Contrastive learning of shared spatiotemporal EEG representations across individuals for naturalistic neuroscience

• Published in: NeuroImage (Orlando, Fla.) Vol. 301; p. 120890
• Main Authors: Shen, Xinke, Tao, Lingyi, Chen, Xuyang, Song, Sen, Liu, Quanying, Zhang, Dan
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.11.2024; Elsevier Limited; Elsevier
• Subjects: Audiobooks; Brain research; Contrastive learning; Datasets; EEG; Electroencephalography; Emotions; Information processing; Inter-subject correlation; Methods; Naturalistic neuroscience; Neural networks; Neurosciences; Sensors; Spatial discrimination learning; Speech; Inter-subject correlation; Naturalistic neuroscience; EEG; Contrastive learning; contrastive learning; naturalistic neuroscience

•Contrastive learning was employed to align EEG patterns across subjects.•Higher intersubject correlation was obtained than state-of-the-art methods.•The model extracts latent patterns that reflect stimulus-relevant properties. Neural representations induced by naturalistic stimuli offer insights in...