When Language Evolution Meets Multimodality: Current Status and Challenges Toward Multimodal Computational Models

Computational models can be considered human-designed computing models inspired by the processes observed in the natural world, which allow simulating and understanding these processes. Computational modelling is notably applied to simulate the behaviour and long-term dynamics of human Language. The research effort made so far in computational modelling of language evolution considers predominantly one modality by arguing for a unimodal origin of Language. This article extends this paradigm to a new perspective that integrates into its structure and learning algorithms principles from multimodal communication. This article gives an overview of the current language evolution models. It discusses the key challenges towards multimodal language evolution modelling by envisioning a conceptual framework to design the multimodal grounding and the language learning processes, as well as their realisation through a multi-agent multimodal referential game. This framework is valuable for many researchers working on language evolution to reveal the key questions they should address and integrate for pursuing a holistic vision that combines all modalities in a multimodal language evolution model.