EEG Based Emotion Recognition by Combining Functional Connectivity Network and Local Activations

Li, Peiyang; Liu, Huan; Si, Yajing; Li, Cunbo; Li, Fali; Zhu, Xuyang; Huang, Xiaoye; Zeng, Ying; Yao, Dezhong; Zhang, Yangsong; Xu, Peng

doi:10.1109/tbme.2019.2897651

Cited by 276 publications

(144 citation statements)

References 66 publications

Supporting

Mentioning

136

Contrasting

Order By: Relevance

“…Furthermore, Li et al reported that connections with significant differences between the negative and neutral valences in the gamma band (30-48 Hz) are much denser than the connections in the beta band (12-30 Hz). The connections mainly occur in the right frontal and parietal-occipital lobes (Li et al, 2019). These studies demonstrate that EEG-based functional connectivity can effectively reflect the specificity of different emotional states.…”

Section: Introductionmentioning

confidence: 58%

“…In this study, we observed that connections with significant differences among the positive, the neutral, and the negative emotions were denser in the high gamma band than in the low gamma band. These denser connections with significant differences between the emotional states in the high frequency bands may indicate that high frequency components in EEG could mediate information transmission when the brain concentrates on processing the emotion-related activities (Li et al, 2019).…”

Section: Neural Signature Of Emotional States In the High Gamma Bandmentioning

confidence: 99%

“…Although the high frequency component of EEG signals has been investigated in emotion processing, most of these pioneering studies have focused on the activities of the local brain areas involved in affective perception. The human brain is a complex system; even a simple brain activity involves interactions among various brain regions (Straaten and Stam, 2013;Li et al, 2019). Emotion is a high-level cognitive function, and the processing of emotions requires the cooperation of multiple brain regions (Bassett and Bullmore, 2015).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

High Gamma Band EEG Closely Related to Emotion: Evidence From Functional Network

Yang

Shu

et al. 2020

Front. Hum. Neurosci.

Self Cite

View full text Add to dashboard Cite

High-frequency electroencephalography (EEG) signals play an important role in research on human emotions. However, the different network patterns under different emotional states in the high gamma band (50-80 Hz) remain unclear. In this paper, we investigate different emotional states using functional network analysis on various frequency bands. We constructed multiple functional networks on different frequency bands and performed functional network analysis and time-frequency analysis on these frequency bands to determine the significant features that represent different emotional states. Furthermore, we verified the effectiveness of these features by using them in emotion recognition. Our experimental results revealed that the network connections in the high gamma band with significant differences among the positive, neutral, and negative emotional states were much denser than the network connections in the other frequency bands. The connections mainly occurred in the left prefrontal, left temporal, parietal, and occipital regions. Moreover, long-distance connections with significant differences among the emotional states were observed in the high frequency bands, particularly in the high gamma band. Additionally, high gamma band fusion features derived from the global efficiency, network connections, and differential entropies achieved the highest classification accuracies for both our dataset and the public dataset. These results are consistent with literature and provide further evidence that high gamma band EEG signals are more sensitive and effective than the EEG signals in other frequency bands in studying human affective perception.

show abstract

Section: Introductionmentioning

confidence: 58%

Section: Neural Signature Of Emotional States In the High Gamma Bandmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

High Gamma Band EEG Closely Related to Emotion: Evidence From Functional Network

Yang

Shu

et al. 2020

Front. Hum. Neurosci.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Neuropsychophysiological connectivity measures could also be applied to the assessment of negative emotionality. In healthy population, increased long-range connections between frontal parietal and temporal areas in beta and gamma bands are related to negative and positive emotional information processing ( 137 ). In this way, fast rhythms communication through the brain could play a role in emotional processing and could be of use in the research of neural markers of alterations in AUD.…”

Section: Evidence In Other Neuropsychophysiological Measuresmentioning

confidence: 99%

Neuropsychophysiological Measures of Alcohol Dependence: Can We Use EEG in the Clinical Assessment?

et al. 2020

View full text Add to dashboard Cite

Addiction management is complex, and it requires a bio-psycho-social perspective, that ought to consider the multiple etiological and developmental factors. Because of this, a large amount of resources has been allocated to assess the vulnerability to dependence, i.e., to identify the processes underlying the transition from substance use to dependence, as well as its course, in order to determine the key points in its prevention, treatment, and recovery. Consequently, knowledge \from neuroscience must be taken into account, which is why different initiatives have emerged with this objective, such as the “Research Domain Criteria” (RDoC), and the “Addiction Neuroclinical Assessment” (ANA). Particularly, neuropsychophysiological measures could be used as markers of cognitive and behavioral attributes or traits in alcohol dependence, and even trace clinical change. In this way, the aim of this narrative review is to provide an overview following ANA clinical framework, to the most robust findings in neuropsychophysiological changes in alcohol dependence, that underlie the main cognitive domains implicated in addiction: incentive salience, negative emotionality, and executive functioning. The most consistent results have been found in event-related potential (ERP) analysis, especially in the P3 component, that could show a wide clinical utility, mainly for the executive functions. The review also shows the usefulness of other components, implicated in affective and substance-related processing (P1, N1, or the late positive potential LPP), as well as event-related oscillations, such as theta power, with a possible use as vulnerability or clinical marker in alcohol dependence. Finally, new tools emerging from psychophysiology research, based on functional connectivity or brain graph analysis could help toward a better understanding of altered circuits in alcohol dependence, as well as communication efficiency and effort during mental operations. This review concludes with an examination of these tools as possible markers in the clinical field and discusses methodological differences, the need for more replicability studies and incipient lines of work. It also uses consistent findings in psychophysiology to draw possible treatment targets and cognitive profiles in alcohol dependence.

show abstract

“…Recently, many research studies attempted to adapt brain connectivity as a proper framework to explore the dimensions of emotion and to improve emotion recognition [15][16][17]. Brain connectivity can inform us about how different brain areas communicate [18] and are encapsulating complementary information compared to brain activity [19].…”

Section: Introductionmentioning

confidence: 99%

A Multiplex Connectivity Map of Valence-Arousal Emotional Model

2020

View full text Add to dashboard Cite

A high number of studies have already demonstrated an electroencephalography (EEG)-based emotion recognition system with moderate results. Emotions are classified into discrete and dimensional models. We focused on the latter that incorporates valence and arousal dimensions. The mainstream methodology is the extraction of univariate measures derived from EEG activity from various frequencies classifying trials into low/high valence and arousal levels. Here, we evaluated brain connectivity within and between brain frequencies under the multiplexity framework. We analyzed an EEG database called DEAP that contains EEG responses to video stimuli and users' emotional self-assessments. We adopted a dynamic functional connectivity analysis under the notion of our dominant coupling model (DoCM). DoCM detects the dominant coupling mode per pair of EEG sensors, which can be either within frequencies coupling (intra) or between frequencies coupling (cross-frequency). DoCM revealed an integrated dynamic functional connectivity graph (IDFCG) that keeps both the strength and the preferred dominant coupling mode. We aimed to create a connectomic mapping of valence-arousal map via employing features derive from IDFCG. Our results outperformed previous findings succeeding to predict in a high accuracy participants' ratings in valence and arousal dimensions based on a flexibility index of dominant coupling modes.

show abstract

EEG Based Emotion Recognition by Combining Functional Connectivity Network and Local Activations

Cited by 276 publications

References 66 publications

High Gamma Band EEG Closely Related to Emotion: Evidence From Functional Network

High Gamma Band EEG Closely Related to Emotion: Evidence From Functional Network

Neuropsychophysiological Measures of Alcohol Dependence: Can We Use EEG in the Clinical Assessment?

A Multiplex Connectivity Map of Valence-Arousal Emotional Model

Contact Info

Product

Resources

About