Social interaction is a dynamic and variable process.However,most hyperscanning studies implicitly assume that inter-brain synchrony(IBS)during task durations is constant and rarely investigate the temporal variability of the multi-brain networks.This study aims to explore the dynamic patterns of IBS during group creative activities,particularly the underlying cognitive and neural basis.In study 1,a representative set of inter-brain network states during different group communication tasks was obtained using f NIRS-based hyperscanning and a dynamic analysis method based on sliding window and K-means clustering.Results of network parameters and temporal occurrence rates showed that dense and efficient inter-brain states and sparse and inefficient inter-brain states alternated periodically,and the occurrence rate of efficient inter-brain states was positively correlated with cooperative behavior and group creativity performance.Moreover,compared to ordinary communication tasks,the occurrence rate and frequency of state transitions of efficient inter-brain states during creative communication tasks were significantly higher,indicating a more active and intertwined neural network.These findings suggest a close correspondence between inter-brain synchronization states and social behavior.In Study 2,momentary inter-brain network states during viewpoint generation of different creativity levels in group creative tasks were obtained using a behavior locking method.Results showed that high creativity answers exhibited efficient inter-brain states and higher intensity of the mirror and mentalizing systems compared to the entire task duration and low creativity answer moments,which may reflect higher levels of mutual understanding and more perspective-taking behaviors among group members during this period.In addition,a predictive model of group creativity performance was established using machine learning methods based on instantaneous inter-brain network states during high creativity answer moments,which showed higher prediction accuracy.The main predictive contributions were from core regions of the default network,executive control network,and mirror neuron network.These results suggest that dynamic inter-brain networks better match the group brain activity pattern during the task period and reflect the neural basis of group creative activities.Based on the findings of the two studies,the inter-brain neural coupling in creative communication within a group shows a dynamic pattern.When the inter-brain neural coupling exhibits a dense and efficient pattern,group members may have engaged in more effective information exchange,which may lead to better innovative ideas.In group creative communication,the increase in brain synchrony in the default network,executive control network,and mirror neuron network-related brain areas of the members may be the neural basis behind the dynamic pattern of group creative communication between brains. |