Four experiments examined the strategies that individuals develop in sentential reasoning. They led to the discovery of five different strategies. According to the theory proposed in the paper, each of the strategies depends on component tactics, which all normal adults possess, and which are based on mental models. Reasoners vary their use of tactics in ways that have no deterministic account. This variation leads different individuals to assemble different strategies, which include the construction of incremental diagrams corresponding to mental models, (...) and the pursuit of the consequences of a single model step by step. Moreover, the difficulty of a problem (i.e., the number of mental models required by the premises) predisposes reasoners towards certain strategies. Likewise, the sentential connectives in the premises also bias reasoners towards certain strategies, e.g., conditional premises tend to elicit reasoning step by step whereas disjunctive premises tend to elicit incremental diagrams. (shrink)

This paper investigates connectionism's potential to solve the frame problem. The frame problem arises in the context of modelling the human ability to see the relevant consequences of events in a situation. It has been claimed to be unsolvable for classical cognitive science, but easily manageable for connectionism. We will focus on a representational approach to the frame problem which advocates the use of intrinsic representations. We argue that although connectionism's distributed representations may look promising from this perspective, doubts can (...) be raised about the potential of distributed representations to allow large amounts of complexly structured information to be adequately encoded and processed. It is questionable whether connectionist models that are claimed to effectively represent structured information can be scaled up to a realistic extent. We conclude that the frame problem provides a difficulty to connectionism that is no less serious than the obstacle it constitutes for classical cognitive science. (shrink)

Working memory limits are best defined in terms of the complexity of the relations that can be processed in parallel. Complexity is defined as the number of related dimensions or sources of variation. A unary relation has one argument and one source of variation; its argument can be instantiated in only one way at a time. A binary relation has two arguments, two sources of variation, and two instantiations, and so on. Dimensionality is related to the number of chunks, because (...) both attributes on dimensions and chunks are independent units of information of arbitrary size. Studies of working memory limits suggest that there is a soft limit corresponding to the parallel processing of one quaternary relation. More complex concepts are processed by or In segmentation, tasks are broken into components that do not exceed processing capacity and can be processed serially. In conceptual chunking, representations are to reduce their dimensionality and hence their processing load, but at the cost of making some relational information inaccessible. Neural net models of relational representations show that relations with more arguments have a higher computational cost that coincides with experimental findings on higher processing loads in humans. Relational complexity is related to processing load in reasoning and sentence comprehension and can distinguish between the capacities of higher species. The complexity of relations processed by children increases with age. Implications for neural net models and theories of cognition and cognitive development are discussed. (shrink)

Fodor's and Pylyshyn's stand on systematicity in thought and language has been debated and criticized. Van Gelder and Niklasson, among others, have argued that Fodor and Pylyshyn offer no precise definition of systematicity. However, our concern here is with a learning based formulation of that concept. In particular, Hadley has proposed that a network exhibits strong semantic systematicity when, as a result of training, it can assign appropriate meaning representations to novel sentences (both simple and embedded) which contain words in (...) syntactic positions they did not occupy during training. The experience of researchers indicates that strong systematicity in any form is difficult to achieve in connectionist systems.Herein we describe a network which displays strong semantic systematicity in response to Hebbian, connectionist training. During training, two-thirds of all nouns are presented only in a single syntactic position (either as grammatical subject or object). Yet, during testing, the network correctly interprets thousands of sentences containing those nouns in novel positions. In addition, the network generalizes to novel levels of embedding. Successful training requires a, corpus of about 1000 sentences, and network training is quite rapid. The architecture and learning algorithms are purely connectionist, but classical insights are discernible in one respect, viz, that complex semantic representations spatially contain their semantic constituents. However, in other important respects, the architecture is distinctly non-classical. (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)

The target article demonstrates how neurocognitive modellers should not be intimidated by challenges such as Jackendoff's and should explore neurally plausible implementations of linguistic constructs. The next step is to take seriously insights offlered by neuroscience, including the robustness allowed by analogue computation with distributed representations and the power of attractor dynamics in turning analogue into nearly discrete operations.

A model is presented that addresses both the distribution and comprehension of different forms of referring expressions in language. This model is expressed in a formalism (Kamp & Reyle, 1993) that uses interpretive rules to map syntactic representations onto representations of discourse. Basic interpretive rules are developed for names, pronouns, definite descriptions, and quantified descriptions. These rules are triggered by syntactic input and interact dynamically with representations of discourse to establish reference and coreference. This interaction determines the ease with which (...) coreference can be established for different linguistic forms given the existing discourse context. The performance of the model approximates that observed in studies of intuitive judgments of grammaticality and studies using online measures of language comprehension. The model uses the same basic interpretive mechanisms for coreference within and between sentences, thereby linking the domain traditionally studied by generative linguists to domains that have been of concern primarily to psychologists and computational linguists. (shrink)

The binding problem is to explain how information processed by different sensory systems is brought together to unify perception. The problem has two sides. First, we want to explain phenomenal binding: the fact that we experience a single world rather than separate perceptual fields for each sensory modality. Second, we must solve a functional problem: to explain how a neural net like the brain links instances to types. I argue that phenomenal binding and functional binding require very different treatments. The (...) puzzle of phenomenal binding rests on a confusion and so can be dissolved. So only functional binding deserves explanation. The general solution to that problem is that information to be bound is arrayed along different dimensions. So sensory coding into separate topographic maps facilitates functional binding and there is no need based on the unity of perception for special mechanisms that bring "back together" information in different maps. (shrink)

Abduction is or subsumes a process of inference. It entertains possible hypotheses and it chooses hypotheses for further scrutiny. There is a large literature on various aspects of non-symbolic, subconscious abduction. There is also a very active research community working on the symbolic (logical) characterisation of abduction, which typically treats it as a form of hypothetico-deductive reasoning. In this paper we start to bridge the gap between the symbolic and sub-symbolic approaches to abduction. We are interested in benefiting from developments (...) made by each community. In particular, we are interested in the ability of non-symbolic systems (neural networks) to learn from experience using efficient algorithms and to perform massively parallel computations of alternative abductive explanations. At the same time, we would like to benefit from the rigour and semantic clarity of symbolic logic. We present two approaches to dealing with abduction in neural networks. One of them uses Connectionist Modal Logic and a translation of Horn clauses into modal clauses to come up with a neural network ensemble that computes abductive explanations in a top-down fashion. The other combines neural-symbolic systems and abductive logic programming and proposes a neural architecture which performs a more systematic, bottom-up computation of alternative abductive explanations. Both approaches employ standard neural network architectures which are already known to be highly effective in practical learning applications. Differently from previous work in the area, our aim is to promote the integration of reasoning and learning in a way that the neural network provides the machinery for cognitive computation, inductive learning and hypothetical reasoning, while logic provides the rigour and explanation capability to the systems, facilitating the interaction with the outside world. Although it is left as future work to determine whether the structure of one of the proposed approaches is more amenable to learning than the other, we hope to have contributed to the development of the area by approaching it from the perspective of symbolic and sub-symbolic integration. (shrink)

Human cognition is unique in the way in which it relies on combinatorial (or compositional) structures. Language provides ample evidence for the existence of combinatorial structures, but they can also be found in visual cognition. To understand the neural basis of human cognition, it is therefore essential to understand how combinatorial structures can be instantiated in neural terms. In his recent book on the foundations of language, Jackendoff described four fundamental problems for a neural instantiation of combinatorial structures: the massiveness (...) of the binding problem, the problem of 2, the problem of variables, and the transformation of combinatorial structures from working memory to long-term memory. This paper aims to show that these problems can be solved by means of neural “blackboard” architectures. For this purpose, a neural blackboard architecture for sentence structure is presented. In this architecture, neural structures that encode for words are temporarily bound in a manner that preserves the structure of the sentence. It is shown that the architecture solves the four problems presented by Jackendoff. The ability of the architecture to instantiate sentence structures is illustrated with examples of sentence complexity observed in human language performance. Similarities exist between the architecture for sentence structure and blackboard architectures for combinatorial structures in visual cognition, derived from the structure of the visual cortex. These architectures are briefly discussed, together with an example of a combinatorial structure in which the blackboard architectures for language and vision are combined. In this way, the architecture for language is grounded in perception. Perspectives and potential developments of the architectures are discussed. Key Words: binding; blackboard architectures; combinatorial structure; compositionality; language; dynamic system; neurocognition; sentence complexity; sentence structure; working memory; variables; vision. (shrink)

How do human infants learn the causal dependencies between events? Evidence suggests that this remarkable feat can be achieved by observation of only a handful of examples. Many computational models have been produced to explain how infants perform causal inference without explicit teaching about statistics or the scientific method. Here, we propose a spiking neuronal network implementation that can be entrained to form a dynamical model of the temporal and causal relationships between events that it observes. The network uses spike-time (...) dependent plasticity, long-term depression, and heterosynaptic competition rules to implement Rescorla–Wagner-like learning. Transmission delays between neurons allow the network to learn a forward model of the temporal relationships between events. Within this framework, biologically realistic synaptic plasticity rules account for well-known behavioral data regarding cognitive causal assumptions such as backwards blocking and screening-off. These models can then be run as emulators for state inference. Furthermore, this mechanism is capable of copying synaptic connectivity patterns between neuronal networks by observing the spontaneous spike activity from the neuronal circuit that is to be copied, and it thereby provides a powerful method for transmission of circuit functionality between brain regions. (shrink)

Theories of binding have recently come into the focus of the consciousness debate. In this review, we discuss the potential relevance of temporal binding mechanisms for sensory awareness. Specifically, we suggest that neural synchrony with a precision in the millisecond range may be crucial for conscious processing, and may be involved in arousal, perceptual integration, attentional selection and working memory. Recent evidence from both animal and human studies demonstrates that specific changes in neuronal synchrony occur during all of these processes (...) and that they are distinguished by the emergence of fast oscillations with frequencies in the gamma-range. (shrink)

In this paper we present Drama, a distributed model of analogical mapping that integrates semantic and structural constraints on constructing analogies. Specifically, Drama uses holographic reduced representations (Plate, 1994), a distributed representation scheme, to model the effects of structure and meaning on human performance of analogical mapping. Drama is compared to three symbolic models of analogy (SME, Copycat, and ACME) and one partially distributed model (LISA). We describe Drama's performance on a number of example analogies and assess the model in (...) terms of neurological and psychological plausibility. We argue that Drama's successes are due largely to integrating structural and semantic constraints throughout the mapping process. We also claim that Drama is an existence proof of using distributed representations to model high‐level cognitive phenomena. (shrink)

By reviewing most of the neurobiology of consciousness, this article highlights some major reasons why a successful emulation of the dynamics of human consciousness by artificial intelligence is unlikely. The analysis provided leads to conclude that human consciousness is epigenetically determined and experience and context-dependent at the individual level. It is subject to changes in time that are essentially unpredictable. If cracking the code to human consciousness were possible, the result would most likely have to consist of a temporal pattern (...) code simulating long-distance signal reverberation and de-correlation of all spatial signal contents from temporal signals. In the light of the massive evidence for complex interactions between implicit (non-conscious) and explicit (conscious) contents of representation, the code would have to be capable of making implicit (non-conscious) processes explicit. It would have to be capable of a progressively less and less arbitrary selection of temporal activity patterns in a continuously developing neural network structure identical to that of the human brain, from the synaptic level to that of higher cognitive functions. The code’s activation thresholds would depend on specific temporal signal coincidence probabilities, vary considerably with time and across individual experience data, and would therefore require dynamically adaptive computations capable of emulating the properties of individual human experience. No known machine or neural network learning approach has such potential. (shrink)

Leech et al.'s model proposes representing relations as primed transformations rather than as structured representations (explicit representations of relations and their roles dynamically bound to fillers). However, this renders the model unable to explain several developmental trends (including relational integration and all changes not attributable to growth in relational knowledge). We suggest looking to an alternative computational model that learns structured representations from examples.

Several approaches to implementing symbol-like representations in neurally plausible models have been proposed. These approaches include binding through synchrony, “mesh” binding, and conjunctive binding. Recent theoretical work has suggested that most of these methods will not scale well, that is, that they cannot encode structured representations using any of the tens of thousands of terms in the adult lexicon without making implausible resource assumptions. Here, we empirically demonstrate that the biologically plausible structured representations employed in the Semantic Pointer Architecture approach (...) to modeling cognition do scale appropriately. Specifically, we construct a spiking neural network of about 2.5 million neurons that employs semantic pointers to successfully encode and decode the main lexical relations in WordNet, which has over 100,000 terms. In addition, we show that the same representations can be employed to construct recursively structured sentences consisting of arbitrary WordNet concepts, while preserving the original lexical structure. We argue that these results suggest that semantic pointers are uniquely well-suited to providing a biologically plausible account of the structured representations that underwrite human cognition. (shrink)

One of the claims of Type Theory with Records is that it can be used to model types learned by agents in order to classify objects and events in the world, including speech events. That is, the types can be represented by patterns of neural activation in the brain. This claim would be empty if it turns out that the types are in principle impossible to represent on a finite network of neurons. We will discuss how to represent types in (...) terms of neural events on a network and present a preliminary computational implementation that maps types to events on a network. The kind of networks we will use are closely related to the transparent neural networks discussed by Strannegård. (shrink)

Despite its strengths, Leech et al.'s model fails to address the important benefits that derive from self-explanation and task feedback in analogical reasoning development. These components encourage explicit, self-reflective processes that do not necessarily link to knowledge accretion. We wonder, therefore, what mechanisms can be included within a connectionist framework to model self-reflective involvement and its beneficial consequences.

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). To address these issues, a connectionist model of sentence production was developed. The model had variables (role‐concept bindings) that were inspired by spatial representations (Landau & Jackendoff, 1993). In order to take advantage of these variables, a novel (...) dual‐pathway architecture with event semantics is proposed and shown to be better at symbolic generalization than several variants. This architecture has one pathway for mapping message content to words and a separate pathway that enforces sequencing constraints. Analysis of the model's hidden units demonstrated that the model learned different types of information in each pathway, and that the model's compositional behavior arose from the combination of these two pathways. The model's ability to balance symbolic and statistical behavior in syntax acquisition and to model aphasic double dissociations provided independent support for the dual‐pathway architecture. (shrink)

We focus on Karmiloff-Smith's Representational redescription model, arguing that it poses some problems concerning the architecture of a redescribing system. To discuss the topic, we consider the implicit/explicit dichotomy and the relations between natur al language and the language of thought. We argue that the model regards how knowledge is employed rather than how it is represented in the system.

Amidst the progress being made in the various (sub-)disciplines of the behavioural and brain sciences a somewhat neglected subject is the problem of how everything fits into one world and, derivatively, how the relation between different levels of discourse should be understood and to what extent different levels, domains, approaches, or disciplines are autonomous or dependent. In this paper I critically review the most recent proposals to specify the nature of interdiscourse relations, focusing on the concept of supervenience. Ideally supervenience (...) is a relation between different discourses which has all the advantages of reduction, but without its disadvantages. I apply the more abstract considerations to two concrete cases: schizophrenia and colour. Usually an interlevel or interdiscourse relation is seen as asymmetrical: the overlaying discourse depends on the underlying discourse (and not vice versa), where the out- or un-spoken assumption is that the ultimate underlying discourse is physical. Instead I argue that scientific categories referred to in interdiscourse relations are, ultimately, dependent on common sense categories and common sense normative criteria. It is the manifest categories and common sense ideas about what is reasonable and what is right that determine the relevant categorisations at the deeper, underlying levels. I suggest that the implications of this are not merely methodological or epistemological. (shrink)

The target article criticises neural-symbolic systems as inadequate for analogical reasoning and proposes a model of analogy as transformation (i.e., learning). We accept the importance of learning, but we argue that, instead of conflicting, integrated reasoning and learning would model analogy much more adequately. In this new perspective, modern neural-symbolic systems become the natural candidates for modelling analogy.

I argue that SHRUTl's ontology is heavily committed to a representational view of mind. This is best seen when one thinks of how SHRUTI could be developed to account for psychological data on deductive reasoning.

There is a gap between two different modes of computation: the symbolic mode and the subsymbolic (neuron-like) mode. The aim of this paper is to overcome this gap by viewing symbolism as a high-level description of the properties of (a class of) neural networks. Combining methods of algebraic semantics and non-monotonic logic, the possibility of integrating both modes of viewing cognition is demonstrated. The main results are (a) that certain activities of connectionist networks can be interpreted as non-monotonic inferences, and (...) (b) that there is a strict correspondence between the coding of knowledge in Hopfield networks and the knowledge representation in weight-annotated Poole systems. These results show the usefulness of non-monotonic logic as a descriptive and analytic tool for analyzing emerging properties of connectionist networks. Assuming an exponential development of the weight function, the present account relates to optimality theory – a general framework that aims to integrate insights from symbolism and connectionism. The paper concludes with some speculations about extending the present ideas. (shrink)

In agreement with some of the ideas expressed by Perruchet & Vinter (P&V), we believe that some phenomena hitherto attributed to processing may in fact reflect a fundamental distinction between direct and reflexive forms of consciousness. This dichotomy, developed by the phenomenological tradition, is substantiated by examples coming from experimental psychology and lesion neuropsychology.

Connectionist attention to variables has been too restricted in two ways. First, it has not exploited certain ways of doing without variables in the symbolic arena. One variable-avoidance method, that of logical combinators, is particularly well established there. Secondly, the attention has been largely restricted to variables in long-term rules embodied in connection weight patterns. However, short-lived bodies of information, such as sentence interpretations or inference products, may involve quantification. Therefore short-lived activation patterns may need to achieve the effect of (...) variables. The paper is mainly a theoretical analysis of some benefits and drawbacks of using logical combinators to avoid variables in short-lived connectionist encodings without loss of expressive power. The paper also includes a brief survey of some possible methods for avoiding variables other than by using combinators. (shrink)

Prior to the twentieth century, theories of knowledge were inherently perceptual. Since then, developments in logic, statis- tics, and programming languages have inspired amodal theories that rest on principles fundamentally different from those underlying perception. In addition, perceptual approaches have become widely viewed as untenable because they are assumed to implement record- ing systems, not conceptual systems. A perceptual theory of knowledge is developed here in the context of current cognitive science and neuroscience. During perceptual experience, association areas in the (...) brain capture bottom-up patterns of activation in sensory-motor areas. Later, in a top-down manner, association areas partially reactivate sensory-motor areas to implement perceptual symbols. The stor- age and reactivation of perceptual symbols operates at the level of perceptual components – not at the level of holistic perceptual expe- riences. Through the use of selective attention, schematic representations of perceptual components are extracted from experience and stored in memory (e.g., individual memories of green, purr, hot). As memories of the same component become organized around a com- mon frame, they implement a simulator that produces limitless simulations of the component (e.g., simulations of purr). Not only do such simulators develop for aspects of sensory experience, they also develop for aspects of proprioception (e.g., lift, run) and introspec- tion (e.g., compare, memory, happy, hungry). Once established, these simulators implement a basic conceptual system that represents types, supports categorization, and produces categorical inferences. These simulators further support productivity, propositions, and ab- stract concepts, thereby implementing a fully functional conceptual system. Productivity results from integrating simulators combinato- rially and recursively to produce complex simulations. Propositions result from binding simulators to perceived individuals to represent type-token relations. Abstract concepts are grounded in complex simulations of combined physical and introspective events. Thus, a per- ceptual theory of knowledge can implement a fully functional conceptual system while avoiding problems associated with amodal sym- bol systems. Implications for cognition, neuroscience, evolution, development, and artificial intelligence are explored. (shrink)

Various defenses of amodal symbol systems are addressed, including amodal symbols in sensory-motor areas, the causal theory of concepts, supramodal concepts, latent semantic analysis, and abstracted amodal symbols. Various aspects of perceptual symbol systems are clarified and developed, including perception, features, simulators, category structure, frames, analogy, introspection, situated action, and development. Particular attention is given to abstract concepts, language, and computational mechanisms.

According to the target article authors, initial experience with a circumstance primes a relation that can subsequently be applied to a different circumstance to draw an analogy. While I broadly agree with their claim about the role of relational priming in early analogical reasoning, I put forward a few concerns that may be worthy of further reflection.

The rise of Artificial Intelligence (AI) has produced prophets and prophecies announcing that the age of artificial consciousness is near. Not only does the mere idea that any machine could ever possess the full potential of human consciousness suggest that AI could replace the role of God in the future, it also puts into question the fundamental human right to freedom and dignity. This position paper takes the stand that, in the light of all we currently know about brain evolution (...) and the never-stopping formation of adaptive neural circuitry for learning, memory, decision making and, ultimately, fully conscious reasoning and creativity in the human species, the idea of an artificial consciousness appears misconceived. The paper highlights some of the major reasons why. While awareness to external stimuli for processes such as perception, recognition, and operational problem solving is under the direct control of functionally specific brain networks associated with sensory and cognitive functions across animal species, consciousness is a unique property of the human mind. (shrink)