Decoding Human Interaction Type from Inter-brain Synchronization by Using EEG Brain Network.

Area of Science:

• Neuroscience
• Social Psychology
• Computational Neuroscience

Background:

• Interpersonal interactions, including cooperation and competition, are fundamental human behaviors.
• Understanding the neural basis of these interactions is crucial for deciphering social cognition.
• Previous research highlighted electrode-paired inter-brain synchronization, but a network-scale perspective was lacking.

Purpose of the Study:

• To investigate neural correlates of interpersonal synchronization at the brain network scale.
• To differentiate network-wise inter-brain synchronization (NIBS) between cooperative and competitive interactions.
• To develop a computational model for classifying interaction types based on NIBS.

Main Methods:

• Advanced a novel Network-wise Inter-Brain Synchronization (NIBS) index for global brain network analysis.
• Utilized electroencephalography (EEG) hyper-scanning data from participants engaged in cooperative and competitive tasks.
• Developed and applied a row-filtered depthwise separable convolution network for NIBS feature classification.

Main Results:

• Demonstrated statistically significant differences in NIBS between cooperative and competitive interactions.
• Found that cross-brain synchronization is more consistent during cooperative tasks compared to competitive ones.
• Achieved a peak classification accuracy of 96.05% for distinguishing cooperation from competition using the neural decoder.

Conclusions:

• NIBS provides a valuable metric for understanding neural dynamics during social interactions.
• Cooperative and competitive behaviors exhibit distinct network-level inter-brain synchronization patterns.
• Computational models can effectively decode interaction types from brain synchronization data.