Fodor, J. and Z. Pylyshyn. (1988). Connectionism and cognitive architecture: a critical analysis. Cognition 28: 3-71.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....a significant proportion of words only occur rarely [21, 22, 23] A language representation must be able to model the function of rarely occurring words. This characteristic is examined in Chapter 2. In a well known critique Fodor and Pylyshyn (1988) identify some key features in natural language [24]. These include systematicity, the ability to move from particular examples to general rules. They also focus on the related concept of compositionality, that is the way in which sequences of words can be grouped together into meaningful elements, both syntactically and semantically. At the ....

....sentences, and analogous linguistic features can be identified. 4. For the related concept of compositionality the syntactic element is captured, again within the bounds of our task. Fodor and Pylyshyn say compositionality argues for . syntactic = semantic structure in sentences [24], and we have shown that elements of syntactic structure can be modelled in our system. Semantic structure is not addressed. Using simplified English We note that the Perkins technical manuals written in PACE can be processed slightly more easily than a computer science text. Taking the ....

J A Fodor and Z W Pylyshyn. Connectionism and cognitive architecture: a critical analysis. Cognition, 1988.

....traces the development of HSP, describes an example of HSP and discusses nature and importance of HSP before nally discussing the current state of HSP. 2 The development of HSP The development of representations supporting HSP was fueled by the attack on connectionism by Fodor Pylyshyn[7] who claimed that NNs did not support symbol processing and that were a way around this to be found it would involve merely reimplementing classical symbolic architectures in a neural substrate. In response, several techniques for representing symbolic structures in NNs appeared, most notably the ....

J. A. Fodor and Z. W. Pylyshyn. Connectionism and cognitive architecture: a critical analysis. Cognition, 28(1-2):3-71, 1988.

....model construction are imposed. The recognition and retrieval of facts (a relation composed of several concepts) would be straight forward if they would be stored as attractors in the network. Nevertheless, when modeling fact retrieval, in light of economy and language productivity considerations [11], one cannot store every fact as an attractor on its own, and therefore another strategy should be pursued; facts will be represented as spatial temporal combinations of their constituents, which themselves will continue to be represented as attractors. Subsequently, the retrieval of facts is ....

Fodor J. and Pylyshyn Z.W., Connectionism and cognitive architecture: a critical analysis. Cognition 28: 3-71, 1988.

....a novel integration of connectionist and symbolic computation, and rests on both symbolic and connectionist technical advances [30, 31, 33, 34, 45, 44] e. SSP combines (a c) to shed new light on a central problem in the foundations of cognitive science, emphasized by Fodor f Pylyshyn [11] in their highly influential critique of counectionist theory: the explanation of how higher cognition can achieve, with finite and fixed resources, competence that is highly systematic, coherent, compositional, and productive [53, 571 . These results constitute direct progress in the ....

....their violation, are typically neglected. As part of the research on unccusativity in French including our HG work, we are carrying out detailed studies based on a data base of axceptability judgments which we have assembled: 8393 sentences involving 183 intransitive and 225 transitive verbs in 11 syntactic environments. Studies of the 3608 sentences in this data base involving intraasitive verbs have corroborated some of the claimed universal semantic and aspectua] tendencies, not corroborated others, and identified new regularities. Our conclusion at this point in our study is that, in ....

[Article contains additional citation context not shown here]

Jerry A. Fodor and Zenon W. Pylyshyn. Connectionism and cognitive architecture: A critical analysis. Cognition, 28:3-71, 1988.

....Whereas we can identify elements like red, green, apple, and pear each with a corresponding (active) neuron, the embodiment of the structural brackets has been much debated, as far back as Hebb, in 1949. Some have even argued that structural representation is impossible in neural networks [2]. In the context of visual perception, the main concern is how to represent and or detect conjunctions such as red and apple. Creating a red apple detector for every location on the retina seems too expensive, at least for every sensible conjunction. The straight forward solution, as also ....

J.A. Fodor and Z.W. Pylyshyn. Connectionism and cognitive architecture: a critical analysis. Cognition, 28:3--71, 1988.

....elements like red, green, apple, and pear each with a neuron that is activated when the element is used. However, the embodiment of the structural brackets has been much debated, as far back as Hebb [2] Some have even argued that such structural representation is impossible in neural networks [3]. # # # local feature detectors globalfeaturedetectors globalconjunctiondetector locationsonaninput grid,or retina Fig. 1. Global conjunction detection via aggregation of activity in local feature detectors. In the context of visual perception, the main concern is how to represent and or ....

J.A. Fodor and Z.W. Pylyshyn. Connectionism and cognitive architecture: a critical analysis. Cognition, 28:3--71, 1988.

....the acquisition and use of language[69] Furthermore, the entire connectionist endeavor represented by the PDP group was called into question as being merely associationist, and therefore not adequate to the tasks of syntactic and semantic representation involved in human cognition as a whole. [36] We use the terms neural and connectionist interchangeably in referring to these networks. The problem boils down to two facts about neural networks that at rst glance seem fundamentally incompatible with important properties of language. First, connectionist representations (at least, those ....

....vectors obtained by the correlation operators. From a more theoretical standpoint, BoltzCONS, with its direct encoding of traditional symbolic operations CAR, CDR, PUSH, and POP, seems to fall into the category of mere implementation to which connectionism is relegated by its strongest critics [36]. This thesis presents an attempt to address both sets of questions, the practical and theoretical,by means of an architecture called In nite RAAM. Representing a new fusion between recurrent neural networks and fractal geometry, In nite RAAM allows us to understand the behavior of these networks ....

[Article contains additional citation context not shown here]

J.A. Fodor and Z.W. Pylyshyn. Connectionism and cognitive architecture: A critical analysis. Cognition, 28:371, 1988.

....gained a wide popularity but have also attracted criticisms. The biological plausibility of neural models is controversial. There is, in fact, a wide gap in terms of complexity between biological and artificial neural systems. A more philosophical criticism was presented by Fodor and Phylyshyn [Fodor and Pylyshyn, 1988] who addressed the capacity of these models to account for sistematicity and compositionality. This critical position, somehow favourable to the symbolic approach, led to a violent debate [Smolensky, 1988, Fodor and McMaughlin, 1990] that showed how the cognitive science community was divided on ....

....placed in the framework of cognitive pragmatics. Two experiments, as well as the results of the simulations, are described. 4.1 Cognitive Modelling with Radial Basis Function Networks The symbolic tradition has been the dominant approach to cognitive modelling for a long time. Fodor and Pylyshyn [Fodor and Pylyshyn, 1988] two of the most representative theorists claim that only a system with symbolic representations, possessing constituent structure, can adequately model cognitive processes. Only the analysis at the level of symbolic processing, they maintain, is relevant to cognitive theorizing, and this ....

Fodor, J. and Pylyshyn, Z. (1988). Connectionism and cognitive architecture: A critical analysis. Cognition, 28:3--71.

....most artificial intelligence tasks such as language processing and reasoning the need arises to represent data structures such as sequences and trees. Originally, these tasks were predominantly tackled using symbolic approaches, while neural networks were claimed to lack this representation power [1]. However, to face this critisism, during the last decade a number of architectures and algorithms were designed demonstrating the capability of connectionist approaches to generate structured representations as well. The best known examples include recursive autoassociative memory [2] ....

J.A. Fodor and Z.W. Pylyshyn. Connectionism and cognitive architecture: A critical analysis. Cognition, 28:3--71, 1988.

....Symbolically Speaking 2 Abstract The ability to combine words into novel sentences has been used to argue that humans have symbolic language production abilities. Critiques of connectionist models of language often center on the inability of these models to generalize symbolically (Fodor Pylyshyn, 1988; Marcus, 1998) A connectionist model of sentence production is described which has variables that are inspired by distinctions in the visual system. For several tests of symbolic generalization, the model with variables is better able to generalize to novel sentences than a model without ....

.... blicket used in this manner. This ability to combine words and sentence frames in the absence of previous experience has led some researchers to argue that language requires symbolic capabilities, where knowledge about language is phrased in terms of variables and operations on those variables (Fodor Pylyshyn, 1988; Marcus, 1998; Pinker Prince, 1988) In addition to evidence that supports symbolic processing, there is also research that shows that people are recording the detailed statistical properties of the sentences that they are hearing and producing. One source of evidence for this is the role of ....

[Article contains additional citation context not shown here]

Fodor, J. A., & Pylyshyn, Z. (1988). Connectionism and cognitive architecture: A critical analysis. Cognition, 28, 3-71.

....firmly and technically based on connectionist primitives in a learning environment. The perspective is motivated by the fundamental differences between the connectionist and classical architectures, in terms of prerequisites, lower level functionality and inherent constraints. 2 1 Introduction Fodor and Pylyshyn (1988) claimed that connectionist models do not provide sufficient means for the (in their view) necessary ability of cognitive agents to compose mental representations for complex expressions and process these systematically according to the syntactic structure of the representation (which directly ....

....is determined by (a function of) the content of its constituents together with its constituent structure. Thus, there is a distinction between atomic and complex (alternatively molecular) representations. The content of each constituent is maintained through sound truth preserving transformations. Fodor and Pylyshyn (1988) opted for context independent combinatorial semantics which naturally allows systematic processing by virtue of the inherent structure, since syntax parallels semantics. For example, if P has property Q and R is a P, P bears the same content in each assertion and, thus, contributes with the same ....

[Article contains additional citation context not shown here]

Fodor, J. A. and Pylyshyn, Z. W. (1988). Connectionism and cognitive architecture: A critical analysis. Cognition, (28):3--71.

....verbs, a stage of error performance characterized by the inappropriate addition of the ed suffix is seen, followed by a gradual recovery of the correct form. This result sparked much criticism, especially from the camps which declared that language could not be modeled by a connectionist system (Fodor and Pylyshyn, 1988; Pinker and Prince, 1988) These critics claimed that the U shaped learning curve actually displayed by children is not as dramatic as that shown by the Rumelhart McClelland model. At no time do children completely lose the ability to correctly form the irregular past tense forms that they ....

Fodor, J. and Pylyshyn, Z. (1988). Connectionism and cognitive architecture: A critical analysis. Cognition, 28:3--71.

....symbolic paradigm, based on logic, and the connectionist paradigm, based on the propagation of the activity of elementary processors. Artificial neural networks do not have the expressive power of general logical representations, since they are not adequate for manipulation of high level symbols [Fodor and Pylyshyn, 1988]. They are usually preferred in a number of situations (such as pattern recognition) because they are able to generalize over the inputs, they are fault tolerant, and they exhibit the ability to learn from experience. But neural networks have a disadvantage: often because of lack of transparency ....

J. A. Fodor and Z. W. Pylyshyn. Connectionism and Cognitive Architecture: A Critical Analysis. Cognition, Vol. 28, 3-71, 1988.

....of simple neuron like processing elements, such as associative content addressable memories. However, their use in models of analogy processing in humans has been limited by the widespread supposition that it is difficult or impossible to represent compositional structure in vector representations (Fodor and Pylyshyn 1988, Ratcliff and McKoon 1989, Thagard, Holyoak, Nelson and Gochfeld, 1990, Gentner and Markman 1993,Forbus, Gentner and Law, 1994, Wharton, Holyoak Downing, Lange, Wickens, and Melz 1994) This supposition is false. Structure can be represented in vectors in a number of ways, e.g. Smolenskys ....

FODOR J A and PYLYSHYN Z W 1988 Connectionism and cognitive architecture: A critical analysis. Cognition 28, 3---71.

....it is never the case that subjects with the capacity to perform addition do not have the capacity to perform subtraction. Such an observation would rule out the associative, but not the relational architecture. This example, though illustrative, captures the essence of the systematicity argument Fodor and Pylyshyn (1988) raised against connectionism. They argued that while human cognitive capacity is organized around groups of related behaviours (i.e. capability over 3 any one behaviour is systematicity related to any other behaviour within the same group) connectionist models are indifferent to this ....

Fodor, J. A., & Pylyshyn, Z. W. (1988). Connectionism and cognitive architecture: A critical analysis. Cognition, 28, 3--71.

....method. This approach became the subject harsh criticism, based on the disparity between the strength of the claims made and the actual results reported [9] as well as the apparent inability of such systems to handle the systematic, compositional aspects of meaning in recursive symbol systems [5]. The second sort of connectionist approach goes beyond the rules and representations view and directly to the heart of what computing actually means, by showing how a recurrent neural network can perform all the operations of a Turing machine, or more [16] Though such proofs may hold a good ....

....we felt justi ed in using the term In nite RAAM (IRAAM) to refer to our decoder networks. Against a traditional approach to problems like logic, in which recursive symbol systems are held to be the only suciently generative models and neural networks are seen as mere nite capacity implementations [5], the formally proven existence of a set of pure a n b n weights provided evidence that a neural network can serve as an in nitely generative model, under a dynamical systems interpretation of the network s behavior. 6 Uni cation based IRAAM Nevertheless, a fundamental problem exists in ....

J.A. Fodor and Z.W. Pylyshyn. Connectionism and cognitive architecture: A critical analysis. Cognition, 28:3{ 71, 1988.

.... symbols and rules school of cognitive science, based on the disparity between the strength of the claims made and the actual results reported (Pinker and Prince 1988) as well as the apparent inability of such systems to handle the systematic, compositional aspects of linguistic meaning (Fodor and Pylyshyn 1988). The second sort of connectionist approach goes beyond the rules and representations view and directly to the heart of what computing actually means, by showing how a recurrent neural network can perform all the operations of a Turing machine, or more (Siegelmann 1995) Though such proofs may ....

....with addresss a, reachable from the attractor on the left transform, are colored dark gray; points with address b, reachable on the right transform, are light gray. The left transients to the attractor are shown as dashed lines, and the right transients as solid lines. merely implementations (Fodor and Pylyshyn 1988) or performance models, the formally proven existence of a set of pure a n b n weights provided evidence that a neural network can serve as both a competence model and a performance model, under a dynamical systems interpretation of the network s behavior. Unification based IRAAM ....

Fodor, J. and Z. Pylyshyn (1988). Connectionism and cognitive architecture: A critical analysis. Cognition 28, 3--71.

....relationship between the processing mechanisms and structures involved in learning vs. those involved in production. That is, learning is accomomplished by supplying values for the parameters defined by metrical theory; these values then form a knowledge base for stress 2 Thus, Fodor Pylyshyn ( Fodor 88, p. 13] the symbol structures in a Classical model are assumed to correspond to real physical structures in the brain and the combinatorial structure of a representation is supposed to have a counterpart in the structural relations among physical properties of the brain . emphasis ....

Fodor, J., & Pylyshyn, Z., Connectionism and cognitive architecture: a critical analysis, in Steven Pinker & Jacques Mehler (eds.), Connections and Symbols, MIT Press, Cambridge, MA, 1988.

....challenger to representational artificial intelligence connectionism hasn t really been a force of significance in common sense psychology. Indeed, it was from the high ground of common sense psychology that Fodor and Pylyshyn made their systematicity argument attack on naive connectionism (Fodor Pylyshyn, 1988). A full exposition of their argument is beyond this thesis, but it is based on the compositional and generative aspects of propositions mentioned in chapter 2, and the problems of implementing these aspects with connectionist systems. Briefly, they argue that because propositional attitudes can ....

Fodor, J. A., & Pylyshyn, Z. W. (1988). Connectionism and Cognitive Architecture: A Critical Analysis. Cognition, 20, 1988.

....from the School of Computer Science, University of Birmingham, UK. 1 would argue contributes to the pessimism of those who hold the AI in crisis view. It did not take long for shortcomings with the earlier work to be discovered and criticisms of the whole approach to be developed (e.g. [5]) and much of recent connectionist research has focused on dealing with these criticisms. Nevertheless the challenge that connectionism made to the symbolic approach has probably led to a better understanding of what is required for a true AI to be built. For connectionism, and AI generally, to ....

....sentences and facts. The languages we will deal with are all compositional the meaning of a sentence is a function of the meaning of its parts. The above emphasises a key point in knowledge representation, namely that the representations are normally compositional. Fodor Pylyshyn [5] argue that compositionality is essential in a 2 cognitive architecture. In traditional symbolic systems compositionality is achieved by building up the representation of a complex proposition directly out of the representation of its constituents. For example the sentence John loves Mary ....

[Article contains additional citation context not shown here]

J. A. Fodor and Z. W. Pylyshyn. Connectionism and cognitive architecture: a critical analysis. Cognition, 28(1--2):3--71, 1988.

....recursive structures. 1. Introduction Connectionism is conventionally seen as standing in opposition to traditional symbol processing, where symbol processing in this context means representing recursive structures and manipulating them according to their structural composition (Smolensky 1988, Fodor and Pylyshyn 1988, Garfield 1997) Connectionist networks and symbol processing systems are often considered to have complementary strengths and weaknesses (Horgan and Tienson 1996, Hadley 1999) Many attempts have been made to combine the virtues of connectionism and symbol processing in a single architecture ....

Fodor, J.A., and Pylyshyn, Z.W., 1988, Connectionism and cognitive architecture: a critical analysis. Cognition, 28 (1 & 2): 3--71.

....as Euclidean distance in an internal multidimensional space may be too restrictive to be applicable to many aspects of cognition. It is typically assumed that the cognitive representation formed of a visually presented object, a sentence or a story, will involve structured representations (e.g. [3, 15, 16, 23, 28, 41, 51]. Structured representations can description an object not just as a set of features, or as a set of numerical values along various dimensions, but in terms of parts and their interrelations, and properties that attach to those parts. Thus, in describing a bird, it is important to specify not just ....

Fodor, J. A., & Pylyshyn, Z. W. (1988) Connectionism and Cognitive Architecture: A Critical Analysis. Cognition, 28, 3--71.

.... properties of intelligent systems like, for example, being massively parallel, context sensitive, adaptable and robust (see e.g. 7] It is strongly believed that intelligent systems must also be able to represent and reason about structured objects and structure sensitive processes (see e.g. [8, 23]) Unfortunately, we are unaware of any connectionist system which can handle structured objects and structure sensitive processes in a satisfying way. Logic systems were designed to cope with such objects and processes and, consequently, it is a long standing research goal to combine the ....

J. A. Fodor and Z. W. Pylyshyn. Connectionism and cognitive architecture: A critical analysis. In Pinker and Mehler, editors, Connections and Symbols, pages 3-71. MIT Press, 1988.

....learning methods. It is widely believed by psycholinguists that humans receive very little (if any) negative feedback during pre school language acquisition. In passing, we should note that our connectionist inspired model is not intended either as a confirmation or as a counter example to Fodor and Pylyshyn (1988) views on systematicity and the limits of connectionist representations. We freely acknowledge that much of our network is implementing higher levels of functionality. Indeed, this has been a major focus of this article. We also believe, though, that connectionist training methods contribute in ....

Fodor, J. A., & Pylyshyn, Z. W. (1988). Connectionism and cognitive architecture: a critical analysis. Cognition, 28, 3--71.

....vs. Enaction Cognitivism, as exemplified by the aforementioned PSSH, can be said to be dominated by a between the ears , centralized and disembodied focus on the mind (Rutkowska 1996) In particular, cognitivism is based on the traditional notion of representationalism (Fodor 1981; Fodor Pylyshyn 1988), characterized by the assumption of a stable relation between manipulable agent internal representations ( knowledge ) and agent external entities in a pregiven external world (cf. Peschl 1996) Hence, the cognitivist notion of cognition is that of computational, i.e. formally defined and ....

Fodor, J. A. & Pylyshyn, Z. (1988) Connectionism and cognitive architecture: A critical analysis. Cognition, 28:3-71.

....symbolic artificial intelligence school. A number of such basic properties displayed by the human mind are examined, namely generativity of thought, systematicity and compositionality of mental representations, and systematicity of inferential processes, properties that have been described in (Fodor and Pylyshyn, 1988). The question of how the connectionist framework fares with respect to their modeling is addressed, leading to the conclusion that experiments which could convince observers believing that some of these properties cannot in principle be neurally modeled are needed. The connectionist theory of ....

....due to Cussins (1990) is then examined. This general theory makes the point that connectionism is not, in principle, restricted in such a way that these properties could not be modeled. The last section of the chapter, finally, presents a refutation of the critique of connectionism found in (Fodor and Pylyshyn, 1988). 1.2 Goal and overview of this thesis Chapter 1 comes to the conclusion that connectionist systems would fail as proper cognitive models if they could not in principle allow: P1) Compositional representations (P2) Systematic representations (P3) Structure sensitive processing (P4) ....

[Article contains additional citation context not shown here]

J.A. Fodor and Z.W. Pylyshyn. Connectionism and cognitive architecture: A critical analysis. Cognition, 28:3--71, 1988.

.... computationally hard problems (usually NP complete problems) can be found in e.g. HT85, PA88, Pin91b, OPS92] In particular, it would be very desirable to merge expressional power of mathematical logic (it is believed that an intelligent system must be able to process symbolical information [FP88] with massive parallelism of connectionist models. Most of the connectionist models [Der87, DL89, KR89, ALM89, H 90b, H 90a, AS90, GN91] in logic are based on spreading activation model [Die90] which is time efficient and relatively simple to analyze, but very wasteful in terms of space. ....

J. A. Fodor and Z. W. Pylyshyn. Connectionism and cognitive architecture: A critical analysis. Cognition, 28:3--71, 1988.

....of the essence of human computation. 4 It is also quite clear that without understanding these complexity issues, connectionists can stumble again and again into the trap of making strong claims for their models, easy to attack for not offering an adequate replacement for established theory. (Fodor Pylyshyn, 1988; Pinker Prince, 1988) But it is only because of long term lack of competition that descriptive theories involving rules and representations can be defended as explanatory theories. Here is an alternative hypothesis for complex syntactic structure: The state space limit of a dynamical ....

Fodor, J. & Pylyshyn, A. (1988). Connectionism and Cognitive Architecture: A Critical Analysis. Cognition, 28, 3-71.

....that Fodor Pylyshyn were wrong to argue that connectionism cannot, in principle, deliver a non Classical explanation of systematicity. What is Systematicity Insofar as the concept of systematicity differs from productivity, it was introduced into cognitive science for the first time in Fodor Pylyshyn s 1988 paper. In view of this, it is a surprising fact that Fodor Pylyshyn characterize systematicity only very vaguely. For example, Fodor has described systematicity as the idea that cognitive capacities come in clumps (Fodor McLaughlin 1990, p 184) In this paper, we focus on just one component ....

Fodor J. A. & Pylyshyn Z. W., (1988), Connectionism and cognitive architecture: A critical analysis, In Connections and symbols, Pinker Steven & Mehler Jacques (Eds), MIT Press, pp 3 - 71.

.... Canberra ACT 0200, Australia tvg coombs.anu.edu.au Lars Niklasson Department of Computer Science, University of Skvde, S 54128 Sweden lars ida.his.se Abstract This paper challenges the widely accepted claim that classical cognitive architectures can explain the systematicity of cognition (Fodor Pylyshyn, 1988). There are plausible ways of rendering more precise the systematicity hypothesis (as standardly formulated) in which it is entailed by classical architectures, and other plausible ways in which it is not. Therefore, it is not a determinate issue whether systematicity is entailed, and hence ....

....inference capacities come in are not the clumps that are entailed by classical architectures. Therefore, in this area at least, systematicity considerations count against the classical conception of cognitive architecture. In their well known paper Connectionism and Cognitive Architecture, Fodor Pylyshyn (1988) argued that connectionism cannot constitute a viable alternative to the so called classical conception of cognitive architecture, on the grounds that, unlike classical architectures, it cannot provide any explanation of the pervasive empirical phenomenon of systematicity. Therefore, the only ....

Fodor, J. A., & Pylyshyn, Z. (1988) Connectionism and cognitive architecture: a critical analysis.

....a high degree of systematicity. For example, in normal cases the ability to multiply 47 and 63 implies the ability to multiply 47 and 65; anything that can do one can do the other. The assumption that cognitive systems are digital computers provides one way to explain such systematicity (See (Fodor Pylyshyn, 1988) for elaboration of this point. What else could it be Until at least the mid 1980s, there were no general hypotheses as to the nature of cognitive systems that could stand as serious competitors to the Computational Hypothesis in scope, power, and precision. 2.9 The Dynamical Hypothesis ....

.... right, who argue that dynamical approaches may force us to substantially reconceive representation, but not to reject it (e.g. Clark, 1997) and the conservative right, who stick steadfastly to the traditional represenationalist picture, cashed out in standard computational terms (e.g. (Fodor Pylyshyn, 1988)) 4. Conclusion Computation is essential to cognitive science, not merely as a practical aid in investigation, but as part of the theoretical fabric. However, the roles played by digital computers in cognitive science depend very much on the specific kind of cognitive science. In computational ....

Fodor, J. A., & Pylyshyn, Z. (1988) Connectionism and cognitive architecture: a critical analysis. Cognition, 28, 3-71.

....Turing Machine is a dynamical system (van Gelder, 1993) So, it could be concluded that there is nothing to gain from introducing a separate dynamicist method of studying cognition. However, Turing Machines and connectionist networks have also been shown to be computationally equivalent (Fodor and Pylyshyn, 1988, p. 10) yet the approaches are vastly disparate in their methods, strengths, and philosophical commitments. Similarly, though Turing Machines are dynamical in the strictest mathematical sense they are nonetheless serial and discrete. Hence, symbolicist models do not behave in the same ideally 24 ....

....can no longer be divorced from brain structure because our knowledge of the biological band has begun to take on a scientific solidity that breeds confidence about the broad outlines of what is going on in the temporal space that reaches from milliseconds to behavior (Newell, 1990, p. 483, c.f. Fodor, 1988 #2442, p. 62 64) However, Newell seems very selective in the biological data he uses in developing Soar. The division of the total cognitive system into perceptual, motor and central subsystems is not biologically supported. If brain structure does not clearly support such a division, an ....

[Article contains additional citation context not shown here]

Fodor, J. and Z. Pylyshyn (1988). "Connectionism and cognitive architecture: A critical analysis." Cognition 28: 3-71.

....being refuted is being ignored. Thus philosophers often advance positions that are so strongly and provocatively stated that other cognitive scientists feel compelled to respond. This is the philosopher as gadfly. An excellent example is the notorious paper on connectionism by Fodor Pylyshyn (Fodor Pylyshyn, 1988). The authors there stated bluntly, and argued deftly, that a whole new research program was either hopeless or an uninteresting variant on their own classical world view. Cognition, they claimed, is systematic, and connectionist models either fail to explain this, or they must be implementing ....

Fodor, J. A., & Pylyshyn, Z. (1988) Connectionism and cognitive architecture: a critical analysis. Cognition, 28, 3-71.

....used in non imagery contexts, will not do. The reason that there has been so much talk (by me and others) about the representations underlying mental imagery being propositional is that there are very good reasons for thinking that much of cognition depends on a language of thought (Fodor, 1975; Fodor Pylyshyn, 1988; Pylyshyn, 1984) For example, propositions, or more correctly, language like tree structured symbolic encodings, are the only form of representation that we know that can take advantage of mechanical reasoning mechanisms, such as computers, and they are also the only ones we know that exhibit ....

.... of representation that we know that can take advantage of mechanical reasoning mechanisms, such as computers, and they are also the only ones we know that exhibit the properties of compositionality, productivity and systematicity that are essential characteristics of at least human thought (see Fodor Pylyshyn, 1988). Although that does not entail that mental images are propositions, the propositional proposal serves as the natural null hypothesis against which to compare any proposal for a special for of representation for mental imagery. It s not that the idea of images having the form of a set of sentences ....

Fodor, J. A., & Pylyshyn, Z. W. (1988). Connectionism and cognitive architecture: A critical analysis. Cognition, 28, 3-71.

No context found.

Fodor, J. and Z. Pylyshyn. (1988). Connectionism and cognitive architecture: a critical analysis. Cognition 28: 3-71.

No context found.

J. A .Fodor, Z. Pylyshyn, "Connectionism and Cognitive Architecture: A Critical Analysis", Cognition, 28 (1988) 3-71.

No context found.

Fodor, J.A. and Pylyshyn, Z.W. (1988) Connectionism and cognitive architecture: a critical analysis. Cognition 28, 3--71

No context found.

Fodor, J.A. & Z.W. Pylyshyn, 1988. Connectionism and cognitive architecture: a critical analysis, Cognition 28, 3--71.

No context found.

J. A. Fodor and Z. Pylyshyn, \Connectionism and cognitive architecture: A critical analysis," in Connections and Symbols (S. Pinker and J. Mehler, eds.), pp. 3-71, Cambridge, MA.: MIT Press, 1988.

No context found.

J. A. Fodor and Z. W. Pylyshyn. Connectionism and cognitive architecture: A critical analysis. In Pinker and Mehler, editors, Connections and Symbols, pages 3--71. MIT Press, 1988.

No context found.

Fodor, J. A., & Pylyshyn, Z. W. (1988). Connectionism and cognitivearchitecture: A critical analysis. Cognition, 28,3--71.

No context found.

Fodor J. A. & Pylyshyn Z. W., "Connectionism and cognitive architecture: a critical analysis", Connections and symbols, (Eds) Pinker S. & Mehler J., Cambridge, MA: MIT Press, Vol. 5, pp 3-71, 1988.

No context found.

Fodor, J.A. and Pylyshyn, Z.W. Connectionism and cognitive architecture: A critical analysis. Cognition, 28. 3-71.

No context found.

J.A. Fodor and Z.W. Pylyshyn. Connectionism and cognitive architecture: A critical analysis. Cognition, 28:3--71, 1988. 26

No context found.

Jerry A. Fodor and Zenon W. Pylyshyn. Connectionism and cognitive architecture: A critical analysis. In Steven Pinker and Jacques Mehler, editors, Connections and Symbols, pages 3--71. MIT Press, 1988.

No context found.

J. A. Fodor and Z. W. Pylyshyn. Connectionism and cognitive architecture: A critical analysis. Cognition, 28:3--71, 1988.

No context found.

Fodor J. A. & Pylyshyn Z. W., "Connectionism and cognitive architecture: a critical analysis", Connections and symbols, (Eds) Pinker S. & Mehler J., Cambridge, MA: MIT Press, Vol. 5, pp 3-71, 1988.

No context found.

Fodor, J. and Pylyshyn, Z. (1988). Connectionism and cognitive architecture: A critical analysis. Cognition, 28:3-71.

No context found.

J. Fodor and Z. Pylyshyn. Connectionism and cognitive architecture: A critical analysis. Cognition, 28:3--71, 1988.

No context found.

Fodor, J. A. and Pylyshyn, Z. W. (1988) "Connectionism and cognitive architecture: A critical analysis," Cognition, 28, pp. 3-71.

First 50 documents  Next 50

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