The Evolution of Meaning-space Structure through Iterated Learning

In order to persist, language must be transmitted from generation to generation through a repeated cycle of use and learning. This process of iterated learning has been explored extensively in recent years using computational and mathematical models. These models have shown how compositional syntax provides language with a stability advantage and that iterated learning can induce linguistic adaptation. This paper presents an extension to previous idealised models to allow linguistic agents flexibility and choice in how they construct the semantics of linguistic expressions. This extension allows us to examine the complete dynamics of mixed compositional and holistic languages, look at how semantics can evolve culturally, and how communicative contexts impact on the evolution of meaning structure.