Deep Learning-Based EEG Emotion Recognition: A Review.

Area of Science:

• Neuroscience
• Artificial Intelligence
• Human-Computer Interaction

Background:

• Emotion recognition is vital across healthcare, security, and HCI.
• Electroencephalographic (EEG) signals are key for non-concealable, brain-activity-based emotion detection.
• Deep learning integration with EEG is rapidly advancing emotion recognition.

Purpose of the Study:

• To review deep learning models used in EEG-based emotion recognition.
• To analyze network architectures, optimization, and feature-based designs.
• To address commonality-individuality limitations and propose future research directions.

Main Methods:

• Comprehensive literature review of deep learning models for EEG emotion recognition.
• Analysis of model designs, optimization strategies, and feature engineering.
• Discussion of commonality-individuality (C-I) challenges and a proposed assessment framework.

Main Results:

• Summary of prevalent deep learning architectures and their applications in EEG emotion recognition.
• Identification of limitations related to balancing generalizability and personalization (C-I).
• Proposal of a minimal C-I framework for model evaluation.

Conclusions:

• Deep learning significantly enhances EEG-based emotion recognition capabilities.
• Addressing the C-I gap is crucial for developing robust and personalized emotion AI.
• This review provides a reference for advancing the field towards more effective emotion recognition systems.