Division of labour and the evolution of extreme specialization

Abstract

Division of labour is a common feature of social groups, from biofilms to complex animal societies. However, we lack a theoretical framework that can explain why division of labour has evolved on certain branches of the tree of life but not others. Here, we model the division of labour over a cooperative behaviour, considering both when it should evolve and the extent to which the different types should become specialized. We found that: (1) division of labour is usually-but not always-favoured by high efficiency benefits to specialization and low within-group conflict; and (2) natural selection favours extreme specialization, where some individuals are completely dependent on the helping behaviour of others. We make a number of predictions, several of which are supported by the existing empirical data, from microbes and animals, while others suggest novel directions for empirical work. More generally, we show how division of labour can lead to mutual dependence between different individuals and hence drive major evolutionary transitions, such as those to multicellularity and eusociality.