Cognitive Neurodynamics最新文献

Attention is a core cognitive domain crucial in facilitating day-to-day life. Using an attention network test (ANT) along with event-related potentials (ERPs) in older individuals with hearing loss would provide excellent information about the impact of hearing loss on attentional processes. Thus, the current study aims to understand the attentional deficits and its cortical dynamics in older individuals with and without hearing loss. The study recruited 40 participants, 20 older individuals with hearing loss and 20 age and education-matched controls with normal hearing. All the participants underwent cognitive assessment using ANT with simultaneous 32-channel EEG recording. Results revealed significant impairment in executive attention and subtle alterations in alerting and orienting attention among older individuals with hearing loss compared to their normal-hearing counterparts. These findings suggest the negative impact of hearing loss on attentional networks. In addition, ANT and ERPs provide insight into the underlying neural mechanisms in specific attention network deficits associated with hearing loss.

Compared to artificial neural networks (ANNs), spiking neural networks (SNNs) present a more biologically plausible model of neural system dynamics. They rely on sparse binary spikes to communicate information and operate in an asynchronous, event-driven manner. Despite the high heterogeneity of the neural system at the neuronal level, most current SNNs employ the widely used leaky integrate-and-fire (LIF) neuron model, which assumes uniform membrane-related parameters throughout the entire network. This approach hampers the expressiveness of spiking neurons and restricts the diversity of neural dynamics. In this paper, we propose replacing the resistor in the LIF model with a discrete memristor to obtain the heterogeneous memristive LIF (MLIF) model. The memristance of the discrete memristor is determined by the voltage and flux at its terminals, leading to dynamic changes in the membrane time parameter of the MLIF model. SNNs composed of MLIF neurons can not only learn synaptic weights but also adaptively change membrane time parameters according to the membrane potential of the neuron, enhancing the learning ability and expression of SNNs. Furthermore, since the proper threshold of spiking neurons can improve the information capacity of SNNs, a learnable straight-through estimator (LSTE) is proposed. The LSTE, based on the straight-through estimator (STE) surrogate function, features a learnable threshold that facilitates the backward propagation of gradients through neurons firing spikes. Extensive experiments on several popular static and neuromorphic benchmark datasets demonstrate the effectiveness of the proposed MLIF and LSTE, especially on the DVS-CIFAR10 dataset, where we achieved the top-1 accuracy of 84.40(%).

Transformer neural networks based on multi-head self-attention are effective in several fields. To capture brain activity on electroencephalographic (EEG) signals and construct an effective pattern recognition model, this paper explores the multi-channel deep feature decoding method utilizing the self-attention mechanism. By integrating inter-channel features with intra-channel features, the self-attention mechanism generates a deep feature vector that encompasses information from all brain activities. In this paper, a time-frequency-spatial domain analysis of motor imagery (MI) based EEG signals from spinal cord injury patients is performed to construct a transformer neural network-based MI classification model. The proposed algorithm is named time-frequency-spatial transformer. The time-frequency and spatial domain feature vectors extracted from the EEG signals are input into the transformer neural network for multiple self-attention depth feature encoding, a peak classification accuracy of 93.56% is attained through the fully connected layer. By constructing the attention matrix brain network, it can be inferred that the channel connections constructed by the attention heads have similarities to the brain networks constructed by the EEG raw signals. The experimental results reveal that the self-attention coefficient brain network holds significant potential for brain activity analysis. The self-attention coefficient brain network can better illustrate correlated connections and show sample differences. Attention coefficient brain networks can provide a more discriminative approach for analyzing brain activity in clinical settings.

Objective. The assessment of mental fatigue (MF) and attention span in educational and healthcare settings frequently relies on subjective scales or methods such as induced-task interruption tools. However, these approaches are deficient in real-time evaluation and dynamic definitions. To address this gap, this paper proposes a Continuous Quantitative Scale (CQS) that allows for the natural and real-time measurement of MF based on group-synchronized electroencephalogram (EEG) data. Approach. In this study, computational psychophysiology was used to measure MF scores during a realistic class. Our methodology continuously monitored participants’ psychological states without interrupting their regular routines, providing an objective evaluation. By analyzing multi-subject brain-computer interface (mBCI) data with a collaborative computing approach, the group-synchronized data were obtained from 10 healthy participants to assess MF levels. Each participant wore an EEG headset for only 10 min of preparation before performing a sustained task for 80 min. Main results. Our findings indicate that a lecture duration of 18.9 min is most effective, while a duration of 43.1 min leads to heightened MF levels. By focusing on the group-level simultaneous data analysis, the effects of individual variability were mitigated and the efficiency of cognitive computing was improved. From the perspective of a neurocomputational measure, these results confirm previous research. Significance. The proposed CQS provides a reliable, objective, memory- and emotion-free approach to the assessment of MF and attention span. These findings have significant implications not only for education, but also for the study of group cognitive mechanisms and for improving the quality of mental healthcare.

Two coordinated dynamic properties (adaptation and sensitization) are observed in retinal ganglion cells (RGCs) under the contrast stimulation. During sustained high-contrast period, adaptation decreases RGCs' responses while sensitization increases RGCs' responses. In mouse retina, adaptation and sensitization respectively show OFF- and ON-pathway-dominance. However, the mechanisms which drive the differentiation between adaptation and sensitization remain unclear. In the present study, multi-electrode recordings were conducted on isolated mouse retina under full-field contrast stimulation. Dynamic property was quantified based on the trend of RGC's firing rate during high-contrast period, light sensitivity was estimated by linear-nonlinear analysis and coding ability was estimated through stimulus reconstruction algorism. γ-Aminobutyric acid (GABA) receptors were pharmacologically blocked to explore the relation between RGCs' dynamic property and the activity of GABA receptors. It was found that GABA_A and GABA_C receptors respectively mediated the adaptation and sensitization processes in RGCs' responses. RGCs' dynamic property changes occurred after the blockage of GABA receptors were related to the modulation of the cells' light sensitivity. Further, the blockage of GABA_A (GABA_C) receptor significantly decreased RGCs' overall coding ability and eliminated the functional benefits of adaptation (sensitization). Our work suggests that the dynamic property of individual RGC is related to the balance between its GABA_A-receptor-mediated inputs and GABA_C-receptor-mediated inputs. Blockage of GABA receptors breaks the balance of retinal circuitry for signal processing, and down-regulates the visual information coding ability.

Supplementary information: The online version contains supplementary material available at 10.1007/s11571-023-09950-2.

[This corrects the article DOI: 10.1007/s11571-021-09728-4.].

Steady-state visual evoked potentials (SSVEPs) based brain–computer interface (BCI) has received considerable attention due to its high information transfer rate (ITR) and available quantity of targets. However, the performance of frequency identification methods heavily hinges on the amount of user calibration data and data length, which hinders the deployment in real-world applications. Recently, generative adversarial networks (GANs)-based data generation methods have been widely adopted to create synthetic electroencephalography data, holds promise to address these issues. In this paper, we proposed a GAN-based end-to-end signal transformation network for Time-window length Extension, termed as TEGAN. TEGAN transforms short-length SSVEP signals into long-length artificial SSVEP signals. Additionally, we introduced a two-stage training strategy and the LeCam-divergence regularization term to regularize the training process of GAN during the network implementation. The proposed TEGAN was evaluated on two public SSVEP datasets (a 4-class and 12-class dataset). With the assistance of TEGAN, the performance of traditional frequency recognition methods and deep learning-based methods have been significantly improved under limited calibration data. And the classification performance gap of various frequency recognition methods has been narrowed. This study substantiates the feasibility of the proposed method to extend the data length for short-time SSVEP signals for developing a high-performance BCI system. The proposed GAN-based methods have the great potential of shortening the calibration time and cutting down the budget for various real-world BCI-based applications.

The basal ganglia (BG) in the brain exhibit diverse functions for motor, cognition, and emotion. Such BG functions could be made via competitive harmony between the two competing pathways, direct pathway (DP) (facilitating movement) and indirect pathway (IP) (suppressing movement). As a result of break-up of harmony between DP and IP, there appear pathological states with disorder for movement, cognition, and psychiatry. In this paper, we are concerned about the Huntington’s disease (HD), which is a genetic neurodegenerative disorder causing involuntary movement and severe cognitive and psychiatric symptoms. For the HD, the number of D2 SPNs ((N_{rm D2})) is decreased due to degenerative loss, and hence, by decreasing (x_{rm D2}) (fraction of (N_{rm D2})), we investigate break-up of harmony between DP and IP in terms of their competition degree (mathcal{C}_d), given by the ratio of strength of DP ((mathcal{S}_{DP})) to strength of IP ((mathcal{S}_{IP})) (i.e., (mathcal{C}_d = mathcal{S}_{DP} / mathcal{S}_{IP})). In the case of HD, the IP is under-active, in contrast to the case of Parkinson’s disease with over-active IP, which results in increase in (mathcal{C}_d) (from the normal value). Thus, hyperkinetic dyskinesia such as chorea (involuntary jerky movement) occurs. We also investigate treatment of HD, based on optogenetics and GP ablation, by increasing strength of IP, resulting in recovery of harmony between DP and IP. Finally, we study effect of loss of healthy synapses of all the BG cells on HD. Due to loss of healthy synapses, disharmony between DP and IP increases, leading to worsen symptoms of the HD.

The electroencephalogram (EEG) signal is being investigated as a more confidential biometric for person identification. Despite recent advancements, a persistent challenge lies in the influence of variations in affective states. Affective states consistently exist during data collection, regardless of the protocol used. Additionally, the inherently non-stationary nature of EEG makes it susceptible to fluctuations in affective states over time. Therefore, it would be highly crucial to perform precise EEG-based person identification under varying affective states. This paper employed an integrated Multi-scale Convolution and Graph Pooling network (MCGP) to mitigate the impact of affective state variations. MCGP utilized multiple 1D convolutions at different scales to dynamically extract and fuse features. Additionally, a graph pooling layer with an attention mechanism was incorporated to generate hierarchical graph embeddings. These embeddings were concatenated as inputs for a fully connected classification layer. Experiments were conducted on the SEED and SEED-V dataset, revealing that MCGP achieved an average accuracy of 85.51% for SEED and 88.69% for SEED-V in cross-session conditions involving mixed affective states. Under single affective state cross-session scenario, MCGP achieved an average accuracy of 85.75% for SEED and 88.06% for SEED-V for the same affective states, while obtaining 79.57% for SEED and 84.52% for SEED-V for different affective states. Results indicated that, compared to the baseline methods, MCGP effectively mitigated the impact of variations in affective states across different sessions. In single affective state cross-session scenario, identification performance for the same affective states was slightly higher than that for different affective states.

Emotion recognition plays an important role in human life and healthcare. The EEG has been extensively researched as an objective indicator of intense emotions. However, current existing methods lack sufficient analysis of shallow and deep EEG features. In addition, human emotions are complex and variable, making it difficult to comprehensively represent emotions using a single-modal signal. As a signal associated with gaze tracking and eye movement detection, Eye-related signals provide various forms of supplementary information for multimodal emotion analysis. Therefore, we propose a Pseudo-Siamese Pyramid Network (PSPN) for multimodal emotion analysis. The PSPN model employs a Depthwise Separable Convolutional Pyramid (DSCP) to extract and integrate intrinsic emotional features at various levels and scales from EEG signals. Simultaneously, we utilize a fully connected subnetwork to extract the external emotional features from eye-related signals. Finally, we introduce a Pseudo-Siamese network that integrates a flexible cross-modal dual-branch subnetwork to collaboratively utilize EEG emotional features and eye-related behavioral features, achieving consistency and complementarity in multimodal emotion recognition. For evaluation, we conducted experiments on the DEAP and SEED-IV public datasets. The experimental results demonstrate that multimodal fusion significantly improves the accuracy of emotion recognition compared to single-modal approaches. Our PSPN model achieved the best accuracy of 96.02% and 96.45% on the valence and arousal dimensions of the DEAP dataset, and 77.81% on the SEED-IV dataset, respectively. Our code link is: https://github.com/Yinyanyan003/PSPN.git.