Re: Modularity

From: HARNAD Stevan (harnad@cogsci.soton.ac.uk)
Date: Tue Jun 04 1996 - 20:53:35 BST


> From: "Hale, Pippa" <pjh295@soton.ac.uk>
> Date: Fri, 24 May 1996 13:32:14 +0100 (BST)
>
> The actual word "module" is used differently by different people.
> Neuroscientists use it to discuss the fact that the brain is structured
> with cells and layers that divide the processing of information as it
> enters the brain.

Kid-sib wouldn't know what you meant by that.

> Jerry Fodor wrote a book which he believed described modularity. In
> this it explained that a module, in cognitive science, is a specialised
> organ within the brain that handles only it own specialised information
> and it is genetically specified. The information is usually related
> only to a specific species. An example of a module, given by Fodor, is
> the human language.

The origin of the idea of cognitive modules was in Chomsky's theory that
syntax is independent of sematics in natural language: We can tell
whether or not a sentence is grammatically correct independently of what
it might mean.

> Fodor specified that nine criteria must be met to define modules as
> cognitive systems. Five of which define the way modules process
> information, three are for the biological status (to distinguish between
> learned habits and behavioural systems). The other is domain
> specificity, which is the fact that the module only deals with one type
> of information.

Kidpsib would again not get much out of that; you should explain, not
just name or list.

> Each module within the brain receives external information indirectly.
> The information is first transformed so that the specific module can
> interpret it. Each module then produces information for the central
> processing. Modules are believed to be unintelligent and inflexible.

Not a bad job, considering that I never went on to cover this topic.
An A would have required an explanation of what it means for one module
to be independent of another; examples are needed too.



This archive was generated by hypermail 2b30 : Tue Feb 13 2001 - 16:23:44 GMT