ABSTRACT

The most important message from this research is that the conceptual system is organised categorically from its inception. Furthermore, it does not appear necessary to start the process off by building in an initial division of animal versus artefact, as several researchers have suggested (Gelman, 1990; Spelke, 1994). The available data suggest that this distinction is easily learnable from the spatial and movement information presented by the visual system. A related issue is whether there are different rates of learning distinctions within these large domains. Caramazza and Shelton (1998) suggested an evolutionary basis for an animate-inanimate distinction because the domain-level categories of animals, plants, and artefacts can be damaged independently of each other. They suggested that this might be due to “specific adaptations for recognising and responding to animal and plant life”. If there were dedicated mechanisms for processing animals and plants, one would expect learning different animal and plant kinds to be especially easy. Although Caramazza and Shelton (1998) do not put it in this way, I take this to be the implication of their hypothesis that there are “specialized mechanisms for the recognition and categorisation of the members of the categories for which specific adaptations have evolved”. However, our data say that conceptualising animal and plant kinds is more difficult for infants than conceptualising kinds of artefacts. The type of learning infants do appears to be the same for animals, plants, and artefacts but, at least in our culture, it is slower for animals and plants. This difference is not what one would predict from an evolutionary hypothesis that emphasises special adaptations for learning in these domains. More seriously for an evolutionary argument, perhaps, even American adults are very poor at distinguishing different kinds of plants and animals, especially non-mammals. As Rosch et al. (1976) discovered, most American college students have little idea about the properties that distinguish one fish from another or one kind of deciduous tree from another, both in terms of what they look like and their other characteristic features. These properties are either taught by the culture or tend to remain unlearned.