A self-organizing spatial vocabulary
Artificial Life
| Hi-index | 0.00 |
To globally approach some language mechanisms, we model linguistic basic interactions by simulating an artificial life, dispatching reactive agents on a grid, where they seek food and reproduce. As a result, we provide a ten stages phylogenetic progression of beings, ranging from a basic vegetative cell, up to a social organization of communicating agents. At each level, agents are improved by new features, so we describe their internal functioning: neural connections between sensors and action effectors. We define a new semantics of a structure, based on it's efficiency to help an agent newly fitted with it. To evaluate the changing in the being fitness, we don't measure the new efficiency of its actions and behavior; but, we simulate in an artificial life, a group of these new agents, we simply measure how much they are, how long they survive; and we compare these new results with the past ones. Within this progression and this new semantics, we obtain a coherent frame to globally study some language mechanisms. When analyzing this got phylogenesis, we notice that the semantics emerges before lexicon and syntax. According to this point of view, we conclude that communication improves the group performances, and that the language seems to be the internal communication of a macro entity: the group.