Identifying relevant EEG channels for subject-independent emotion recognition using attention network layers
BackgroundElectrical activity recorded with electroencephalography (EEG) enables the development of predictive models for emotion recognition. These models can be built using two approaches: subject-dependent and subject-independent. Although subject-independent models offer greater practical utilit...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-02-01
|
| Series: | Frontiers in Psychiatry |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fpsyt.2025.1494369/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823860978479005696 |
|---|---|
| author | Camilo E. Valderrama Camilo E. Valderrama Anshul Sheoran |
| author_facet | Camilo E. Valderrama Camilo E. Valderrama Anshul Sheoran |
| author_sort | Camilo E. Valderrama |
| collection | DOAJ |
| description | BackgroundElectrical activity recorded with electroencephalography (EEG) enables the development of predictive models for emotion recognition. These models can be built using two approaches: subject-dependent and subject-independent. Although subject-independent models offer greater practical utility compared to subject-dependent models, they face challenges due to the significant variability of EEG signals between individuals. ObjectiveOne potential solution to enhance subject-independent approaches is to identify EEG channels that are consistently relevant across different individuals for predicting emotion. With the growing use of deep learning in emotion recognition, incorporating attention mechanisms can help uncover these shared predictive patterns.MethodsThis study explores this method by applying attention mechanism layers to identify EEG channels that are relevant for predicting emotions in three independent datasets (SEED, SEED-IV, and SEED-V). ResultsThe model achieved average accuracies of 79.3% (CI: 76.0-82.5%), 69.5% (95% CI: 64.2-74.8%) and 60.7% (95% CI: 52.3-69.2%) on these datasets, revealing that EEG channels located along the head circumference, including Fp1, Fp2, F7, F8, T7, T8, P7, P8, O1, and O2, are the most crucial for emotion prediction. ConclusionThese results emphasize the importance of capturing relevant electrical activity from these EEG channels, thereby facilitating the prediction of emotions evoked by audiovisual stimuli in subject-independent approaches. |
| format | Article |
| id | doaj-art-a044dad0b7554657b92dbc92e4c7d019 |
| institution | Kabale University |
| issn | 1664-0640 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Psychiatry |
| spelling | doaj-art-a044dad0b7554657b92dbc92e4c7d0192025-02-10T06:48:53ZengFrontiers Media S.A.Frontiers in Psychiatry1664-06402025-02-011610.3389/fpsyt.2025.14943691494369Identifying relevant EEG channels for subject-independent emotion recognition using attention network layersCamilo E. Valderrama0Camilo E. Valderrama1Anshul Sheoran2Department of Applied Computer Science, University of Winnipeg, Winnipeg, MB, CanadaDepartment of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, AB, CanadaDepartment of Applied Computer Science, University of Winnipeg, Winnipeg, MB, CanadaBackgroundElectrical activity recorded with electroencephalography (EEG) enables the development of predictive models for emotion recognition. These models can be built using two approaches: subject-dependent and subject-independent. Although subject-independent models offer greater practical utility compared to subject-dependent models, they face challenges due to the significant variability of EEG signals between individuals. ObjectiveOne potential solution to enhance subject-independent approaches is to identify EEG channels that are consistently relevant across different individuals for predicting emotion. With the growing use of deep learning in emotion recognition, incorporating attention mechanisms can help uncover these shared predictive patterns.MethodsThis study explores this method by applying attention mechanism layers to identify EEG channels that are relevant for predicting emotions in three independent datasets (SEED, SEED-IV, and SEED-V). ResultsThe model achieved average accuracies of 79.3% (CI: 76.0-82.5%), 69.5% (95% CI: 64.2-74.8%) and 60.7% (95% CI: 52.3-69.2%) on these datasets, revealing that EEG channels located along the head circumference, including Fp1, Fp2, F7, F8, T7, T8, P7, P8, O1, and O2, are the most crucial for emotion prediction. ConclusionThese results emphasize the importance of capturing relevant electrical activity from these EEG channels, thereby facilitating the prediction of emotions evoked by audiovisual stimuli in subject-independent approaches.https://www.frontiersin.org/articles/10.3389/fpsyt.2025.1494369/fullemotion recognitionelectroencephalogramaffective computingdeep learningattention mechanismEEG signal processing |
| spellingShingle | Camilo E. Valderrama Camilo E. Valderrama Anshul Sheoran Identifying relevant EEG channels for subject-independent emotion recognition using attention network layers Frontiers in Psychiatry emotion recognition electroencephalogram affective computing deep learning attention mechanism EEG signal processing |
| title | Identifying relevant EEG channels for subject-independent emotion recognition using attention network layers |
| title_full | Identifying relevant EEG channels for subject-independent emotion recognition using attention network layers |
| title_fullStr | Identifying relevant EEG channels for subject-independent emotion recognition using attention network layers |
| title_full_unstemmed | Identifying relevant EEG channels for subject-independent emotion recognition using attention network layers |
| title_short | Identifying relevant EEG channels for subject-independent emotion recognition using attention network layers |
| title_sort | identifying relevant eeg channels for subject independent emotion recognition using attention network layers |
| topic | emotion recognition electroencephalogram affective computing deep learning attention mechanism EEG signal processing |
| url | https://www.frontiersin.org/articles/10.3389/fpsyt.2025.1494369/full |
| work_keys_str_mv | AT camiloevalderrama identifyingrelevanteegchannelsforsubjectindependentemotionrecognitionusingattentionnetworklayers AT camiloevalderrama identifyingrelevanteegchannelsforsubjectindependentemotionrecognitionusingattentionnetworklayers AT anshulsheoran identifyingrelevanteegchannelsforsubjectindependentemotionrecognitionusingattentionnetworklayers |