Distinguishing cognitive states using electroencephalography local activation and functional connectivity patterns

Suhail, T.A.; Indiradevi, K.P.; Suhara, E M; Poovathinal, Suresh A; Ayyappan, Anitha

doi:10.1016/j.bspc.2022.103742

Cited by 13 publications

(6 citation statements)

References 66 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, the temporal dynamics of directed connectivity networks can be combined with other EEG features to improve the classification accuracy of cognitive states. A recent study [63] showed that the fusion of EEG local activation parameters and brain connectivity patterns provides a better classification performance in detecting cognitive states compared to a single EEG feature.…”

Section: Discussionmentioning

confidence: 99%

Microstates of Dynamic Directed Connectivity Networks Revealing Visual Color Influences on the Brain Information Processing During Learning

Chai

Tang

2023

IEEE Access

View full text Add to dashboard Cite

Color has the exceptional ability to capture visual attention and is also capable of enhancing positive emotions, leading to a significant impact on human learning and memory. However, the influence of color on the spatiotemporal dynamics of brain connectivity networks during learning has remained unexplored. This study aimed to propose an analytical approach based on time-frequency decomposition and microstate analysis to capture temporal variations in dynamic directed connectivity networks using electroencephalography (EEG) signals for investigating the influence of visual color on network dynamics of the brain during a learning task. Wavelet transform and phase slope index were employed to estimate the dynamic directed connectivity networks of EEG signals. The estimated dynamic directed connectivity networks were then characterized using graph theoretical analysis. The recurring patterns of dynamic directed connectivity networks were classified using cluster analysis before the temporal dynamics of directed connectivity networks were quantified using microstate analysis. Forty-five healthy participants participated in the experiment, which included memorizing learning materials presented in three different colors (achromatic, cool, and warm). The results revealed that the dynamic directed connectivity networks could be grouped into several quasi-stable states and the presence of common and unique brain states repetitive across frequency bands under individual conditions. A joint analysis of all conditions revealed that the temporal dynamics (coverage, mean duration, and state transition probability) differed significantly between the achromatic and colored conditions. Few dynamic brain states were shared between conditions and tended to remain in particular brain states for a longer duration in specific frequency bands. Our observations provided the first evidence of temporal dynamics of frequency-specific directed connectivity networks in the brain during multimedia learning tasks, that is, increased coverage of top-down interactions in the θ and α bands and switching between top-down and bottom-up interactions (information flow from anterior to posterior regions and vice versa) in the α band, in colored conditions compared to that of achromatic conditions. Therefore, these results suggest that several frequency-specific directed connectivity networks cooperate during knowledge acquisition and may change over time (from one state to another). The proposed framework captures the temporal dynamics of directed connectivity networks, and provides implications for monitoring and assessing emotional and cognitive processes in various contexts.INDEX TERMS Complex Morlet wavelet, directionality index, dynamic directed connectivity network, electroencephalography, k-means clustering, microstate analysis, phase slope index.The associate editor coordinating the review of this manuscript and approving it for publication was Ludovico Minati .

show abstract

Section: Discussionmentioning

confidence: 99%

Microstates of Dynamic Directed Connectivity Networks Revealing Visual Color Influences on the Brain Information Processing During Learning

Chai

Tang

2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Because of the large individual variability of subjects’ EEG signals, when a dataset containing multi-subject data is used to construct a model, a large model width is required to ensure the performance and stability of the model ( Suhail et al, 2022 ). The training samples become larger and each sample needs to go through all the computations of the model, which leads to a large increase in the training cost.…”

Section: Methodsmentioning

confidence: 99%

“…Because of the large individual variability of subjects' EEG signals, when a dataset containing multi-subject data is used to construct a model, a large model width is required to ensure the performance and stability of the model (Suhail et al, 2022).…”

Section: Sparsity Improvement In Transformer Networkmentioning

confidence: 99%

EMPT: a sparsity Transformer for EEG-based motor imagery recognition

Liu,

Shi

et al. 2024

Front. Neurosci.

View full text Add to dashboard Cite

IntroductionTransformer network is widely emphasized and studied relying on its excellent performance. The self-attention mechanism finds a good solution for feature coding among multiple channels of electroencephalography (EEG) signals. However, using the self-attention mechanism to construct models on EEG data suffers from the problem of the large amount of data required and the complexity of the algorithm.MethodsWe propose a Transformer neural network combined with the addition of Mixture of Experts (MoE) layer and ProbSparse Self-attention mechanism for decoding the time-frequency-spatial domain features from motor imagery (MI) EEG of spinal cord injury patients. The model is named as EEG MoE-Prob-Transformer (EMPT). The common spatial pattern and the modified s-transform method are employed for achieving the time-frequency-spatial features, which are used as feature embeddings to input the improved transformer neural network for feature reconstruction, and then rely on the expert model in the MoE layer for sparsity mapping, and finally output the results through the fully connected layer.ResultsEMPT achieves an accuracy of 95.24% on the MI EEG dataset for patients with spinal cord injury. EMPT has also achieved excellent results in comparative experiments with other state-of-the-art methods.DiscussionThe MoE layer and ProbSparse Self-attention inside the EMPT are subjected to visualisation experiments. The experiments prove that sparsity can be introduced to the Transformer neural network by introducing MoE and kullback-leibler divergence attention pooling mechanism, thereby enhancing its applicability on EEG datasets. A novel deep learning approach is presented for decoding EEG data based on MI.

show abstract

“…Intensive research has been conducted in recent years to develop advanced sensory technologies allowing non-invasive monitoring and classification of neural correlates of cognitive processing in human brains via EEG with the aid of convolutional neural networks [ 1 ], for example. Some interesting studies have focused on exploring neural correlates of learning and working memory tasks, for example, with the use of EEG to compare several classifiers [ 2 ]. As indicated by [ 3 ], EEG technologies have some advantages and good applications for Developmental Cognitive Neuroscience, while new trends in quantitative analysis are emerging in a broad spectrum of applications, including medical applications [ 4 ].…”

Section: Introductionmentioning

confidence: 99%

Information-Theoretical Analysis of the Cycle of Creation of Knowledge and Meaning in Brains under Multiple Cognitive Modalities

Davis,

Schübeler,

Kozma

2024

Sensors

View full text Add to dashboard Cite

It is of great interest to develop advanced sensory technologies allowing non-invasive monitoring of neural correlates of cognitive processing in people performing everyday tasks. A lot of progress has been reported in recent years in this research area using scalp EEG arrays, but the high level of noise in the electrode signals poses a lot of challenges. This study presents results of detailed statistical analysis of experimental data on the cycle of creation of knowledge and meaning in human brains under multiple cognitive modalities. We measure brain dynamics using a HydroCel Geodesic Sensor Net, 128-electrode dense-array electroencephalography (EEG). We compute a pragmatic information (PI) index derived from analytic amplitude and phase, by Hilbert transforming the EEG signals of 20 participants in six modalities, which combine various audiovisual stimuli, leading to different mental states, including relaxed and cognitively engaged conditions. We derive several relevant measures to classify different brain states based on the PI indices. We demonstrate significant differences between engaged brain states that require sensory information processing to create meaning and knowledge for intentional action, and relaxed-meditative brain states with less demand on psychophysiological resources. We also point out that different kinds of meanings may lead to different brain dynamics and behavioral responses.

show abstract

Distinguishing cognitive states using electroencephalography local activation and functional connectivity patterns

Cited by 13 publications

References 66 publications

Microstates of Dynamic Directed Connectivity Networks Revealing Visual Color Influences on the Brain Information Processing During Learning

Microstates of Dynamic Directed Connectivity Networks Revealing Visual Color Influences on the Brain Information Processing During Learning

EMPT: a sparsity Transformer for EEG-based motor imagery recognition

Information-Theoretical Analysis of the Cycle of Creation of Knowledge and Meaning in Brains under Multiple Cognitive Modalities

Contact Info

Product

Resources

About