Classifying different emotional states by means of EEG-based functional connectivity patterns

Area of Science:

• Neuroscience
• Psychology
• Biomedical Engineering

Background:

• Understanding the neural correlates of emotional states is crucial.
• Electroencephalography (EEG) offers a non-invasive method to measure brain activity.
• Functional connectivity analysis reveals how different brain regions interact.

Purpose of the Study:

• To classify emotional states using EEG-based functional connectivity.
• To investigate the relationship between brain activity and subjective affect.
• To evaluate the efficacy of pattern classification for emotion recognition.

Main Methods:

• Forty young participants viewed film clips inducing neutral, positive, or negative emotions.
• EEG signals were recorded to estimate functional connectivity using correlation, coherence, and phase synchronization.
• Quadratic Discriminant Analysis was employed for pattern classification of emotional states.

Main Results:

• Significant differences in EEG-based functional connectivity were observed across emotional states.
• Pattern classification analysis successfully distinguished between emotional states with accuracy above chance.
• Connectivity patterns varied distinctly depending on the evoked emotional valence.

Conclusions:

• EEG-based functional connectivity is a viable tool for classifying emotional states.
• This approach provides insights into the neural dynamics underlying emotions.
• The findings support the use of EEG for objective emotion assessment.