The most advanced form of sociality is eusociality, characterized by overlapping generations, cooperative care of the young, and reproductive division of labor, which includes sterility or near-sterility of the overwhelming majority of colony members.
With few exceptions, all the practitioners of eusociality are insects of the orders Hymenoptera (ants, bees, and wasps), Isoptera (termites), Thysanoptera (thrips), and Hemiptera (aphids).
Social insects can exhibit division of labor with respect to non-reproductive tasks, in addition to the aforementioned reproductive one.
Evolutionary biologists are still debating the fitness-advantage gained by social insects due to their advanced division of labor and task allocation, but hypotheses include: increased resilience against a fluctuating environment, reduced energy costs of continuously switching tasks, increased longevity of the colony as a whole, or reduced rate of pathogen transmission.
[9][10][11] Beyond the rationale, there is well-documented empirical evidence of communication related to tasks; examples include the waggle dance of honey bee foragers, trail marking by ant foragers such as the red harvester ants, and the propagation via pheromones of an alarm state in Africanized honey bees.
The term soldiers may be apt, as in Cephalotes, but in many species members of the larger caste act primarily as foragers or food processors.
In honeybees, the youngest workers exclusively clean cells, which is then followed by tasks related to brood care and nest maintenance from about 2–11 days of age.
[19] A dominant theory of explaining the self-organized division of labor in social insect societies such as honey bee colonies is the Response-Threshold Model.
[20] The Response-Threshold Model only provides for effective task allocation in the honey bee colony if thresholds are varied among individual workers.
This variation originates from the considerable genetic diversity among worker daughters of a colony due to the queen’s multiple matings.
[21] To explain how colony-level complexity arises from the interactions of several autonomous individuals, a network-based approach has emerged as a promising area of social insect research.
As decentralized networks, colonies are capable of distributing information rapidly which facilitates robust responsiveness to their dynamic environments.
Social insect networks are often non-randomly distributed, wherein a few individuals act as ‘hubs,’ having disproportionately more connections to other nestmates than other workers in the colony.
[23] Computer simulations of this particular interaction network demonstrated that inter-individual variation in connectivity patterns expedites information flow among nestmates.
Task allocation within a social insect colony can be modeled using a network-based approach, in which workers are represented by nodes, which are connected by edges that signify inter-node interactions.
[22] This approach is potentially problematic because connections between workers are not permanent, and some information is broadcast globally, e.g. through pheromones, and therefore does not rely on interaction networks.
To demonstrate how time and space constraints of individual-level interactions affect colony function, social insect network approaches can also incorporate spatiotemporal dynamics.
For example, the rate of information flow through Temnothorax rugatulus ant colonies is slower than would be predicted if time spent traveling and location within the nest were not considered.
They conclude that "... normalized matrix-input generalizations of Shannon's and Simpson's index ... should be the indices of choice when one wants to simultaneously examine division of labor amongst all individuals in a population".