A Systematic Review for Human EEG Brain Signals Based Emotion Classification, Feature Extraction, Brain Condition,...

Area of Science:

• Neuroscience and Affective Computing
• Interdisciplinary research combining electroencephalography (EEG) and music psychology.

Background:

• Electroencephalography (EEG) signal analysis for human emotion detection is an emerging research area.
• Existing research is fragmented across academic databases, necessitating a consolidated overview.
• Music's influence on human emotions, measured via brain signals, is a key focus.

Purpose of the Study:

• To systematically review and categorize academic articles on EEG and music-evoked human emotions.
• To map the research landscape and establish a taxonomy for this field.
• To identify trends, challenges, and future directions in EEG-based emotion research.

Main Methods:

• Systematic literature search across ScienceDirect, Web of Science, and IEEE Xplore (1999-2016).
• Three-iteration screening process: duplicate exclusion, title/abstract filtering, and full-text review.
• Categorization of 100 selected articles into five distinct classes based on research focus.

Main Results:

• Articles were classified into: emotion classification using AI (39%), EEG techniques (21%), feature extraction (8%), comparative studies (26%), and music as stimulus (6%).
• Analysis covered study characteristics such as participant demographics, music listening duration, and publication origins.
• Key characteristics, challenges, and recommendations for EEG-based emotion measurement were identified.

Conclusions:

• The review provides a taxonomy of EEG and music-emotion research, identifying AI's role in classification.
• It highlights the need for further research into EEG signal analysis for understanding music's emotional impact.
• Recommendations are proposed to advance the utilization of EEG in affective computing and neuroscience.