Abstract:
In this work, we investigated the emergence of diversity in language. Fundamental game theoretical model for emergence of language has been established and used in several studies. In these studies most basic assumption is that understanding everyone in the population gives evolutionary advantage to an agent. However, being in interaction with everyone is not achievable in reality where various limited resources such as memory and physical availability bind agents. Thus population should be organized in a way that each agent can interact only some percentage of population. In our model, agents select their teachers, who are responsible for the transmission of language, from neighbors of their parents, which forms a social network in population. Our ndings include that emerged number of di erent languages in population is related to the percentage size of neighborhood and is independent of population size. Furthermore, we observe that seeking language-wise similarity in teacher selection makes no evolutionary di erence in emergence of language, instead of seeking physical closeness. As a result, we see this study as an important contribution towards understanding the emergence of diversity in languages.
