Categories and word meanings are not single entities in the mind

But the idea that categories and word meanings are single mental entities can also have a distributed connectionist interpretation. In distributed connectionist models an entity is not represented by a single network unit as in localist connectionist models. However, categories or word meanings can be considered as corresponding to some particular pattern of activation on an ensemble of network units and therefore still interpreted as single entities. Categories and word meanings tend to be interpreted as single entities that are activated in the mind when an individual categorizes the world or uses language. We argue that this view is mistaken. Connectionist simulations of populations of neural networks that live and reproduce in an environment show that categories and word meanings are open and changing collections of activation patterns in the network’s internal units rather than single activation patterns.