phil 230 paper 2 (connectionist model)

phil 230 paper 2 (connectionist model) - 713966879...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
713966879 Philosophy 230 Professor Sabo 1498 words Connectionist and Classical Models of Cognition A luring alternative to classical models, connectionist networks are commonly advanced as a model of the cognitive mind. The connectionist model has become popular and is praised by a wide spectrum of academic fields because it seemingly holds several advantages over classical models. There are several key differences between the two and the connectionist model is still criticized by Jerry Fodor and Zenon Pylyshyn in their paper “Connectionism and Cognitive Architecture: A Critical Analysis.” This paper will examine what a connectionist model is, the major differences between the classical and connectionist networks, the advantages of connectionist networks, and examine Fodor and Pylyshyn’s criticisms of the connectionist networks. Connectionist models are essentially defined as a network with cognitive capabilities from interconnected “units,” or nodes. Connectionist models consist of three main parts: a connectionist architecture, nodes, and weighted connections. The connected architecture is the defining characteristic of a connectionist network. There are arbitrary numbers of interconnected nodes which form one interconnected network. Because of these connections the network is able to carry out its process in the manner it does; each node contributes equally to the final output. Nodes themselves are the individual processing units and function to receive input (information), process it, and produce an output (results). The weights on the connections are how the input becomes an output. Essentially, the weights are the computation of the each node; weights are passed from node to node. Subsequently, nodes influence each others’ computation. Because the network functions as a whole, the sum of all the weights on the entire network is the final output. Each node is identical except for its connection weight. As a process, connectionist networks
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
713966879 Philosophy 230 Professor Sabo 1498 words work by receiving information into the network, distributing the processing between the interconnected nodes and producing a final output. Connectionist models do not use symbols; they process syntax-less information to produce an output with semantical value. There are several crucial differences between connectionist networks and classical models, which Fodor and Pylyshyn highlight. They claim classical models have combinatorial syntax and semantics and structure sensitive processes (operations) while connectionist models use a parallel distributed process. Combinatorial syntax and semantics is that a set of basic
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 03/24/2008 for the course PHIL 230 taught by Professor Sabo during the Fall '07 term at UNC.

Page1 / 6

phil 230 paper 2 (connectionist model) - 713966879...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online