Simulating evolution: connectionist metaphors for studying human cognitive behaviour

The use of intelligent systems to perform computer simulation in studying evolving behaviour is, we believe, beneficial for researchers interested in understanding the complexity and behaviour of the algorithms used in simulating evolution through computational learning. Intelligent systems based on connectionist networks or connectionism use computer simulation to talk about human consciousness, thought and evolution. We present a cognitive modelling study using neural networks to simulate childrens´ progression of language capabilities during the crucial two-year period (9-24 months) of child language development. We argue that in order to simulate the uniquely human language learning, it is very important to assess the aptness of connectionist architectures for cognitive modelling. We provide a connectionist interpretation to the basic aspects of cognition which are relevant to the simulation of child language development, thereby providing a mapping of psycholinguistic notions to connectionist concepts