Data Associated with "Dancing on the Edge: Honey Bee Recruitment Networks are Sparse and Affected by Individual Waggle Dance Calibrations"

Here we studied a particular kind of social network in honey bees: the recruitment of nestmates to profitable food resources via the waggle dance.
We housed experimental colonies of approximately 3,000 individually-marked worker honey bees (plus the queen) in observation hives, trained bees to collect sugar solution from an artificial feeder, and then traced the spread of information about the location of the feeder as bees waggle danced to recruit one another to it.
First, using these empirical data, we calculated network density for our observed recruitment networks. To do this in a biologically relevant way, we used deterministic susceptible-infected models to simulate empirical upper bounds for density values, rather than using the default reference of a complete graph. We emphasize that calculating and interpreting bread-and-butter network descriptive statistics taking the biological constraints of the system at hand into account is an important step in studying social networks in animals.
Second, we calculated burstiness of waggle dance bouts in the observed networks, and found that bees danced regularly, not burstily.
Finally, we investigated the role of individuality in waggle dance behavior in network structure. Each honey bee communicates the distance to a resource in a part of the waggle dance called the waggle run, whose duration is proportional to the distance she flew. However, bees differ in the proportion by which they scale the real-world distance to a waggle run duration. We found that this individuality supported network propagation when dancers had higher calibrations than their followers did.
We hope that this work will be interesting across bee biology as well as bio-inspired computing.