Cross-subject EEG signals-based emotion recognition using contrastive learning

• Published in: Scientific reports Vol. 15; no. 1; pp. 28295 - 17
• Main Authors: Alghamdi, Ahmed Mohammed, Ashraf, M. Usman, Bahaddad, Adel A., Almarhabi, Khalid Ali, Al Shehri, Waleed A., Daraz, Amil
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 03.08.2025; Nature Publishing Group; Nature Portfolio
• Subjects: 639/166; 639/4077; Accuracy; Adult; Algorithms; Artificial Intelligence; Biochips; Brain; Brain - physiology; Brain-Computer Interfaces; CNN; Computer applications; Datasets; Deep learning; EEG; Electroencephalography; Electroencephalography - methods; Emotions; Emotions - physiology; Ensemble learning; Female; Humanities and Social Sciences; Humans; Implants; Literature reviews; Machine Learning; Male; multidisciplinary; Neural networks; Physiology; Science; Science (multidisciplinary); Signal processing; Signal Processing, Computer-Assisted; Wavelet transforms; Deep learning; CNN; Ensemble learning; Artificial Intelligence
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-025-13289-5

Electroencephalography (EEG) signals based emotion brain computer interface (BCI) is a significant field in the domain of affective computing where EEG signals are the cause of reliable and objective applications. Despite these advancements, significant challenges persist, including individual differences in EEG signals across subjects during emotion recognition. To cope this challenge, current study introduces a cutting-edge cross subject contrastive learning (CSCL) scheme for EEG signals representation of brain region. The proposed scheme addresses the generalisation across subjects directly, which is a primary challenge in EEG signals-based emotions recognition. The proposed CSCL scheme captures the complex patterns effectively by employing emotions and stimulus contrastive losses within hyperbolic space. CSCL is designed primarily to learn representations that can effectively distinguish signals originating from different brain regions. Further, we evaluate the significance of our proposed CSCL scheme on five different datasets, including SEED, CEED, FACED and MPED, and obtain 97.70%, 96.26%, 65.98%, and 51.30% respectively. The experimental results show that our proposed CSCL scheme demonstrates strong effectiveness while addressing the challenges related to cross subject variability and label noise in the EEG-based emotion recognition system.