Re: Categorisation and Prototypes

From: HARNAD Stevan (harnad@cogsci.soton.ac.uk)
Date: Fri Jun 07 1996 - 23:01:35 BST


> From: "Murfitt, Gareth" <gjrm195@soton.ac.uk>
> Date: Wed, 29 May 1996 13:13:40 +0100 (BST)
>
> Categorisation in general is sorting things. Things fall into
> catagories of some sort, and we have to decide how to sort them. For
> example, an oak or a fir are in the catagory "tree", but a daffodil
> isn't in this catagory.
>
> The classical view of catagorisation is where the classification is
> decided by specific features. For example, cats can be classified as
> animals having both (a) fur and (b) a tail, and elephants, whilst it
> has a tail, does not have fur therefore it is not catagorised as a cat.
> Only creatures with both (a) and (b) are cats. This is clearly a very
> simplistic example.
> Catagories can be much more detailed and complicated than this, with
> many, many conditions. Eventually each catagory will only be filled by
> one individual item, but this is the issue of vanishing intersections,
> which we do not need to go into here. Basically, members of a catagory
> have certain necessary and sufficient condidtions for being a member of
> that catagory. One circumstance might be that each of two conditions is
> neccesary but neither is sufficient on its own, as shown in my earlier
> example.

You need to read the threads on Categorisation, Prototypes and
Vanishing Intersections to sort this ouut better. The Classical
View is meant to contrast with the Prototype view. They need to
be described. Kid-sib would have very little idea of what the
classical theory was from this reply.



This archive was generated by hypermail 2b30 : Tue Feb 13 2001 - 16:23:47 GMT