Social networks drive collective thermoregulation in honey bees

Social insects may be uniquely poised to deal with a changing environment, due to their ability to utilize both ecological and social information to behave. Further, they can coordinate to affect massive change within their colonies to regulate internal conditions. Although we have observed and defined many of these behaviors, we are still working to understand the mechanisms of this resiliency. How do animals coordinate to affect this change? We use honey bee thermoregulation to explore this question. Honey bees perform a collective fanning behavior to maintain internal temperatures at 35°C. Higher temperatures result in brood loss and shortened adult lifespans. Honey bees are more likely to fan in large groups and in higher density, and are less likely to fan when physically separated from nestmates. We utilized machine learning video tracking software to analyze social interactions before the fanning response was initiated. We found that more connected individuals were more likely to fan than less connected individuals. Developing social networks, especially for scalable behaviors like fanning, allow us to explore how components within a system communicate information, and how resiliency or failure of distributed systems may be built as a result of this communication.