Over the past decade, many findings in cognitive about the contents of consciousness: we will not address neuroscience have resulted in the view that selective what might be called the ‘enabling factors’ for conscious- attention, working memory and cognitive control ness (e.g. appropriate neuromodulation from the brain- stem, etc.). involve competition between widely distributed rep-.

Cognitive developmental disorders cannot be properly understood without due attention to the developmental process, and we commend the authors’simulations in this regard. We note the contribution of these simulations to the nascent field of connectionist modeling of developmental disorders and outline a set of criteria for assessing individual models in the hope of furthering future modeling efforts.

This is an overview of recent philosophical discussion about connectionism and the foundations of cognitive science. Connectionist modeling in cognitive science is described. Three broad conceptions of the mind are characterized, and their comparative strengths and weaknesses are discussed: (1) the classical computation conception in cognitive science; (2) a popular foundational interpretation of connectionism that John Tienson and I call “non‐sentential computationalism”; and (3) an alternative interpretation of connectionism we call “dynamical cognition.” Also discussed are two recent philosophical (...) attempts to enlist connectionism in defense of eliminativism about folk psychology. (shrink)

Connectionist models provide a promising alternative to the traditional computational approach that has for several decades dominated cognitive science and artificial intelligence, although the nature of connectionist models and their relation to symbol processing remains controversial. Connectionist models can be characterized by three general computational features: distinct layers of interconnected units, recursive rules for updating the strengths of the connections during learning, and “simple” homogeneous computing elements. Using just these three features one can construct surprisingly elegant and (...) powerful models of memory, perception, motor control, categorization, and reasoning. What makes the connectionist approach unique is not its variety of representational possibilities or its departure from explicit rule-based models, or even its preoccupation with the brain metaphor. Rather, it is that connectionist models can be used to explore systematically the complex interaction between learning and representation, as we try to demonstrate through the analysis of several large networks. (shrink)

The ability to make informative comparisons is central to human cognition. Comparison involves aligning two representations and placing their elements into correspondence. Detecting correspondences is a necessary component of analogical inference, recognition, categorization, schema formation, and similarity judgment. Connectionist Analogy Builder (CAB) determines correspondences through a simple iterative computation that matches elements in one representation with elements playing compatible roles in the other representation while simultaneously enforcing structural constraints. CAB shows promise as a process model of comparison as its (...) performance can be related to human performance (e.g., solution trajectory, error patterns, time‐on‐task). Furthermore, CAB's bounded working memory allows it to account for the inherent capacity limitations of human processing. CAB's strengths are its parsimony, transparency of operations, and ability to generate performance predictions. In this paper, CAB is evaluated against benchmark phenomena from the analogy literature. (shrink)

The contents and structure of semantic memory have been the focus of much recent research, with major advances in the development of distributional models, which use word co‐occurrence information as a window into the semantics of language. In parallel, connectionist modeling has extended our knowledge of the processes engaged in semantic activation. However, these two lines of investigation have rarely been brought together. Here, we describe a processing model based on distributional semantics in which activation spreads throughout a (...) semantic network, as dictated by the patterns of semantic similarity between words. We show that the activation profile of the network, measured at various time points, can successfully account for response times in lexical and semantic decision tasks, as well as for subjective concreteness and imageability ratings. We also show that the dynamics of the network is predictive of performance in relational semantic tasks, such as similarity/relatedness rating. Our results indicate that bringing together distributional semantic networks and spreading of activation provides a good fit to both automatic lexical processing (as indexed by lexical and semantic decisions) as well as more deliberate processing (as indexed by ratings), above and beyond what has been reported for previous models that take into account only similarity resulting from network structure. (shrink)

This paper discusses the relation between cognitive and implementational levels of analysis. Chalmers (1990, 1993) argues that a connectionist implementation of a classical cognitive architecture possesses a compositional semantics, and therefore undercuts Fodor and Pylyshyn's (1988) argument that connectionist networks cannot possess a compositional semantics. I argue that Chalmers argument misconstrues the relation between cognitive and implementational levels of analysis. This paper clarifies the distinction, and shows that while Fodor and Pylyshyn's argument survives Chalmers' critique, it cannot be (...) used to establish the irrelevance of neurophysiological implementation to cognitive modeling; some aspects of Chater and Oaksford's (1990) response to Fodor and Pylyshyn, though not all, are therefore cogent. (shrink)

Connectionism—also known as parallel distributed processing, or neural network modeling—offers promise as a framework to unite clinical and cognitive psychology, and as a tool for studying conscious and unconscious mental activity. This paper describes a neural network model of the case study of Lucy R., from Freud and Breuer's Studies on Hysteria. Though very simple in architecture, the network spontaneously displays analogues of repression and hallucination, corresponding to Lucy R.'s symptoms. Salient elements of Lucy's conscious experience are represented in (...) the model by the activations of neuronlike processors in a fully interconnected network, without hidden units. The model learns to associate elements of experience that were associated in Lucy's case history. Some of these configurations of elements were traumatic for Lucy; trauma is modeled in the network by "emphatic learning," learning accomplished at an abnormally high learning rate. The model suggests that changing associations among conscious elements are sufficient to generate the symptoms Freud observed: apparent repression, hallucination, and recovery through therapy. In the case of Lucy R., Freud's theoretical inference regarding active but unconscious thought is not required by his data. Instead, the unconscious can be understood as a set of complex dispositions embodied in connections between elements of conscious experience. (shrink)

A set of hypotheses is formulated for a connectionist approach to cognitive modeling. These hypotheses are shown to be incompatible with the hypotheses underlying traditional cognitive models. The connectionist models considered are massively parallel numerical computational systems that are a kind of continuous dynamical system. The numerical variables in the system correspond semantically to fine-grained features below the level of the concepts consciously used to describe the task domain. The level of analysis is intermediate between those of (...) symbolic cognitive models and neural models. The explanations of behavior provided are like those traditional in the physical sciences, unlike the explanations provided by symbolic models.Higher-level analyses of these connectionist models reveal subtle relations to symbolic models. Parallel connectionist memory and linguistic processes are hypothesized to give rise to processes that are describable at a higher level as sequential rule application. At the lower level, computation has the character of massively parallel satisfaction of soft numerical constraints; at the higher level, this can lead to competence characterizable by hard rules. Performance will typically deviate from this competence since behavior is achieved not by interpreting hard rules but by satisfying soft constraints. The result is a picture in which traditional and connectionist theoretical constructs collaborate intimately to provide an understanding of cognition. (shrink)

Evidence from numerous studies using the visual world paradigm has revealed both that spoken language can rapidly guide attention in a related visual scene and that scene information can immediately influence comprehension processes. These findings motivated the coordinated interplay account (Knoeferle & Crocker, 2006) of situated comprehension, which claims that utterance‐mediated attention crucially underlies this closely coordinated interaction of language and scene processing. We present a recurrent sigma‐pi neural network that models the rapid use of scene information, exploiting an utterance‐mediated (...) attentional mechanism that directly instantiates the CIA. The model is shown to achieve high levels of performance (both with and without scene contexts), while also exhibiting hallmark behaviors of situated comprehension, such as incremental processing, anticipation of appropriate role fillers, as well as the immediate use, and priority, of depicted event information through the coordinated use of utterance‐mediated attention to the scene. (shrink)

The forms of words as they appear in text and speech are central to theories and models of lexical processing. Nonetheless, current methods for simulating their learning and representation fail to approach the scale and heterogeneity of real wordform lexicons. A connectionist architecture termed thesequence encoderis used to learn nearly 75,000 wordform representations through exposure to strings of stress‐marked phonemes or letters. First, the mechanisms and efficacy of the sequence encoder are demonstrated and shown to overcome problems with traditional (...) slot‐based codes. Then, two large‐scale simulations are reported that learned to represent lexicons of either phonological or orthographic wordforms. In doing so, the models learned the statistics of their lexicons as shown by better processing of well‐formed pseudowords as opposed to ill‐formed (scrambled) pseudowords, and by accounting for variance in well‐formedness ratings. It is discussed how the sequence encoder may be integrated into broader models of lexical processing. (shrink)

I consider the role of cognitive modeling in cognitive science. Modeling, and the computers that enable it, are central to the field, but the role of modeling is often misunderstood. Models are not intended to capture fully the processes they attempt to elucidate. Rather, they are explorations of ideas about the nature of cognitive processes. In these explorations, simplification is essential—through simplification, the implications of the central ideas become more transparent. This is not to say that simplification (...) has no downsides; it does, and these are discussed. I then consider several contemporary frameworks for cognitive modeling, stressing the idea that each framework is useful in its own particular ways. Increases in computer power (by a factor of about 4 million) since 1958 have enabled new modeling paradigms to emerge, but these also depend on new ways of thinking. Will new paradigms emerge again with the next 1,000‐fold increase? (shrink)

The forms of words as they appear in text and speech are central to theories and models of lexical processing. Nonetheless, current methods for simulating their learning and representation fail to approach the scale and heterogeneity of real wordform lexicons. A connectionist architecture termed thesequence encoderis used to learn nearly 75,000 wordform representations through exposure to strings of stress‐marked phonemes or letters. First, the mechanisms and efficacy of the sequence encoder are demonstrated and shown to overcome problems with traditional (...) slot‐based codes. Then, two large‐scale simulations are reported that learned to represent lexicons of either phonological or orthographic wordforms. In doing so, the models learned the statistics of their lexicons as shown by better processing of well‐formed pseudowords as opposed to ill‐formed (scrambled) pseudowords, and by accounting for variance in well‐formedness ratings. It is discussed how the sequence encoder may be integrated into broader models of lexical processing. (shrink)

In this commentary I explore nonclassical connectionism as a coherent framework for evaluation in the spirit of the Newell Test. Focusing on knowledge integration, development, real-time performance, and flexible behavior, I argue that NCC's “within-theory rank ordering” would place subsymbolic modeling in a better position. Failure to adopt a symbolic level of thought cannot be interpreted as a weakness.

Glenberg's account falls short in several respects. Besides requiring clearer explication of basic concepts, his account fails to recognize the autonomous nature of perception. His account of what is remembered, and its description, is too static. His strictures against connectionist modeling might be overcome by combining the notions of psychological space and principled learning in an embodied and situated network.

Are similar, or even identical, mechanisms used in the computational modeling of speech segmentation, serial image processing, and music processing? We address this question by exploring how TRACX2, a recognition‐based, recursive connectionist autoencoder model of chunking and sequence segmentation, which has successfully simulated speech and serial‐image processing, might be applied to elementary melody perception. The model, a three‐layer autoencoder that recognizes “chunks” of short sequences of intervals that have been frequently encountered on input, is trained on the tone (...) intervals of melodically simple French children's songs. It dynamically incorporates the internal representations of these chunks into new input. Its internal representations cluster in a manner that is consistent with “human‐recognizable” melodic categories. TRACX2 is sensitive to both contour and proximity information in the musical chunks that it encounters in its input. It shows the “end‐of‐word” superiority effect demonstrated by Saffran et al. (1999) for short musical phrases. The overall findings suggest that the recursive autoassociative chunking mechanism, as implemented in TRACX2, may be a general segmentation and chunking mechanism, underlying not only word‐ and image‐chunking, but also elementary melody processing. (shrink)

The debate between Classicism and Connectionism can be properly characterized as a debate concerning the appropriate levels of analysis for psychological theorizing. Classicists maintain that the level of analysis defined by the Classical architecture is the level of analysis at which psychological theorizing should reside. This level is called the symbolic level. On the other hand, Connectionists claim that the proper level of analysis for cognitive modeling is at the subsymbolic level which is considered a level lower than the (...) symbolic level. This thesis argues that both Classicism and Connectionism characterized in the above way are single-level approaches. It is argued in this thesis that Classicism is wrong in denying the cognitive relevancy of the lower-level theorizing; and some forms of Connectionism should be rejected because they underestimate the theoretical significance of the higher-level modeling. ;The concept of level is widely employed by philosophers and cognitive scientists, but has not received sufficient analysis. This thesis offers a general theory of levels, and evaluates Classicism and Connectionism on the basis of this general theory of levels. Basically this thesis argues that a cognitive system should be regarded as a system organized into a hierarchy of levels of organization. Different levels of organization may demand different levels of analysis. This thesis argues that each level of analysis enjoys a degree of autonomy, i.e., it is not totally reducible to the lower-level analysis. The traditional dichotomy of the cognitive and the implementational levels is too simplistic and overlooks the complexity of a cognitive system. Not all levels of organization of a system are cognitively relevant. But much empirical evidence has shown that many levels of organization are either strongly relevant or at least weakly relevant to cognitive modeling. It would be theoretically more plausible and fruitful to adopt a multi-level approach which takes many levels of analysis seriously in cognitive modeling. A very general picture of a multi-level approach to the mind is depicted in this thesis. (shrink)

A set of hypotheses is formulated for a connectionist approach to cognitive modeling. These hypotheses are shown to be incompatible with the hypotheses underlying traditional cognitive models. The connectionist models considered are massively parallel numerical computational systems that are a kind of continuous dynamical system. The numerical variables in the system correspond semantically to fine-grained features below the level of the concepts consciously used to describe the task domain. The level of analysis is intermediate between those of (...) symbolic cognitive models and neural models. The explanations of behavior provided are like those traditional in the physical sciences, unlike the explanations provided by symbolic models. (shrink)

This paper reviews four significant advances on the feedforward architecture that has dominated discussions of connectionism. The first involves introducing modularity into networks by employing procedures whereby different networks learn to perform different components of a task, and a Gating Network determines which network is best equiped to respond to a given input. The second consists in the use of recurrent inputs whereby information from a previous cycle of processing is made available on later cycles. The third development involves developing (...) compressed representations of strings in which there is no longer an explicit encoding of the components but where information about the structure of the original string can be recovered and so is present functionally. The final advance entails using connectionist learning procedures not just to change weights in networks but to change the patterns used as inputs to the network. These advances significantly increase the usefulness of connectionist networks for modeling human cognitive performance by, among other things, providing tools for explaining the productivity and systematicity of some mental activities, and developing representations that are sensitive to the content they are to represent. (shrink)

Connectionism is a style of modeling based upon networks of interconnected simple processing devices. This style of modeling goes by a number of other names too. Connectionist models are also sometimes referred to as 'Parallel Distributed Processing' (or PDP for short) models or networks.1 Connectionist systems are also sometimes referred to as 'neural networks' (abbreviated to NNs) or 'artificial neural networks' (abbreviated to ANNs). Although there may be some rhetorical appeal to this neural nomenclature, it is (...) in fact misleading as connectionist networks are commonly significantly dissimilar to neurological systems. For this reason, I will avoid using this terminology, other than in direct quotations. Instead, I will follow the practice I have adopted above and use 'connectionist' as my primary term for systems of this kind. (shrink)

In the late 1980s, there were many who heralded the emergence of connectionism as a new paradigm – one which would eventually displace the classically symbolic methods then dominant in AI and Cognitive Science. At present, there remain influential connectionists who continue to defend connectionism as a more realistic paradigm for modeling cognition, at all levels of abstraction, than the classical methods of AI. Not infrequently, one encounters arguments along these lines: given what we know about neurophysiology, it is (...) just not plausible to suppose that our brains are digital computers. Thus, they could not support a classical architecture. I argue here for a middle ground between connectionism and classicism. I assume, for argument's sake, that some form(s) of connectionism can provide reasonably approximate models – at least for lower-level cognitive processes. Given this assumption, I argue on theoretical and empirical grounds that most human mental skills must reside in separate connectionist modules or sub-networks. Ultimately, it is argued that the basic tenets of connectionism, in conjunction with the fact that humans often employ novel combinations of skill modules in rule following and problem solving, lead to the plausible conclusion that, in certain domains, high level cognition requires some form of classical architecture. During the course of argument, it emerges that only an architecture with classical structure could support the novel patterns of information flow and interaction that would exist among the relevant set of modules. Such a classical architecture might very well reside in the abstract levels of a hybrid system whose lower-level modules are purely connectionist. (shrink)

Taking to heart Massaro's [(1988) Some criticisms of connectionist models of human performance, Journal of Memory and Language, 27, 213-234] criticism that multi-layer perceptrons are not appropriate for modeling human cognition because they are too powerful (i.e. they can simulate just about anything, which gives them little explanatory power), Regier develops the notion of constrained connectionism. The model that he discusses is a distributed network but with numerous constraints added that are (more or less) motivated by real psychophysical (...) and neurophysical constraints. His model learns static prepositions of spatial location such as in, above, to the left of, to the right of, under, etc., as well as dynamic prepositions such as through and the Russian iz-pod, meaning out from under. The network learns these prepositions by viewing a number of examples of them. Very importantly, this book tackles-and goes a long way towards resolving-the problem of the lack of negative exemplars (i.e. we are only very rarely told when something is not above something else), which should lead to overgeneralization, but does not. This book is a significant contribution to connectionist literature. (shrink)

Perruchet & Vinter do not fully resolve issues about the role of consciousness and the unconscious in cognition and learning, and it is doubtful that consciousness has been computationally implemented. The cascade-correlation (CC) connectionist model develops high-order feature detectors as it learns a problem. We describe an extension, knowledge-based cascade-correlation (KBCC), that uses knowledge to learn in a hierarchical fashion.

Intermediate constructs are required as bridges between complex behaviors and realistic models of neural circuitry. For cognitive scientists in general, schemas are the appropriate functional units; brain theorists can work with neural layers as units intermediate between structures subserving schemas and small neural circuits.After an account of different levels of analysis, we describe visuomotor coordination in terms of perceptual schemas and motor schemas. The interest of schemas to cognitive science in general is illustrated with the example of perceptual schemas in (...) high-level vision and motor schemas in the control of dextrous hands.Rana computatrix, the computational frog, is introduced to show how one constructs an evolving set of model families to mediate flexible cooperation between theory and experiment. Rana computatrix may be able to do for the study of the organizational principles of neural circuitry what Aplysia has done for the study of subcellular mechanisms of learning. Approach, avoidance, and detour behavior in frogs and toads are analyzed in terms of interacting schemas. Facilitation and prey recognition are implemented as tectal-pretectal interactions, with the tectum modeled by an array of tectal columns. We show how layered neural computation enters into models of stereopsis and how depth schemas may involve the interaction of accommodation and binocular cues in anurans. (shrink)

Although Paul Churchland and Jerry Fodor both subscribe to the so-called theory-theory– the theory that folk psychology (FP) is an empirical theory of behavior – they disagree strongly about FP’s fate. Churchland contends that FP is a fundamentally flawed view analogous to folk biology, and he argues that recent advances in computational neuroscience and connectionist AI point toward development of a scientifically respectable replacement theory that will give rise to a new common-sense psychology. Fodor, however, wagers that FP will (...) be largely preserved and vindicated by scientific investigations of behavior. Recent findings by developmental psychologists, I argue, will push both Churchlandians and Fodorians toward the pessimistic view that FP is a misguided theory that will never be displaced, because it is, so to speak, built into our cognitive systems. I explore the possibility of preserving optimism by rejecting the theory-theory and adopting the simulation theory, a competing view developed by Robert Gordon, Alvin Goldman, and Jane Heal. According to simulationists, common-sense interpretation of behavior is accomplished by means of pretense-like operations that deploy the cognitive system’s own reasoning capabilities in a disengaged manner. Since on this view no theory-like set of principles would be needed, the simulation theory seems to enjoy a simplicity advantage over the theory-theory. Steven Stich and Shawn Nichols, however, contend that as the cognitive system would require special mechanisms for disengaged operation, the simplicity question cannot be resolved until suitable computational models are developed. I describe a set of models I have constructed to meet this need, and I discuss the contribution such models can make to determining FP’s fate. (shrink)

It is often assumed that similar domain-specific behavioural impairments found in cases of adult brain damage and developmental disorders correspond to similar underlying causes, and can serve as convergent evidence for the modular structure of the normal adult cognitive system. We argue that this correspondence is contingent on an unsupported assumption that atypical development can produce selective deficits while the rest of the system develops normally (Residual Normality), and that this assumption tends to bias data collection in the field. Based (...) on a review of connectionist models of acquired and developmental disorders in the domains of reading and past tense, as well as on new simulations, we explore the computational viability of Residual Normality and the potential role of development in producing behavioural deficits. Simulations demonstrate that damage to a developmental model can produce very different effects depending on whether it occurs prior to or following the training process. Because developmental disorders typically involve damage prior to learning, we conclude that the developmental process is a key component of the explanation of endstate impairments in such disorders. Further simulations demonstrate that in simple connectionist learning systems, the assumption of Residual Normality is undermined by processes of compensation or alteration elsewhere in the system. We outline the precise computational conditions required for Residual Normality to hold in development, and suggest that in many cases it is an unlikely hypothesis. We conclude that in developmental disorders, inferences from behavioural deficits to underlying structure crucially depend on developmental conditions, and that the process of ontogenetic development cannot be ignored in constructing models of developmental disorders. Key Words: Acquired and developmental disorders; connectionist models; modularity; past tense; reading. (shrink)

The aim of this paper is to grasp the relevant distinctions between various ways in which models and simulations in Artificial Intelligence (AI) relate to cognitive phenomena. In order to get a systematic picture, a taxonomy is developed that is based on the coordinates of formal versus material analogies and theory-guided versus pre-theoretic models in science. These distinctions have parallels in the computational versus mimetic aspects and in analytic versus exploratory types of computer simulation. The proposed taxonomy cuts across the (...) traditional dichotomies between symbolic and embodied AI, general intelligence and symbol and intelligence and cognitive simulation and human/non-human-like AI. -/- According to the taxonomy proposed here, one can distinguish between four distinct general approaches that figured prominently in early and classical AI, and that have partly developed into distinct research programs: first, phenomenal simulations (e.g., Turing’s “imitation game”); second, simulations that explore general-level formal isomorphisms in pursuit of a general theory of intelligence (e.g., logic-based AI); third, simulations as exploratory material models that serve to develop theoretical accounts of cognitive processes (e.g., Marr’s stages of visual processing and classical connectionism); and fourth, simulations as strictly formal models of a theory of computation that postulates cognitive processes to be isomorphic with computational processes (strong symbolic AI). -/- In continuation of pragmatic views of the modes of modeling and simulating world affairs, this taxonomy of approaches to modeling in AI helps to elucidate how available computational concepts and simulational resources contribute to the modes of representation and theory development in AI research—and what made that research program uniquely dependent on them. (shrink)

Connectionism and computationalism are currently vying for hegemony in cognitive modeling. At first glance the opposition seems incoherent, because connectionism is itself computational, but the form of computationalism that has been the prime candidate for encoding the "language of thought" has been symbolic computationalism (Dietrich 1990, Fodor 1975, Harnad 1990c; Newell 1980; Pylyshyn 1984), whereas connectionism is nonsymbolic (Fodor & Pylyshyn 1988, or, as some have hopefully dubbed it, "subsymbolic" Smolensky 1988). This paper will examine what is and is (...) not a symbol system. A hybrid nonsymbolic/symbolic system will be sketched in which the meanings of the symbols are grounded bottom-up in the system's capacity to discriminate and identify the objects they refer to. Neural nets are one possible mechanism for learning the invariants in the analog sensory projection on which successful categorization is based. "Categorical perception" (Harnad 1987a), in which similarity space is "warped" in the service of categorization, turns out to be exhibited by both people and nets, and may mediate the constraints exerted by the analog world of objects on the formal world of symbols. (shrink)

At the end of Personal Knowledge, Polanyi discusses human development, arguing for a view of the human person as emerging out of but not constituted by its material substrate. As part of this view, he argues that the human person can never be likened to a computer, an inference machine, or a neural model because all are based in formalized processes of automation, processes that cannot account for the contribution of unformalizable, tacit knowing. This paper revisits Polanyi’s discussion of the (...) emergence of consciousness and his rejection of neural models in light of recent developments in connectionism. Connectionist neural modeling proposes an emergentist account of brain structure and, in many ways, is compatible with Polanyi’s philosophy, even if it ultimately neglects questions of meaning. (shrink)

What would Glenberg 's attractive ideas look like when computationally fleshed out? I suggest that the most helpful next step in formalizing them is neither a connectionist nor a symbolic implementation, but rather an implementation- general analysis of the task in terms of the informational content required.

Green's target article is an attack on most current connectionist models of cognition. Our commentary will suggest that there is an essential component missing in his discussion of modeling, namely, the idea that the appropriate level of the model needs to be specified. We will further suggest that the precise form of connectionist networks will fall out as ever more detailed constraints are placed on their function.

Cognitive function certainly poses the biggest challenge for computational neuroscience. As we argue, past efforts to build neural models of cognition (the target article included) had too narrow a focus on implementing rule-based language processing. The problem with these models is that they sacrifice the advantages of connectionism rather than building on them. Recent and more promising approaches for modeling cognition build on the mathematical properties of distributed neural representations. These approaches truly exploit the key advantages of connectionism, that (...) is, the high representational power of distributed neural codes and similarity-based pattern recognition. The architectures for cognitive computing that emerge from these approaches are neural associative memories endowed with additional mapping operations to handle invariances and to form reduced representations of combinatorial structures. (shrink)

The model of phoneme monitoring proposed by Norris et al. is implausible when implemented in a localist connectionist network. Lexical representations mysteriously inform phoneme decision nodes as to the presence or absence of a target phoneme.

David Marr's (1982) three‐level analysis of computational cognition argues for three distinct levels of cognitive information processing—namely, the computational, representational, and implementational levels. But Marr's levels are—and were meant to be—descriptive, rather than interactive and dynamic. For this reason, we suggest that, had Marr been writing today, he might well have gone even farther in his analysis, including the emergence of structure—in particular, explicit structure at the conceptual level—from lower levels, and the effect of explicit emergent structures on the level (...) (or levels) that gave rise to them. The message is that today's cognitive scientists need not only to understand how emergent structures—in particular, explicit emergent structures at the cognitive level—develop but also to understand how they feed back on the sub‐structures from which they emerged. (shrink)

There is a broad consensus about the fundamental role of thehippocampal system (hippocampus and its adjacent areas) in theencoding and retrieval of episodic memories. This paper presents afunctional model of this system. Although memory is not asingle-unit cognitive function, we took the view that the wholesystem of the smooth, interrelated memory processes may have acommon basis. That is why we follow the Ockham's razor principleand minimize the size or complexity of our model assumption set.The fundamental assumption is the requirement of (...) solving the socalled ``homunculus fallacy'', which addresses the issue ofinterpreting the input. Generative autoassociators seem to offer aresolution of the paradox. Learning to represent and to recallinformation, in these generative networks, imply maximization ofinformation transfer, sparse representation and noveltyrecognition. A connectionist architecture, which integrates theseaspects as model constraints, is derived. Numerical studiesdemonstrate the novelty recognition and noise filtering propertiesof the architecture. Finally, we conclude that the derivedconnectionist architecture can be related to the neurobiologicalsubstrate. (shrink)

An influential view of the nature of the language system is that of an evolved biological system in which a set of rules is combined with a lexicon that contains the words of the language together with a representation of their context. Alternative views, usually based on connectionist modeling, attempt to explain the structure of language on the basis of complex associative processes. Here, I put forward a third view that stresses experience-dependent structural development of the brain circuits (...) supporting language as a core principle of the organization of the language system. In this view, embodied in a recent neuroconstructivist neural network of past tense development and processing, initial domain-general predispositions enable the development of functionally specialized brain structures through interactions between experience-dependent brain development and statistical learning in a structured environment. Together, these processes shape a biological adult language system that appears to separate into distinct mechanism for processing rules and exceptions, whereas in reality those subsystems co-develop and interact closely. This view puts experience-dependent brain development in response to a specific language environment at the heart of understanding not only language development but adult language processing as well. (shrink)

Learning the structure of event sequences is a ubiquitous problem in cognition and particularly in language. One possible solution is to learn a probabilistic generative model of sequences that allows making predictions about upcoming events. Though appealing from a neurobiological standpoint, this approach is typically not pursued in connectionist modeling. Here, we investigated a sequential version of the restricted Boltzmann machine, a stochastic recurrent neural network that extracts high-order structure from sensory data through unsupervised generative learning and can (...) encode contextual information in the form of internal, distributed representations. We assessed whether this type of network can extract the orthographic structure of English monosyllables by learning a generative model of the letter sequences forming a word training corpus. We show that the network learned an accurate probabilistic model of English graphotactics, which can be used to make predictions about the letter following a given context as well as to autonomously generate high-quality pseudowords. The model was compared to an extended version of simple recurrent networks, augmented with a stochastic process that allows autonomous generation of sequences, and to non-connectionist probabilistic models. We conclude that sequential RBMs and stochastic simple recurrent networks are promising candidates for modeling cognition in the temporal domain. (shrink)

Consciousness remains a mystery—“a phenomenon that people do not know how to think about—yet” (Dennett, , p. 21). Here, I consider how the connectionist perspective on information processing may help us progress toward the goal of understanding the computational principles through which conscious and unconscious processing differ. I begin by delineating the conceptual challenges associated with classical approaches to cognition insofar as understanding unconscious information processing is concerned, and to highlight several contrasting computational principles that are constitutive of the (...) connectionist approach. This leads me to suggest that conscious and unconscious processing are fundamentally connected, that is, rooted in the very same computational principles. I further develop a perspective according to which the brain continuously and unconsciously learns to redescribe its own activity itself based on constant interaction with itself, with the world, and with other minds. The outcome of such interactions is the emergence of internal models that are metacognitive in nature and that function so as to make it possible for an agent to develop a (limited, implicit, practical) understanding of itself. In this light, plasticity and learning are constitutive of what makes us conscious, for it is in virtue of our own experiences with ourselves and with other people that our mental life acquires its subjective character. The connectionist framework continues to be uniquely positioned in the Cognitive Sciences to address the challenge of identifying what one could call the “computational correlates of consciousness” (Mathis & Mozer, ) because it makes it possible to focus on the mechanisms through which information processing takes place. (shrink)

McRae, Brown and Elman argue against the view that events are structured as frequently‐occurring sequences of world stimuli. They underline the importance of temporal structure defining event types and advance a more complex temporal structure, which allows for some variance in the component elements.

The notion of levels has been widely used in discussions of cognitive science, especially in discussions of the relation of connectionism to symbolic modeling of cognition. I argue that many of the notions of levels employed are problematic for this purpose, and develop an alternative notion grounded in the framework of mechanistic explanation. By considering the source of the analogies underlying both symbolic modeling and connectionist modeling, I argue that neither is likely to provide an adequate (...) analysis of processes at the level at which cognitive theories attempt to function: One is drawn from too low a level, the other from too high a level. If there is a distinctly cognitive level, then we still need to determine what are the basic organizational principles at that level. (shrink)