Broadbent, D. (1985). A question of levels: Comments on McClelland and Rumelhart. Journal of Experimental Psychology: General, 114, 189-192.  Home/Search   Document Not in Database   Summary   Related Articles   Check

.... We are stressing the Representationalist character of Connectionist theorizing because much Connectionist methodological writing has been preoccupied with the question What level of explanation is appropriate for theories of cognitive architecture (see, for example, the exchange between Broadbent, 1985, and Rumelhart McClelland, 1985) And, as we re about to see, what one says about the levels question depends a lot on what stand one takes about whether there are representational states. It seems certain that the world has causal structure at very many different levels of analysis, with the ....

Broadbent, D. (1985). A question of levels: Comments on McClelland and Rumelhart. Journal of Experimental Psychology: General, 114, 189-192.

.... science has become the issue of levels [Estes 1988] must connectionist models be interpreted at the level of physical instantiation 134 Connectionist explanation: taking positions in the mind brain dilemma or at the level of symbol manipulation Critics of the connectionist movement (e.g. [Broadbent 1985] and [Fodor and Pylyshyn 1988] argue that it cannot be considered an alternative to the classical cognitivist paradigm. They characterize connectionism as an attempt to define a brain like implementation of symbol manipulating models. Supporters of connectionism, however, emphasize that it is an ....

....of this type of connectionist models is made explicit. The conceptual dependence of these models on cognitivism implies that they have to deal with the standards set in that paradigm. They must answer the criticism from the cognitivist tradition (like [Fodor and Pylyshyn 1988] and [Broadbent 1985]) for instance that the representations built in connectionist models must satisfy constraints such as compositionality. Moreover, the analysis presented shows that the contribution of these models to our understanding of mental phenomena is limited since they only echo what their designers had ....

Broadbent D.: A question of levels: comments on McClelland and Rumelhart, Journal of Experimental Psychology 114, 189--192, 1985.

.... With this we want to highlight that arguments such as anything your system can do, ours can too, so what s new with your system , may not be valid arguments when comparing the two systems we are interested in (this point is also made by Rumelhart and McClelland (1985) in their response to Broadbent (1985)) 2 Representations The computational level of the classical cognitive architecture describes what a system computes, and why. At this level, the two systems of interest here compute the same function, i.e. their 1 external behaviour is indistinguishable. We have to move one level down from ....

Broadbent, D. (1985). A question of levels -- comments on McClelland and Rumelhart. Journal of Experimental Psychology General, 114(2):189--192.

....years. With this we want to highlight that arguments such as anything your system can do, ours can too, so what s new with your system , may not be valid arguments when comparing the two systems we are interested in (this point is also made by Rumelhart McClelland (1985) in their response to Broadbent (1985)) 2 Representations The computational level of the classical cognitive architecture describes what a system computes, and why. At this level, the two systems of interest here compute the same function, i.e. their external behaviour is indistinguishable. We have to move one level down from the ....

Broadbent, D. (1985). A question of levels -- comments on McClelland and Rumelhart. Journal of Experimental Psychology General, 114(2):189--192.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC