Temporal properties of higher-order interactions in social networks

Abstract Human social interactions in local settings can be experimentally detected by recording the physical proximity and orientation of people. Such interactions, approximating face-to-face communications, effectively represented as time varying networks with links being unceasingly created destroyed over time. Traditional analyses temporal have addressed mostly pairwise where describe dyadic connections among individuals. However, many network dynamics are hardly ascribable to but often comprise larger groups, which better described higher-order interactions. Here we investigate organizations analyzing five publicly available datasets collected different settings. We find that ubiquitous and, similarly their counterparts, characterized heterogeneous dynamics, bursty trains rapidly recurring events separated long periods inactivity. evolution formation groups looking at transition rates between structures. more spontaneous settings, group slower disaggregation, while work these phenomena abrupt, possibly reflecting pre-organized dynamics. Finally, observe reinforcement suggesting longer a stays together higher probability same interaction pattern persist future. Our findings suggest importance considering structure when investigating human