Cognitive Neurodynamics最新文献

Magnetoencephalography (MEG) records the extremely weak magnetic fields on the surface of the scalp through highly sensitive sensors. Multi-channel MEG data provide higher spatial and temporal resolution when measuring brain activities, and can be applied for brain-computer interfaces as well. However, a large number of channels leads to high computational complexity and can potentially impact decoding accuracy. To improve the accuracy of MEG decoding, this paper proposes a new coherence-based channel selection method that effectively identifies task-relevant channels, reducing the presence of noisy and redundant information. Riemannian geometry is then used to extract effective features from selected channels of MEG data. Finally, MEG decoding is achieved by training a support vector machine classifier with the Radial Basis Function kernel. Experiments were conducted on two public MEG datasets to validate the effectiveness of the proposed method. The results from Dataset 1 show that Riemannian geometry achieves higher classification accuracy (compared to common spatial patterns and power spectral density) in the single-subject visual decoding task. Moreover, coherence-based channel selection significantly (P = 0.0002) outperforms the use of all channels. Moving on to Dataset 2, the results reveal that coherence-based channel selection is also significantly (P <0.0001) superior to all channels and channels around C3 and C4 in cross-session mental imagery decoding tasks. Additionally, the proposed method outperforms state-of-the-art methods in motor imagery tasks.

The beneficial psychological effects of exercise might be explained by self-determination theory and autonomy. However, the underlying neurophysiological mechanisms are even less elucidated. Previously neglected, aperiodic (1/f) brain activity is suggested to indicate enhanced cortical inhibition when the slope is steeper. This is thought to be associated with an increased cognitive performance. Therefore, we hypothesize that running with a self-selected intensity and thus given autonomy leads to stronger neural inhibition accompanied by psychological improvements. Twenty-nine runners performed two 30-min runs. First, they chose their individual feel-good intensity (self-selected run; SR). After a 4-weeks washout, the same speed was blindly prescribed (imposed run; IR). Acute effects on mood (Feeling Scale, Felt Arousal Scale, MoodMeter®), cognition (d2-R, digit span test) and electrocortical activity (slope, offset, 1/f-corrected alpha and low beta band) were analyzed before and after the runs. Both runs had an equal physical workload and improved mood in the Felt Arousal Scale, but not in the Feeling Scale or MoodMeter®. Cognitive performance improved after both runs in the d2-R, while it remained stable in the digit span test after SR, but decreased after IR. After running, the aperiodic slope was steeper, and the offset was reduced. Alpha activity increased after SR only, while low beta activity decreased after both conditions. The aperiodic features partially correlated with mood and cognition. SR was not clearly superior regarding psychological effects. Reduced aperiodic brain activity indicates enhanced neural inhibition after both runs. The 1/f-corrected alpha band may emphasize a different neural processing between both runs.

To illustrate the occurrences of extreme events in the neural system we consider a pair of Chialvo neuron maps. Importantly, we explore the dynamics of the proposed system by including a flux term between the neurons. Primarily, the dynamical behaviors of the coupled Chialvo neurons are examined using the Lyapunov spectrum and bifurcation analysis. We find the transitions between the periodic and chaotic dynamics in relation to the injected ion current of the first and second neurons and the flux coupling strength. It is interesting to note that, the extreme events can occur in the chaotic zone for some parameters. The analysis is then extended to a network of Chialvo neurons with various network connectivities. We discover that coexisting coherent and incoherent behaviour can arise and that nodes in the network can exhibit extreme event features. The findings of this study could help to better understand the rare large-amplitude events that occur in neural networks, which can help detect and prevent a variety of neurological disorders.

The recent exceptional demand for emotion recognition systems in clinical and non-medical applications has attracted the attention of many researchers. Since the brain is the primary object of understanding emotions and responding to them, electroencephalogram (EEG) signal analysis is one of the most popular approaches in affect classification. Previously, different approaches have been presented to benefit from brain connectivity information. We envisioned analyzing the interactions between brain electrodes with the information potential and providing a new index to quantify the connectivity matrix. The current study proposed a simple measure based on the cross-information potential between pairs of EEG electrodes to characterize emotions. This measure was tested for different EEG frequency bands to realize which EEG waves could be fruitful in recognizing emotions. Support vector machine and k-nearest neighbor (kNN) were implemented to classify four emotion categories based on two-dimensional valence and arousal space. Experimental results on the Database for Emotion Analysis using Physiological signals revealed a maximum accuracy of 90.14%, a sensitivity of 89.71%, and an F-score of 94.57% using kNN. The gamma frequency band obtained the highest recognition rates. Furthermore, low valence-low arousal was classified more effectively than other classes.

Electroencephalography (EEG) signals provide information about the brain activities, this study bridges neuroscience and machine learning by introducing an astronomy-inspired feature extraction model. In this work, we developed a novel feature extraction function, black-white hole pattern (BWHPat) which dynamically selects the most suitable pattern from 14 options. We developed BWHPat in a four-phase feature engineering model, involving multileveled feature extraction, feature selection, classification, and cortex map generation. Textural and statistical features are extracted in the first phase, while tunable q-factor wavelet transform (TQWT) aids in multileveled feature extraction. The second phase employs iterative neighborhood component analysis (INCA) for feature selection, and the k-nearest neighbors (kNN) classifier is applied for classification, yielding channel-specific results. A new cortex map generation model highlights the most active channels using median and intersection functions. Our BWHPat-driven model consistently achieved over 99% classification accuracy across three scenarios using the publicly available EEG pain dataset. Furthermore, a semantic cortex map precisely identifies pain-affected brain regions. This study signifies the contribution to EEG signal classification and neuroscience. The BWHPat pattern establishes a unique link between astronomy and feature extraction, enhancing the understanding of brain activities.

Expectation States Theory suggests that social status carries emotions, with higher statuses producing positive emotions and lower statuses leading to negative emotions. However, the theory is broad and lacks empirical evidence. This study investigated whether positive and negative evaluations from positions of higher and lower social hierarchies affect decisions. We examined whether decision making is influenced when evaluations were given in a first (L1) versus second language (L2). Bilinguals read scenarios in which they imagined themselves in the middle of the hierarchy. They then made a series of decisions, each of which was preceded with an evaluative word from other individuals whose hierarchical positions were higher or lower. The behavioral results showed that negative evaluations from higher positions exerted greater impact on decisions than when negative evaluations came from a lower position. At the neural level, after receiving negative evaluations, a higher hierarchy elicited greater activation in the right inferior frontal gyrus (IFG), left supplementary motor area (SMA), right precentral gyrus, left fusiform gyrus, bilateral inferior occipital gyrus (IOG), and right AI compared to a lower hierarchy, which may be caused by the view that a negative evaluation from a higher hierarchy is criticism. Conversely, after receiving positive evaluations, the lower hierarchy elicited greater activation in the right IFG, left SMA, right precentral gyrus, bilateral IOG, right AI and right IPS compared to the higher hierarchy, which may be due to the fact that positive evaluations from positions of lower hierarchies are perceived as encouraging. Together, these findings support Expectation States Theory in that regardless of whether evaluative advice is given in an L1 or L2, there is an internal association between social status and social-emotional neural responses that are localized in the frontal–parietal and visual cortices.

Vibration is an indispensable part of the tactile perception, which is encoded to oscillatory synaptic currents by receptors and transferred to neurons in the brain. The A2 and B1 neurons in the drosophila brain postsynaptic to the vibration receptors exhibit selective preferences for oscillatory synaptic currents with different frequencies, which is caused by the specific voltage-gated Na⁺ and K⁺ currents that both oppose the variations in membrane potential. To understand the peculiar role of the Na⁺ and K⁺ currents in shaping the filtering property of A2 and B1 neurons, we develop a linearized modeling framework that allows to systematically change the activation properties of these ionic channels. A data-driven conductance-based biophysical model is used to reproduce the frequency filtering of oscillatory synaptic inputs. Then, this data-driven model is linearized at the resting potential and its frequency response is calculated based on the transfer function, which is described by the magnitude–frequency curve. When we regulate the activation properties of the Na⁺ and K⁺ channels by changing the biophysical parameters, the dominant pole of the transfer function is found to be highly correlated with the fluctuation of the active current, which represents the strength of suppression of slow voltage variation. Meanwhile, the dominant pole also shapes the magnitude–frequency curve and further qualitatively determines the filtering property of the model. The transfer function provides a parsimonious description of how the biophysical parameters in Na⁺ and K⁺ channels change the inhibition of slow variations in membrane potential by Na⁺ and K⁺ currents, and further illustrates the relationship between the filtering properties and the activation properties of Na⁺ and K⁺ channels. This computational framework with the data-driven conductance-based biophysical model and its linearized model contributes to understanding the transmission and filtering of vibration stimulus in the tactile system.

Recognizing familiar faces holds great value in various fields such as medicine, criminal investigation, and lie detection. In this paper, we designed a Complex Trial Protocol-based familiar and unfamiliar face recognition experiment that using self-face information, and collected EEG data from 147 subjects. A novel neural network-based method, the EEG-based Face Recognition Model (EEG-FRM), is proposed in this paper for cross-subject familiar/unfamiliar face recognition, which combines a multi-scale convolutional classification network with the maximum probability mechanism to realize individual face recognition. The multi-scale convolutional neural network extracts temporal information and spatial features from the EEG data, the attention module and supervised contrastive learning module are employed to promote the classification performance. Experimental results on the dataset reveal that familiar face stimuli could evoke significant P300 responses, mainly concentrated in the parietal lobe and nearby regions. Our proposed model achieved impressive results, with a balanced accuracy of 85.64%, a true positive rate of 73.23%, and a false positive rate of 1.96% on the collected dataset, outperforming other compared methods. The experimental results demonstrate the effectiveness and superiority of our proposed model.

The accumulation of amyloid β peptide (left( {text{A}}beta right) ) is assumed to be one of the main causes of Alzheimer’s disease (left( {text{AD}}right) ). There is increasing evidence that astrocytes are the primary targets of Aβ. Aβ can cause abnormal synaptic glutamate, aberrant extrasynaptic glutamate, and astrocytic calcium dysregulation through astrocyte glutamate transporters facing the synaptic cleft (GLT-syn), astrocyte glutamate transporters facing the extrasynaptic space (GLT-ess), metabotropic glutamate receptors in astrocytes (mGluR), N-methyl-D-aspartate receptors in astrocytes (NMDAR), and glutamatergic gliotransmitter release (Glio-Rel). However, it is difficult to experimentally identify the extent to which each pathway affects synaptic glutamate, extrasynaptic glutamate, and astrocytic calcium signaling. Motivated by these findings, this work established a concise mathematical model of astrocyte ({text{Ca}}^{2+}) dynamics, including the above Aβ-mediated glutamate-related multiple pathways. The model results presented the extent to which five mechanisms acted upon by Aβ affect synaptic glutamate, extrasynaptic glutamate, and astrocytic intracellular ({text{Ca}}^{2+}) signals. We found that GLT-syn is the main pathway through which Aβ affects synaptic glutamate. GLT-ess and Glio-Rel are the main pathways through which A(beta ) affects extrasynaptic glutamate. GLT-syn, mGluR, and NMDAR are the main pathways through which Aβ affects astrocytic intracellular ({text{Ca}}^{2+}) signals. Additionally, we discovered a strong, monotonically increasing relationship between the mean glutamate concentration and the mean ({text{Ca}}^{2+}) oscillation amplitude (or frequency). Our results may have therapeutic implications for slowing cell death induced by the combination of glutamate imbalance and ({text{Ca}}^{2+}) dysregulation in AD.

Brain-inspired neuromorphic computing has emerged as a promising solution to overcome the energy and speed limitations of conventional von Neumann architectures. In this context, in-memory computing utilizing memristors has gained attention as a key technology, harnessing their non-volatile characteristics to replicate synaptic behavior akin to the human brain. However, challenges arise from non-linearities, asymmetries, and device variations in memristive devices during synaptic weight updates, leading to inaccurate weight adjustments and diminished recognition accuracy. Moreover, the repetitive weight updates pose endurance challenges for these devices, adversely affecting latency and energy consumption. To address these issues, we propose a Siamese network learning approach to optimize the training of multi-level memristor neural networks. During neural inference, forward propagation takes place within the memristor neural network, enabling error and noise detection in the memristive devices and hardware circuits. Simultaneously, high-precision gradient computation occurs on the software side, initially updating the floating-point weights within the Siamese network with gradients. Subsequently, weight quantization is performed, and the memristor conductance values requiring updates are modified using a sparse update strategy. Additionally, we introduce gradient accumulation and weight quantization error compensation to further enhance network performance. The experimental results of MNIST data recognition, whether based on a MLP or a CNN model, demonstrate the rapid convergence of our network model. Moreover, our method successfully eliminates over 98% of weight updates for memristor conductance weights within a single epoch. This substantial reduction in weight updates leads to a significant decrease in energy consumption and time delay by more than 98% when compared to the basic closed-loop update method. Consequently, this approach effectively addresses the durability requirements of memristive devices.