuron encoding and energy consumption for fingertip under compression remains unclear. In this study, a star-like neuron network is constructed using the Wang-Zhang model as the node, and it is combined with a contact mechanics model to simulate the phenomenon when a sphere being compressed into the fingertip. The remote synchronization characteristics is explored via average maximum correlation coefficient and Kuramoto order parameter, and energy encoding rules of the network are discussed. The results show that the star-like network can achieve remote synchronization between central and peripheral neurons. The energy consumption of central neurons is much higher than that of peripheral neurons due to signal integration and direct compression. The neuron energy consumption exhibits a spatial distribution of "high in the center and low in the periphery". It is found that there is an optimal value for the number of network layers, at which energy consumption and information processing efficiency reach a balance. This study reveals the neurometabolic mechanism of tactile perception and provides a new theoretical reference for the study of tactile neuronal encoding.
Multi-scroll generation mechanism, dynamic analysis, and DSP implementation of a dual-memristor-coupled Sprott-C system.
Bi Y, Yang S, Xu X … +2 more , Banerjee S, Mou J
Cogn Neurodyn · 2026 Dec · PMID 41836197 · Full text
A five-dimensional multi-scroll chaotic system is presented by introducing two memristive elements into a three-dimensional chaotic system. The resulting model generates multi-scroll attractors whose scroll count can be...
A five-dimensional multi-scroll chaotic system is presented by introducing two memristive elements into a three-dimensional chaotic system. The resulting model generates multi-scroll attractors whose scroll count can be regulated by tuning the memristors' internal parameters. We analyze the equilibria and then quantify the dynamic behaviors using phase portraits, Poincaré sections, bifurcation diagrams, and Lyapunov exponents. Coexisting multi-scroll attractors are observed, and their attraction basins are mapped to visualize the corresponding spatial domains. Parameter-driven adjustment of local amplitude is also demonstrated, enabling flexible modulation of the system output. A DSP-based implementation is further provided to validate the realizability of the proposed design. The study advances memristor-assisted multi-scroll construction and supports engineering-oriented hardware realization of high-dimensional chaotic systems.
Modular memristor circuits for Pavlov associative memory with scalability.
Wang X, Ge Z, Wu M … +5 more , Zhang X, Zhang Z, Li XL, Iu HH, Kang SM
Cogn Neurodyn · 2026 Dec · PMID 41836196 · Full text
This paper proposes a novel neuronal circuit and a Pavlov associative memory network based on memristors, which possesses various characteristics such as adjustable conductance and can effectively simulate the synaptic c...
This paper proposes a novel neuronal circuit and a Pavlov associative memory network based on memristors, which possesses various characteristics such as adjustable conductance and can effectively simulate the synaptic connection strength between neurons in a neural network. Furthermore, this paper designs a corresponding digital logic circuit for the Pavlovian associative memory task to reproduce the associative memory process observed in Pavlov's experiments. Experimental results show that the proposed circuit can accurately implement associative memory and weight modulation, and can simulate the Pavlovian associative memory process. Combining memristor-based synaptic circuits and memristor-Complementary Metal Oxide Semiconductor digital logic gates, an associative memory circuit with modularity and simple structure is realized and verified using LTSpice. The proposed circuit realizes the basic learning and forgetting functions of Pavlovian associative circuits, which provides a novel circuit implementation form for further research based on Pavlovian association.
MedIntelliCare: neurodynamic-inspired AI for medical decision support by integrating retrieval-augmented generation with multimodal cognitive processing.
Kunekar P, Mankar S, Cholke P … +3 more , Kulkarni A, Nooji P, Gadhave R
Cogn Neurodyn · 2026 Dec · PMID 41836195 · Full text
MedIntelliCare is an AI-powered medical assistant designed to enhance diagnostic accuracy, reduce cognitive load on healthcare professionals, and integrate real-time medical data. While current AI-driven medical systems...
MedIntelliCare is an AI-powered medical assistant designed to enhance diagnostic accuracy, reduce cognitive load on healthcare professionals, and integrate real-time medical data. While current AI-driven medical systems focus on information retrieval and response generation, MedIntelliCare leverages Retrieval-Augmented Generation (RAG) combined with principles from neural computation and decision-making processes. This study explores the system's ability to simulate biologically inspired information processing by integrating brain-like computing, predictive modeling, and multimodal analysis, including EEG and neuroimaging data. By aligning MedIntelliCare with advances in computational neuroscience and intelligent diagnostics, we aim to establish a model that enhances clinical decision support through adaptive information retrieval. The system's future implications include cognitive disorder modeling, brain-computer collaboration, and advanced AI-driven diagnostics inspired by neural processing frameworks. Experimental validation using cosine similarity metrics demonstrates that MedIntelliCare achieves a 73% alignment with expert-generated reports, reinforcing its potential in neuro-inspired medical intelligence.
Application of neurodynamics theory in the study of neural circuits in major depressive disorder: a review on neural energy approaches.
Li Y, Wang R, Yan C … +6 more , Xu X, Wang Y, Pan X, Song Y, Zhang B, Liu Z
Cogn Neurodyn · 2026 Dec · PMID 41822235 · Full text
Major depressive disorder (MDD) is accompanied by abnormal reward processing, altered dopamine transmission in the ventral tegmental area-nucleus accumbens-medial prefrontal cortex (VTA-NAc-mPFC) dopaminergic pathway, an...
Major depressive disorder (MDD) is accompanied by abnormal reward processing, altered dopamine transmission in the ventral tegmental area-nucleus accumbens-medial prefrontal cortex (VTA-NAc-mPFC) dopaminergic pathway, and disruptions in both neural dynamics and brain energy metabolism. Yet, how these abnormalities converge within a unified framework of neural dynamics and neural energy coding remains unclear. The purpose of this review is to integrate and critically assess computational models of neural dynamics and neural energy coding in MDD, with a particular emphasis on the multiscale modeling approaches developed in our recent work, and to organize these advances into a coherent conceptual framework linking dopamine-related circuit dysfunction to alterations in neural energy consumption. First, we constructed Hodgkin-Huxley (H-H) models for the NAc medium spiny neuron (MSN) to simulate its neurodynamics. Then, using the neural energy model, we explored the energy consumption characteristics of MSNs and found that, in the MDD condition, MSN energy consumption during spiking was lower than in controls, demonstrating the feasibility and sensitivity of this energy-based methodology. To further examine how these mechanisms scale to functional circuits, we constructed a neural network dynamical model for the VTA-NAc-mPFC dopaminergic pathway and applied an augmented neural-energy computation framework to characterize its energy consumption features. Simulations demonstrated that neural energy consumption was substantially lower in the MDD condition, primarily due to decreased mPFC energy expenditure. Distinct energy-coding patterns emerged across neuronal types, and the energy required to encode a single action potential in both MSNs and pyramidal neurons increased under MDD low dopamine situation, indicating reduced energy efficiency. Moreover, the correlation between membrane potential and instantaneous power was moderate (0.6-0.9) rather than tight, and it changed substantially with dopamine levels. This shows that neural energy consumption carries additional neural information that is not reflected directly in membrane potential signals. Together, these findings establish a unified computational framework that links dopamine deficiency, ion-channel-level dysfunction, microcircuit dynamics impairment, and large-scale reductions in neural energy consumption. Our work highlights neural energy coding as a promising mechanistic indicator and potential biomarker for MDD, and provides a generalizable methodology for investigating other neuropsychiatric disorders.
Effect of changes in the arm physical parameters on the minimum torque-change trajectories of human reaching movements.
Muramatsu K, Kagawa T, Ogihara N
Cogn Neurodyn · 2026 Dec · PMID 41798052 · Full text
UNLABELLED: The minimum torque-change model is a computational model describing the trajectory formation of the point-to-point reaching movement in humans. This model roughly predicts a straight hand trajectory with a be...
UNLABELLED: The minimum torque-change model is a computational model describing the trajectory formation of the point-to-point reaching movement in humans. This model roughly predicts a straight hand trajectory with a bell-shaped velocity profile, as observed in human reaching movements. However, the minimum torque-change criterion is a dynamic quantity, and the calculated trajectories could be, at least to some extent, affected by changes in the arm's physical parameters such as mass, moment of inertia, and viscosity of each link. This study systematically investigates how changes in the arm's physical parameters affect the optimal arm trajectories calculated based on the minimum torque-change criterion. The calculated optimal trajectories were largely curved, particularly when the physical parameters of the forearm were doubled or halved from the original physical parameters. Furthermore, when the original parameters were modified to be biomechanically more appropriate, the trajectories were also largely curved, unlike those in actual human reaching movements. The results suggest that the hand trajectory in human reaching movements may be determined by a dynamic optimization criterion that is less sensitive to variations in the biomechanical properties. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s11571-026-10428-0.
Cognitive load alters cortical dynamics during gait in Parkinson's disease but not in neurologically healthy individuals.
Imaizumi LFI, Fukuchi CA, Simieli L … +5 more , Silveira CRA, Dos Santos PCR, Rodrigues ST, Polastri PF, Barbieri FA
Cogn Neurodyn · 2026 Dec · PMID 41767408 · Full text
The level of difficulty of a secondary cognitive task (DT) can affect gait and cortical activity distinctly in individuals with Parkinson's disease (PD). During a simpler ST, individuals with PD may use a compensatory ne...
The level of difficulty of a secondary cognitive task (DT) can affect gait and cortical activity distinctly in individuals with Parkinson's disease (PD). During a simpler ST, individuals with PD may use a compensatory neural mechanism by reallocating neural resources to preserve gait performance; for difficult DT, this compensation may not be the case. However, whether different levels of difficulty of a single-domain DT would distinctively affect gait and cortical activity in individuals with PD compared to neurologically healthy individuals is still unknown. Fourteen individuals with PD and 14 healthy individuals performed walking trials at self-selected speed, under six conditions of walking with an auditory DT and varying levels of difficulty (very easy: VE-SCT, easy: E-SCT, moderate: M-SCT, difficult: D-SCT, and very difficult: VD-SCT). Gait kinematics and cortical activity data were recorded. RM-ANOVAs identified that individuals with PD showed higher DT cost for both step length and step velocity when the cognitive task was D-SCT or VD-SCT, compared to easier tasks ( < 0.005). Cortical activity showed a different pattern. During more difficult tasks (M-SCT, D-SCT, VD-SCT), PD individuals had a lower DT cost in delta frequency (frontal and motor areas) and beta frequency (parietal area) compared to the easier tasks (VE-SCT, E-SCT) ( < 0.005). These findings suggest that individuals with PD exhibit a distinct pattern of cognitive-motor interaction during dual-task walking, characterized by increased cortical dual-task cost in lower vs. greater gait deterioration in higher task demands. These findings suggest that individuals with PD over-engage cognitive resources while walking with relatively easier DT.
Multi-branch Domain Adversarial Neural Network with dynamic weight allocation for multi-source EEG classification.
Gao Y, Ma Y, Liu Y … +2 more , Yin G, Qin Y
Cogn Neurodyn · 2026 Dec · PMID 41767407 · Full text
To address challenges such as the strong non-stationarity and inter-subject distribution shifts of EEG data, as well as the limitations of conventional DANN-based methods in feature representation and multi-source domain...
To address challenges such as the strong non-stationarity and inter-subject distribution shifts of EEG data, as well as the limitations of conventional DANN-based methods in feature representation and multi-source domain adaptation, a Multi-branch Domain Adversarial Neural Network with Multi-scale Channel Attention (MBCA-DANN) is proposed. To enhance feature richness, a multi-scale channel attention Module (MSCA) is designed, which provides multi-scale features and adaptively adjusts the feature channel weights, improving the feature capture ability of the network. A multi-branch architecture is constructed by combining auxiliary Maximum Mean Discrepancy (MMD), domain discriminators, and label discriminators, ensuring optimal matching between the source and target domains. Furthermore, a multi-source domain method with dynamic weight allocation is introduced, enhancing classification performance and robustness. Experimental results demonstrate that the classification accuracy for single-source domain transfer on the MII and MIII datasets is 71.89% and 71.82%, respectively, while the multi-source domain transfer classification accuracy improves to 79.83% and 82.87%. The model achieves a classification accuracy of 98.69% on the fatigue detection dataset, outperforming all currently known state-of-the-art algorithms, validating its strong generalization ability and providing an effective solution for multi-source cross-subject EEG classification.
Predictive modeling of vocal biomarkers for the diagnosis of Parkinson's disease.
Emegano DI, Mustapha MT, Isaac EP … +3 more , Ozsahin I, Uzun B, Ozsahin DU
Cogn Neurodyn · 2026 Dec · PMID 41728211 · Full text
Parkinson's disease (PD) is among the two most prevalent neurodegenerative disorders (NDDs), affecting about 2-3% of individuals aged 65 and older. This NDD exhibits characteristic motor symptoms and several other non-mo...
Parkinson's disease (PD) is among the two most prevalent neurodegenerative disorders (NDDs), affecting about 2-3% of individuals aged 65 and older. This NDD exhibits characteristic motor symptoms and several other non-motor features. Vocal deficits have been identified as one of the earliest quantifiable indicators of PD, which makes speech evaluation a viable, painless diagnostic instrument. We aim to apply machine learning (ML) models to vocal biomarkers for the early detection of PD, and use explainable artificial intelligence (XAI) techniques to interpret the predictions. The dataset is from Kaggle, a publicly reputable database, containing 1000 Parkinson's samples and 24 acoustic variables. We performed feature selection to identify the crucial vocal biological markers. Multiple machine learning (ML) models: Adaptive Boosting (AdaBoost), Random Forest (RF), Support Vector Machine (SVM), Multilayer Perceptron (MLP), Gaussian Naïve Bayes (GNB), Extreme Gradient Boosting (XGB), LightGBM (LGBM), CatBoost, Gradient Boosting (GB), Histogram-Based Gradient Boosting (HGB), and K-Nearest Neighbors (KNN) were employed. We also used SHAP (Shapley Additive exPlanations), LIME (Local Interpretable Model-agnostic Explanations), and Partial Dependence Plot (PDP) to explain the model performances. The HGB model ranked highest (1.00) based on accuracy, precision, recall, and F1-score, respectively. Also, the Confidence intervals (CI) (1.00,1.00) and p-value of < 0.001 of HGB were computed. XAI showed that jitter and shimmer-based biomarkers were the strongest contributors to the prediction of PD. In this study, the results showed that vocal base biomarker screening is not only economical but also an accessible diagnostic tool. In subsequent studies, we hope to include more varied datasets to improve both model and therapeutic relevance.
Reduced task-switching flexibility in parietal-cingulate and frontal circuits associated with brooding.
Singh S, Shaw SB, Becker S
Cogn Neurodyn · 2026 Dec · PMID 41728210 · Full text
UNLABELLED: Ruminative brooding is marked by its perseverative nature. Existing mechanistic theories attribute this to cognitive control deficits linked to elevated functional connectivity within the default mode network...
UNLABELLED: Ruminative brooding is marked by its perseverative nature. Existing mechanistic theories attribute this to cognitive control deficits linked to elevated functional connectivity within the default mode network and abnormal prefrontal activity. Here, we conceptualize ruminative brooding as an emergent property of a neural attractor state within the default mode network. Stable attractors are mathematically defined by two key properties: (1) convergence over time (assessing attractivity), and (2) resistance to perturbation (assessing stability). We tested whether brain states associated with brooding exhibited these properties in healthy volunteers using EEG during a task-switching protocol that interleaved cued rumination, working memory, and autobiographical memory tasks. Since cued rumination and working memory are thought to engage anticorrelated networks (default mode vs. central executive), switching from cued rumination to working memory effectively "perturbs" this system. Cued rumination was associated with beta power in the posterior cingulate cortex, with rumination disengagement marked by a reduction in beta power in posterior parietal and cingulate cortices. Moreover, high trait rumination was associated with impaired disengagement of these rumination-related dynamics and reduced recruitment of the dlPFC when transitioning from cued rumination to the working memory task, consistent with the "resistance to perturbation" criterion of a stable attractor. Furthermore, trait brooding was positively associated with a reduction in variance in posterior parietal and cingulate cortices time series over the course of cued rumination trials, consistent with the "convergence" criterion. These results provide support for framing brooding-related neural dynamics as pathological attractor states, providing a mechanistic account of rumination's perseverative quality. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s11571-026-10425-3.
Multistability analysis of state-dependent switching CVNNs with discontinuous nonmonotonic piecewise linear activation function and its application in associative memory.
Gong W, Yang L, Li Q … +3 more , Huang Z, Zhang L, Du F
Cogn Neurodyn · 2026 Dec · PMID 41728209 · Full text
This paper investigates the multistability of complex-valued neural networks (CVNNs) with state-dependent switching rules and discontinuous nonmonotonic piecewise linear activation functions featuring peaks. By leveragi...
This paper investigates the multistability of complex-valued neural networks (CVNNs) with state-dependent switching rules and discontinuous nonmonotonic piecewise linear activation functions featuring peaks. By leveraging Brouwer's fixed point theorem and the properties of strictly diagonally dominant matrices, we analyze the existence, stability, and instability of equilibrium points through state space decomposition. Our results demonstrate that an -neuron switching CVNNs can possess up to [Formula: see text] equilibrium points, among which [Formula: see text] are stable. These findings significantly extend existing results and enrich the stability theory of neural networks. Numerical examples validate the theoretical conclusions and illustrate potential applications in associative memory.
MountPat: investigations on the EEG signals.
Ince U, Goktas OF, Sercek I … +5 more , Kirik S, Barua PD, Baygin M, Dogan S, Tuncer T
Cogn Neurodyn · 2026 Dec · PMID 41728208 · Full text
To extract information from the brain, the most cost-effective method is electroencephalography (EEG) signal acquisition. Therefore, many researchers have used EEG signals to capture brain activity. EEG signals are compl...
To extract information from the brain, the most cost-effective method is electroencephalography (EEG) signal acquisition. Therefore, many researchers have used EEG signals to capture brain activity. EEG signals are complex; hence, computer-aided models-especially machine learning (ML)-are generally employed to interpret them. The primary objective of this research is to demonstrate the feature-extraction capability of a new, novel method. The proposed feature-extraction approach employs a deterministic feature-engineering transformation, designed to restructure multi-strided signal representations through fixed linear operations. The resulting transformation graph exhibits a mountain-like structure; therefore, we term the model MountPat. To evaluate MountPat's performance, we present an explainable feature engineering (XFE) model with four main phases. In the first phase, we extract informative features using MountPat. In the second phase, we select the most informative features using cumulative weighted iterative neighborhood component analysis (CWNCA). In the third phase, we generate classification results by applying t-algorithm-based k-nearest neighbors (tkNN). In the fourth phase, we extract explainable insights from the EEG signals using the Directed Lobish (DLob) explainable artificial intelligence (XAI) method. To demonstrate the general classification ability of the MountPat-based XFE framework, we use six EEG datasets. Under rigorous subject-independent (LOSO) validation, the model achieves 76.36%-98.88% accuracy, demonstrating strong cross-subject generalization. Sample-wise tenfold CV results exceed 89% on all six datasets. Moreover, by deploying the DLob XAI method, we generate interpretable results for each dataset. These results clearly illustrate that the MountPat-based XFE framework is an effective feature-extraction approach for multichannel signal processing.
Machine learning for missing data imputation in Alzheimer's research: predicting medial temporal lobe dynamic flexibility.
Moallemian S, Saghafi A, Deshpande R … +5 more , Perez JM, Budak M, Fausto BA, Elahi FM, Gluck MA
Cogn Neurodyn · 2026 Dec · PMID 41684835 · Full text
Alzheimer's disease (AD) pathology begins years before symptoms appear, and dynamic flexibility of the medial temporal lobe (MTL) may serve as an early functional biomarker. Using data from 656 older adults in the Rutger...
Alzheimer's disease (AD) pathology begins years before symptoms appear, and dynamic flexibility of the medial temporal lobe (MTL) may serve as an early functional biomarker. Using data from 656 older adults in the Rutgers Aging and Brain Health Alliance study, we evaluated whether cognitive, genetic, biochemical, and demographic predictors could estimate MTL dynamic flexibility, despite substantial missingness (1,866 missing values; 25.86%). Only 42 participants (6.40%) had complete data; therefore, we compared case deletion with five imputation strategies (MICE, GAIN, MissForest, MIWAE, ReMasker) and eight regression models, assessing prediction accuracy using repeated 5-fold cross-validation. Complete-case analysis yielded limited performance (average [Formula: see text], [Formula: see text]). After imputation, all methods improved accuracy, with MissForest paired with Bagging Trees or Random Forest achieving the lowest prediction error ([Formula: see text]). The greatest improvement in concordance occurred when GAIN was combined with Bagging Trees/Random Forest ([Formula: see text]), representing a 57% gain over the best complete-case model. A Scheirer-Ray-Hare ANOVA confirmed significant differences across imputation strategies ([Formula: see text]). Runtime analyses showed GAIN and MissForest to be both accurate and computationally efficient, while deep generative imputers were slower. These findings demonstrate that robust imputation is essential for maximizing data utility and predictive reliability in high-missingness neuroimaging studies and highlight the potential of ensemble tree models combined with advanced imputation techniques for estimating MTL dynamic flexibility in aging populations.
Mctsleepnet: a multiscale waveform and composite attention network with temporal dependency learning for robust EEG-based sleep staging.
Liu Z, Wu Y, Tan K … +1 more , Gao Y
Cogn Neurodyn · 2026 Dec · PMID 41684834 · Full text
Sleep staging is a critical indicator for assessing sleep quality and sleep disorders. Although significant progress has been made in sleep staging research, the representation of prominent waveforms and the capture of d...
Sleep staging is a critical indicator for assessing sleep quality and sleep disorders. Although significant progress has been made in sleep staging research, the representation of prominent waveforms and the capture of dynamic transitions between sleep stages still pose challenges. To address these issues, we propose MCTSleepNet, an Sleep staging Network containing Multiscale waveform representation, Composite attention and Time dependency learning modules based on single-channel electroencephalography (EEG). Firstly, multiscale waveform representation is learned from EEG signals using a dual-scale convolutional neural network (CNN). Then, a Composite Attention module is employed to enhance signal feature representation by considering both local and global contextual dependencies, thereby more effectively capturing prominent waveform features. Finally, a Bidirectional Gated Recurrent Unit (Bi-GRU) is used to learn the time dependent feature between EEG signals, enabling MCTSleepNet to model dynamic transitions between different sleep stages. Furthermore, considering the data imbalance between different sleep stages, this paper introduces an adaptive cross-entropy polynomial loss function to adjust the weights of different classes, thereby enhancing the model's attention to minority classes. Evaluation results on the publicly available Sleep-EDF-20 and Sleep-EDF-78 datasets demonstrate that MCTSleepNet performs exceptionally well in the sleep staging task.
DeBERTa-BiLSTM: a multi-label classification model for depression emotions.
Sarkar A, Majumder A
Cogn Neurodyn · 2026 Dec · PMID 41684833 · Full text
Multi-label depressive emotion classification remains challenging due to the co-occurrence of multiple mental health-related emotions, implicit linguistic expressions, long-range contextual dependencies, and the presence...
Multi-label depressive emotion classification remains challenging due to the co-occurrence of multiple mental health-related emotions, implicit linguistic expressions, long-range contextual dependencies, and the presence of both active and passive depressive signals. This paper presents a comprehensive study on multi-label depressive emotion detection using the DepressionEmo dataset, which contains textual instances annotated with eight clinically relevant depression emotions: Anger, Cognitive Dysfunction, Emptiness, Hopelessness, Loneliness, Sadness, Suicide Intent and Worthlessness. The objective is to develop an effective and computationally efficient multi-label classification framework capable of accurately identifying both explicit active and latent passive depressive emotions from text. To address this problem, a broad spectrum of transformer-based and hybrid architectures is evaluated, including BERT, RoBERTa, DistilBERT, T5, BART, and DeBERTa with BiLSTM integration, as well as seq2seq BART and seq2seq RoBERTa-BART models. The proposed DeBERTa-BiLSTM architecture integrates disentangled self-attention for rich contextual representation with a BiLSTM layer for sequential dependency learning and history-state fusion, enabling effective modeling of long-range depressive cues. Experimental results demonstrate that the proposed DeBERTa-BiLSTM model consistently outperforms baseline seq2seq BART, BERT, T5, GAN-BERT, and all other developed variants, achieving an F1-Micro score of 0.83 and an F1-Macro score of 0.80, along with the lowest Hamming Loss (0.15) and the highest Jaccard Index (0.71). The model further achieves micro-precision of 0.81 and micro-recall of 0.85 indicating robust detection of both frequent and minority emotion labels. Runtime analysis shows notable inference efficiency, reducing time per sample by 26.32% at batch size 4 and 21.39% at batch size 32 compared to seq2seq BART. Despite these advantages, the model remains computationally heavier than lightweight transformers, is influenced by the dataset's modest size, and requires further validation across broader mental health domains.
A brain-constrained neural model of cognition and language with NEST: transitioning from the Felix framework.
Carriere M, Dobler F, Plesser HE … +4 more , Feledyn A, Tomasello R, Wennekers T, Pulvermüller F
Cogn Neurodyn · 2026 Dec · PMID 41657965 · Full text
UNLABELLED: We introduce a brain-constrained neurocomputational model designed to simulate higher cognitive functions of the human brain, implemented using NEST, a widely used open-source simulator optimised for high-per...
UNLABELLED: We introduce a brain-constrained neurocomputational model designed to simulate higher cognitive functions of the human brain, implemented using NEST, a widely used open-source simulator optimised for high-performance spiking neural network simulations. Previously implemented in the custom-built C-based Felix simulation library, transitioning the model to NEST enhances accessibility, reproducibility, and computational efficiency. At the cellular level, the model comprises spiking excitatory neurons and local inhibitory neurons, whereas at the network level, it replicates the structural and functional organisation of 12 cortical regions spanning frontal, temporal, and occipital cortices, along with their associated inter-area connectivity. Additionally, global inhibition mechanisms and neuronal noise are integrated. Learning in the model follows biologically plausible Hebbian plasticity principles, incorporating both long-term potentiation and long-term depression. To validate the NEST implementation, we replicated previous simulation findings obtained with the Felix-based model. The new implementation successfully reproduced the same topographical distribution of cell assemblies following associative learning of object and action words within action and perception systems, replicating a range of previous neuroimaging results. Although the NEST model produced larger cell assemblies than Felix, the overall topographical patterns remained similar, indicating preservation of fundamental network characteristics. Moreover, the transition to NEST significantly enhanced computational efficiency, reducing simulation runtime nearly sixfold compared to Felix. This improvement in computational speed is crucial for expanding the model to include additional cortical regions, such as extending to the right hemisphere, which necessitates increased computational resources. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s11571-026-10415-5.
Optimized cortical EEG modeling for Parkinson disease diagnosis with snow Shepherd Stride tuning mechanism.
Kolla M, Madapuri RK, Kandukuri P … +3 more , Salvadi S, Tadepalli S, Gajula R
Cogn Neurodyn · 2026 Dec · PMID 41657964 · Full text
Parkinson's disease (PD) is a neurodegenerative disease that causes extensive impacts on cognitive and motor function, hence making correct and early diagnosis essential for efficient clinical management and enhancement...
Parkinson's disease (PD) is a neurodegenerative disease that causes extensive impacts on cognitive and motor function, hence making correct and early diagnosis essential for efficient clinical management and enhancement of the quality of life. Analysis using EEG has now become a safe and possible method for the identification of neural abnormalities in PD. Nevertheless, current models must struggle with several constraints: high false detection rates, poor generalizability across subjects, sensitivity to EEG noise pollution, and the inability to extract deep cortical representations, which have the capability to distinguish between healthy and Parkinson patterns. To alleviate these issues, the current paper proposes a novel CortiMoS-Net (Cortical Modeling with Stacked Autoencoder and MobileNet) capable of accurately detecting Parkinson's disease from EEG signals. CortiMoS-Net architecture combines deep stacked autoencoders with low-computation MobileNet convolution blocks such that low-complexity learning of complex cortical activity patterns is supplemented with computational scalability. To achieve further enhanced model convergence and optimization of learnable parameters, the present work also proposes an enhanced hybrid optimization technique named Snow Shepherd Stride Configuration Tuning (S3C-Tune). The proposed pipeline is initiated with raw EEG signal recording, preprocessing, and peak picking for the intent of artifact removal and detection of neurologically intriguing events. Model parameters are tuned by the S3C-Tune algorithm to realize maximal training accuracy. Such a pipeline hybrid enables extensive cortical modeling as well as efficient optimization and results in correct PD vs. healthy subject classification. Experimental results confirm the effectiveness of the suggested approach with better accuracy, precision, recall, and F1-score of 0.99 and minimum error rate and minimum loss of 0.01 and 0.05, respectively. The suggested model also indicates maximum prediction correctness of 0.99 and mean efficiency measure of 0.95 as compared to a large number of state-of-the-art hybrid deep learning approaches.
Ethical risks and considerations of the integration of Brain-Computer Interfaces with Artificial Intelligence.
Cao Y, Yang H, Xue Y … +4 more , Wang F, Li T, Zhao L, Fu Y
Cogn Neurodyn · 2026 Dec · PMID 41657963 · Full text
In recent years, with the rapid development of Brain-Computer Interface (BCI) and Artificial Intelligence (AI) technologies, a trend of increasing integration between the two has emerged. In some practical applications,...
In recent years, with the rapid development of Brain-Computer Interface (BCI) and Artificial Intelligence (AI) technologies, a trend of increasing integration between the two has emerged. In some practical applications, BCI has been combined with AI to enhance the overall performance of systems, including usability, user experience, and satisfaction, especially in terms of intelligent capabilities. However, this technological integration also introduces new or exacerbates existing ethical risks, such as neural privacy breaches, cross-domain misuse, and unclear system responsibility attribution. This paper discusses the novel or more severe ethical challenges arising from the fusion of BCI and AI technologies, as well as measures and strategies to address these ethical issues, calling for the establishment of more comprehensive ethical guidelines and governance frameworks. It is hoped that this paper will contribute to a deeper understanding and reflection on the ethical risks and corresponding regulations related to the integration of BCI and AI technologies.
Responses of fast-spiking basket cells to theta-modulated oscillatory synaptic input.
Liu M, Sun X
Cogn Neurodyn · 2026 Dec · PMID 41657962 · Full text
UNLABELLED: Fast-spiking basket cells (FSBCs) govern hippocampal oscillations through their rapid and sustained firing patterns, which drive rhythmic inhibition onto postsynaptic neurons, thereby enforcing population syn...
UNLABELLED: Fast-spiking basket cells (FSBCs) govern hippocampal oscillations through their rapid and sustained firing patterns, which drive rhythmic inhibition onto postsynaptic neurons, thereby enforcing population synchrony in the gamma and other frequency bands that support cognitive processes. Despite the established role of FSBCs in hippocampal oscillations, the precise mechanisms by which their dendrites influence membrane potential responses across different frequency bands remain unclear. In this study, we simulate oscillation-like input protocols to explore how dendrites modulate the spectral responses of the membrane potentials of FSBCs. Our results show that FSBCs exhibit both slow and fast oscillatory components, which are shaped by their action potentials. Input synchrony is essential for determining both the fast-band response frequency and its coupling with the slow frequency, while the neuron's intrinsic firing dynamics maintain the stability of the fast-band peak frequency across theta-range inputs. Although dendritic Na[Formula: see text]/A-type K[Formula: see text] channel blockade and cp-AMPA enhancement both increase fast-band frequency, they differentially affect phase-amplitude coupling, with blockade reducing and cp-AMPA enhancement increasing it, highlighting the role of intrinsic dendritic conductances and cp-AMPA inputs in promoting coupling. Furthermore, we show that the spatial distribution of synaptic inputs along dendrites affects the response frequencies, with distinct frequencies observed at different dendritic locations according to their electrotonic distance. These findings provide insights into how the intrinsic properties of FSBCs influence their response to oscillatory inputs. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s11571-026-10418-2.
Spatial-temporal representation of cortical neural activity evoked by acupuncture stimulation.
Yu H, Hu Z, Lin Z … +4 more , Wang J, Liu C, Liu J, Li G
Cogn Neurodyn · 2026 Dec · PMID 41647148 · Full text
UNLABELLED: Acupuncture modulates cognitive functions through acupoint stimulation and demonstrates significant regulatory effects on brain disorders. However, the underlying neurodynamic mechanisms of acupuncture remain...
UNLABELLED: Acupuncture modulates cognitive functions through acupoint stimulation and demonstrates significant regulatory effects on brain disorders. However, the underlying neurodynamic mechanisms of acupuncture remain unclear due to a lack of effective measures of brain activity. In this study, we developed an acupuncture-related potential (ARP) method based on Electroencephalogram (EEG) to elucidate the dynamic representation mechanisms underlying acupuncture stimulation. By analyzing ARP signal features and functional networks to capture stimulus-evoked brain activity, we derived spatiotemporal representations of neural manifolds and located across whole brain regions. It is exhibited that acupuncture induced significant four-phase event-related potentials (ERPs) waveforms predominantly in the parietal, frontal, central, and temporal lobes, with the parietal lobe exhibiting the highest amplitude at the P1 component (first positive peak). Latency gradients confirmed that the cortical neural activity originated in the parietal lobe and propagated through the central region to the frontal and temporal lobes. Dynamic network analysis revealed phase-specific reorganization: local frontal propagation (P1 component), global integration (P2 component), and novel topological pattern formation (P3 component). Neural manifold analysis uncovered a low-dimensional, ring-shaped representation encompassing the frontal, parietal, central, and temporal lobes. Acupuncture modulates brain function by activating key parietal lobe nodes, triggering distance-attenuated inter-regional signal transmission that dynamically reorganizes functional networks for multi-regional collaboration. The neural manifold representation revealed perception and integration of mechanisms of acupuncture information in the human brain. This ARP method provided a novel framework for investigating acupuncture-modulated spatiotemporal brain dynamics while enabling quantitative evaluation of its therapeutic effects. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s11571-025-10408-w.
About
- Frequency
- Sun
- Papers found
- 200
- RSS feed
- Subscribe