Talk:Computational theory of mind

From Wikipedia, the free encyclopedia

Socrates
Philosophy Portal
This article is within the scope of the WikiProject Philosophy, which collaborates on articles related to philosophy. To participate, you can edit this article or visit the project page for more details.
??? This article has not yet received a rating on the quality scale.
??? This article has not yet received an importance rating on the importance scale.

Contents

[edit] Remove minimal computationalism

I think this section should just be an external link, and I will make it so if there are no objections. The article linked is pretty crappy anyway, making a lot of amateur mistakes in interpreting complex mathematical statements into a philosophy. The idea of minimal computationalism is certainly not a dominant perspective in the field, and doesn't even seem to be a major perspective, so I believe having its own section in lieu of other conjectures shows bias. SamuelRiv 06:02, 10 November 2007 (UTC)

[edit] Todo

The introduction of this article seems accurate to me, however the article is far from comprehensive or encyclopedic, so I've marked it with an expert tag.

  1. Describe the roots of the idea in Hobbes, Descartes, Liebniz, Hume.
  2. Describe the influence of Alan Newell and Herbert Simon's work 1956-1970, and how this inspired the movement.
  3. Y Done Describe the behaviorist milieu that computationalism was designed to refute.
  4. Y Done Mention Jerry Fodor's version.
  5. Y Done Mention Hilary Putnam's version.
  6. Mention Zenon Pylyshyn's version.
  7. Y Done Mention the popularity of the idea with folks like Steven Pinker and Daniel Dennett
  8. Refute the idea: John Searle, Hubert Dreyfus.

Good references:
Stanford Encyclopedia of Philosophy: The Computational Theory of Mind
Pinker, Steven How the Mind Works
Haugaland, John Artificial Intelligence: The Very Idea", 1986ish,
 as well as the (unused) references given for the article. ---- CharlesGillingham (talk) 07:59, 27 December 2007 (UTC)

I removed Bruno Marchal, because I don't believe that his ideas (although they are fascinating) are a central thread of computational theory of mind. ---- CharlesGillingham (talk) 02:03, 1 June 2008 (UTC)
I removed GOFAI, because I already mentioned Newell and Simon. They explicitly claimed that their programs were models of human cognition. Other researchers in the "GOFAI" tradition (John McCarthy, Marvin Minsky, Seymour Papert, Edward Feigenbaum, etc) did not, and some (McCarthy especially) argued that human cognition and machine cognition were essentially different. ---- CharlesGillingham (talk) 02:03, 1 June 2008 (UTC)

[edit] Further Refutation

I think the above ToDo list is a great start for improving this article. In response to the "who else?" question above I submit Hubert Dreyfus's What Computers Can't Do and the follow-up What Computers Still Can't Do. Hubert Dreyfus is a philosopher at UC Berkeley. He also co-wrote Mind Over Machine with his brother Stuart Dreyfus, a professor in industrial engineering also at UC Berkeley. Cheers, Wolfworks (talk) 15:38, 13 May 2008 (UTC)

[edit] new material

This page needs a lot of work, and I like fixer-uppers. I can't do the whole to-do list myself, but I can add some meat. Leadwind (talk) 19:05, 31 May 2008 (UTC)

[edit] Removed

I removed this unsourced section, originally added by User:Peterburton.

Overview

To put substance into this metaphor, at least three components of a 'computable' system model must be specified. Firstly, the data-structure which specifies the least coherent element of 'computation' must be identified. Secondly, the rules of syntax under which these least data-structures may be combined must be specified. Thirdly, some plausible form of brain control over these data-structures must be invoked.

As with all computation, the elegance and flexibility of the final 'program' is largely dependent upon the elegance of the data-structure definitions, around which other issues revolve. In the real brain, presumably the problem is one of finding a data-structure model at the right degree of abstraction such that contact remains with the active neuroscience of the real brain while contact is gained with the process attributes of a mind. The barrier to the latter has been a sufficiently scientific conception of consciousness, surely the precursor concept of any mind, that could even in principle be engineered.

One approach that seeks a resolution of these issues is the Cognitive Process Consciousness model, which seeks to identify human consciousness with a 'computable' and defined system of cognitive processes. Computational Theory is an advanced subject used extensily in the aritificial intelligence field.

This section does not, it seems to me, present a standard introduction to the idea of computationalism. ----CharlesGillingham (talk) 02:50, 1 June 2008 (UTC)

[edit] Looking better

Thanks to some good work by User:Leadwind, this article is looking better. It still needs work, of course, but at least it mentions the right names and stays on topic. I've removed the "expert" tag for now. ---- CharlesGillingham (talk) 03:00, 1 June 2008 (UTC)