An Agent-based model of the gradual emergence of modern linguistic complexity

Abstract

A central question in the evolution of human language is whether it emerged as a result of one specific event or from a mosaic-like constellation of different phenomena and their interactions. Three potential processes have been identified by recent research as the potential primum mobile for the origins of modern linguistic complexity: Self-domestication, characterized by a reduction in reactive aggression and often associated with a gracilization of the face; changes in early brain development manifested by globularization of the skull; and demographic expansion of H. sapiens during the Middle Pleistocene. We developed an agent-based model to investigate how these three factors influence transmission of information within a population. Our model shows that there is an optimal degree of both hostility and mental capacity at which the amount of transmitted information is the largest. It also shows that linguistic communities formed within the population are strongest under circumstances where individuals have high levels of cognitive capacity available for information processing and there is at least a certain degree of hostility present. In contrast, we find no significant effects related to population size.