Selection for group-level efficiency leads to self-regulation of population size

In general, a population will grow until a limiting factor, such as resource availability, is reached. However, increased task efficiency can also regulate the size of a population during task development. Through the use of digital evolution, we demonstrate that the evolution of a group-level task, requiring a small number of individuals, can cause a population to self-regulate its size, even in the presence of abundant energy. We also show that as little as a 1% transfer of energy from a parent group to its offspring produces significantly better results than no energy transfer. A potential application of this result is the configuration and management of real-world distributed agent-based systems.