For many people, consciousness is one of the defining characteristics of mental states. Thus, it is quite surprising that consciousness has, until quite recently, had very little role to play in the cognitive sciences. Three very popular multi-authored overviews of cognitive science, Stillings et al. [33], Posner [26], and Osherson et al. [25], do not have a single reference to consciousness in their indexes. One reason this seems surprising is that the cognitive revolution was, in large part, a repudiation of (...) behaviorism's proscription against appealing to inner mental events. When researchers turned to consider inner mental events, one might have expected them to turn to conscious states of mind. But in fact the appeals were to postulated inner events of information processing. The model for many researchers of such information processing is the kind of transformation of symbolic structures that occurs in a digital computer. By positing procedures for performing such transformation of incoming information, cognitive scientists could hope to account for the performance of cognitive agents. Artificial intelligence, as a central discipline of cognitive science, has seemed to impose some of the toughest tests on the ability to develop information processing accounts of cognition: it required its researchers to develop running programs whose performance one could compare with that of our usual standard for cognitive agents, human beings. As a result of this focus, for AI researchers to succeed, at least in their primary task, they did not need to attend to consciousness; they simply had to design programs that behaved appropriately (no small task in itself!). This is not to say that conscious was totally ignored by artificial intelligence researchers. Some aspect of our conscious experience seemed critical to the success of any information processing model. For example, conscious agents exhibit selective attention. Some information received through their senses is attended to; much else is ignored.. (shrink)

The relationship between philosophy and science has always been complementary. Today, while science moves increasingly fast and philosophy shows some problems in catching up with it, it is not always possible to ignore such relationships, especially in some disciplines such as philosophy of mind, cognitive science, and neuroscience. However, the methodological procedures used to analyze these data are based on principles and assumptions that require a profound dialogue between philosophy and science. Following these ideas, this work aims to raise the (...) problems that a classical connectionist theory can cause and problematize them in a cognitive framework, considering both philosophy and cognitive sciences but also the disciplines that are near to them, such as AI, computer sciences, and linguistics. For this reason, we embarked on an analysis of both the computational and theoretical problems that connectionism currently has. The second aim of this work is to advocate for collaboration between neuroscience and philosophy of mind because the promotion of deeper multidisciplinarity seems necessary in order to solve connectionism’s problems. In fact, we believe that the problems that we detected can be solved by a thorough investigation at both a theoretical and an empirical level, and they do not represent an impasse but rather a starting point from which connectionism should learn and be updated while keeping its original and profoundly convincing core. (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)

This chapter explores the epistemic roles played by connectionist models of cognition, and offers a formal analysis of how connectionist models explain. It looks at how other types of computational models explain. Classical artificial intelligence (AI) programs explain using abductive reasoning, or inference to the best explanation; they begin with the phenomena to be explained, and devise rules that can produce the right outcome. The chapter also looks at several examples of connectionist models of cognition, observing what (...) sorts of constraints are used in their design, and how their results are evaluated. It argues that the point of implementing networks roughly analogous to neural structures is to discover and explores the generic mechanisms at work in the brain, not to deduce the precise activities of specific structures. The chapter explores a formal analysis of the explanations offered, which interprets connectionist models and the cognitive theories they represent as sharing membership in a type of mechanism. (shrink)

Classical symbolic computational models of cognition are at variance with the empirical findings in the cognitive psychology of memory and inference. Standard symbolic computers are well suited to remembering arbitrary lists of symbols and performing logical inferences. In contrast, human performance on such tasks is extremely limited. Standard models donot easily capture content addressable memory or context sensitive defeasible inference, which are natural and effortless for people. We argue that Connectionism provides a more natural framework in which to model this (...) behaviour. In addition to capturing the gross human performance profile, Connectionist systems seem well suited to accounting for the systematic patterns of errors observed in the human data. We take these arguments to counter Fodor and Pylyshyn's (1988) recent claim that Connectionism is, in principle, irrelevant to psychology. (shrink)

Tae Wan Kim, John Hooker, and Thomas Donaldson make an attempt, in recent articles, to solve the alignment problem. As they define the alignment problem, it is the issue of how to give AI systems moral intelligence. They contend that one might program machines with a version of Kantian ethics cast in deontic modal logic. On their view, machines can be aligned with human values if such machines obey principles of universalization and autonomy, as well as a deontic utilitarian principle. (...) Programming machines to do so might be useful, in their view, for applications such as future autonomous vehicles. Their proposal draws both on traditional logic-based and contemporary connectionist approaches, to fuse factual information with normative principles. I will argue that this approach makes demands of machines that go beyond what is currently feasible, and may extend past the limits of the possible for AI. I also argue that a deontological ethics for machines should place greater stress on the formula of humanity of the Kantian categorical imperative. On this principle, one ought never treat a person as a mere means. Recognition of what makes a person a person requires ethical insight. Similar insight is needed to tell treatment as a means from treatment as a mere means. The resources in Kim, Hooker, and Donaldson’s approach is insufficient for this reason. Hesitation regarding deployment of autonomous machines is warranted in light of these alignment concerns. (shrink)

Critics of the computational connectionism of the last decade suggest that it shares undesirable features with earlier empiricist or associationist approaches, and with behaviourist theories of learning. To assess the accuracy of this charge the works of earlier writers are examined for the presence of such features, and brief accounts of those found are given for Herbert Spencer, William James and the learning theorists Thorndike, Pavlov and Hull. The idea that cognition depends on associative connections among large networks of neurons (...) is indeed one with precedents, although the implications of this for psychological issues have been interpreted variously — not all versions of connectionism are alike. (shrink)

The current renewal of connectionist techniques using networks of neuron-like units has started to have an influence on cognitive modelling. However, compared with classical artificial intelligence methods, the position of connectionism is still not clear. In this article artificial intelligence and connectionism are systematically compared as cognitive models so as to bring out the advantages and shortcomings of each. The problem of structured representations appears to be particularly important, suggesting likely research directions.

It is claimed that there are pre-objective phenomena, which cognitive science should explain by employing the notion of non-conceptual representational content. It is argued that a match between parallel distributed processing (PDP) and non-conceptual content (NCC) not only provides a means of refuting recent criticisms of PDP as a cognitive architecture; it also provides a vehicle for NCC that is required by naturalism. A connectionist cognitive mapping algorithm is used as a case study to examine the affinities between PDP (...) and NCC. (shrink)

This article compares the potential of classical and connectionist computational concepts for explanations of innate infant knowledge and of its development. It focuses on issues relating to: the perceptual process; the control and form(s) of perceptual-behavioural coordination; the role of environmental structure in the organization of action; and the construction of novel forms of activity. There is significant compatibility between connectionist and classical views of computation, though classical concepts are, at present, better able to provide a comprehensive computational (...) view of the infant. However, Varela's “enaction” perspective poses a significant challenge for both approaches. (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)

In this paper I put forward a reconstruction of the evolution of certain explanatory hypotheses on the neural basis of association and learning that are the premises of connectionism in the cybernetic age and of present-day connectionism. The main point of my reconstruction is based on two little-known case studies. The first is the project, published in 1913, of a hydraulic machine through which its author believed it was possible to simulate certain essential elements of the plasticity of nervous connections. (...) The author, S. Bent Russell, was an engineer deeply influenced by the neurological hypotheses on nervous conduction of Herbert Spencer, Max Meyer and Edward L. Thorndike. The second is the project, published in 1929, of an electromechanical machine in which the author, the psychologist J.M. Stephens, believed it was possible to embody Thorndike's law of effect. Thus both Bent Russell and Stephens referred to the principles of learning that Thorndike defined as connectionist . Their attempt was that of simulating by machines at least certain simple aspects of inhibition, association and habit formation that are typical of living organisms. I propose to situate their projects within the frame of thediscovery of a simulative (modelling) methodology which I believe might be considered an important topic of the Culture of the Artificial . Certain more recent steps toward such a methodology made by both connectionism of the 1950s and present-day connectionism are briefly pointed out in the paper. (shrink)

According to the standard (recent) history of connectionism (see for example the accounts offered by Hecht-Nielsen (1990: pp. 14-19) and Dreyfus and Dreyfus (1988), or Papert's (1988: pp. 3-4) somewhat whimsical description), in the early days of Classical Computational Theory of Mind (CCTM) based AI research, there was also another allegedly distinct approach, one based upon network models. The work on network models seems to fall broadly within the scope of the term 'connectionist' (see Aizawa 1992), although the term (...) had yet to be coined at the time. These two approaches were "two daughter sciences" according to Papert (1988: p. 3). The fundamental difference between these two 'daughters', lay (according to Dreyfus and Dreyfus (1988: p. 16)) in what they took to be the paradigm of intelligence. Whereas the early connectionists took learning to be fundamental, the traditional school concentrated upon problem solving. (shrink)

ABSTRACT Usually, the problems in AI may be many times related to Philosophy of Mind, and perhaps because this reason may be in essence very disputable. So, for instance, the famous question: Can a machine think? It was proposed by Alan Turing [16]. And it may be the more decisive question, but for many people it would be a nonsense. So, two of the very fundamental and more confronted positions usually considered according this line include the Connectionism and the Computational (...) Theory of Mind. We analyze here its content, with their past disputes, and current situation. (shrink)

The expression ‘‘artificial intelligence’’ (AI) was introduced by John McCarthy, and the official birth of AI is unanimously considered to be the 1956 Dartmouth Conference. Thus, AI turned fifty in 2006. How did AI begin? Several differently motivated analyses have been proposed as to its origins. In this paper a brief look at those that might be considered steps towards Dartmouth is attempted, with the aim of showing how a number of research topics and controversies that marked the short history (...) of AI were touched on, or fairly well stated, during the year immediately preceding Dartmouth. The framework within which those steps were taken was the development of digital computers. Earlier computer applications in areas such as complex decision making and management, at that time dealt with by operations research techniques, were important in this story. The time was ripe for AI’s intriguingly tumultuous development, marked as it has been by hopes and defeats, successes and difficulties. (shrink)

In the first section of the article, we examine some recent criticisms of the connectionist enterprise: first, that connectionist models are fundamentally behaviorist in nature (and, therefore, non-cognitive), and second that connectionist models are fundamentally associationist in nature (and, therefore, cognitively weak). We argue that, for a limited class of connectionist models (feed-forward, pattern-associator models), the first criticism is unavoidable. With respect to the second criticism, we propose that connectionist modelsare fundamentally associationist but that this (...) is appropriate for building models of human cognition. However, we do accept the point that there are cognitive capacities for which any purely associative model cannot provide a satisfactory account. The implication that we draw from is this is not that associationist models and mechanisms should be scrapped, but rather that they should be enhanced.In the next section of the article, we identify a set of connectionist approaches which are characterized by “active symbols” — recurrent circuits which are the basis of knowledge representation. We claim that such approaches avoid criticisms of behaviorism and are, in principle, capable of supporting full cognition. In the final section of the article, we speculate at some length about what we believe would be the characteristics of a fully realized active symbol system. This includes both potential problems and possible solutions (for example, mechanisms needed to control activity in a complex recurrent network) as well as the promise of such systems (in particular, the emergence of knowledge structures which would constitute genuine internal models). (shrink)

In contemporary artificial intelligence, the challenge is making intricate connectionist systems—comprising millions of parameters—more comprehensible, defensible, and rationally grounded. Two prevailing methodologies address this complexity. The inaugural approach amalgamates symbolic methodologies with connectionist paradigms, culminating in a hybrid system. This strategy systematizes extensive parameters within a limited framework of formal, symbolic rules. Conversely, the latter strategy remains staunchly connectionist, eschewing hybridity. Instead of internal transparency, it fabricates an external, transparent proxy system. This ancillary system’s mandate is elucidating (...) the principal system’s decisions, essentially approximating its outcomes. Leveraging natural language processing as our analytical lens, this paper elucidates both methodologies: the hybrid method is underscored by the compositional vector semantics, whereas the purely connectionist method evolves as a derivative of neural semantic parsers. This discourse extols the merits of the purely connectionist approach for its inherent flexibility and for a pivotal delineation: segregating the explanatory apparatus from the operational core, thereby rendering artificial intelligence systems reminiscent of human cognition. (shrink)

Abstract The notion of representation lies at the crossroads of questions about the nature of belief and knowledge, meaning, and intentionality. But there is some hope that it might be simpler than all those. If we could understand it clearly, it might then help to explicate those more difficult notions. In this paper, my central aim is to find a principled criterion, along lines that make biological sense, for deciding just when it becomes theoretically plausible to ascribe to some process (...) or state a representational role. I shall be especially concerned with some differences, in this regard, between classical and connectionist models. The relation between ?standard? artificial intelligence and connectionism turns out to illustrate a ?first in, last out? principle: What we most easily understand (and so can program) is what we have most recently invented; tasks we ourselves perform best, by contrast, are a lot harder to understand. Classical AI has modelled the former; connectionism tries to tackle the latter. I end with some speculations about the possible implications of these considerations for our understanding of understanding. (shrink)

Horgan and Tienson do a wonderful job of explicating the dynamical system perspective and contrasting that view with classical AI approaches. However, their arguments for replacing a classical conception of connectionism with system dynamics rely on philosophical distinctions that do not make a difference. In particular, (1) their generalized version of Man's three levels of analysis collapses into itself; (2) their description of attractor dynamics works better than their metaphor of forces; and (3) their versions of “soft laws” and physical (...) laws amount to the same thing. (shrink)

According to how AI has defined itself from its beginning, thinking in non-living matter, i.e., without life, is possible. The premise of symbolic AI is that operating on representations of reality machines can understand it. When this assumption did not work as expected, the mathematical model of the neuron became the engine of artificial “brains.” Connectionism followed. Currently, in the context of Machine Learning success, attempts are made at integrating the symbolic and connectionist paths. There is hope that Artificial (...) General Intelligence (AGI) performance can be achieved. As encouraging as neuro-symbolic AI seems to be, it remains unclear whether AGI is actually a moving target as long as AI itself remains ambiguously defined. This paper makes the argument that the intelligence of machines, expressed in their performance, reflects how adequate the means used for achieving it are. Therefore, energy use and the amount of data necessary qualify as a good metric for comparing natural and artificial performance. (shrink)

In this paper I shall describe the symbolic search space paradigm which is the dominant model for most of AI. Coupled with the mechanisms of logic it yields the predominant methodology underlying expert systems which are the most successful application of AI technology to date. Human decision making, more precisely, expert human decision making is the function that expert systems aspire to emulate, if not surpass.Expert systems technology has not yet proved to be a decisive success — it appears to (...) fare better in some areas of human expertise than others. As a result subdomains of human expertise are variously categorised and we shall examine a few of the suggested classification schemes. A particular line of argument explored is one which maintains that certain types of human decision making, at least, are not adequately approximated by the symbolic search space paradigm of AI. Furthermore, attempts to project this inadequate model of human decision making via implementations of expert systems will be detrimental to both our image of ourselves and the future possibilities for AI software.Finally, we examine one possible route to the realization of AI, perhaps even practical applications of AI, that is a significant alternative to the model offered by the symbolic search space paradigm. (shrink)

An academic anthology on Jewish studies and education in memory of dr. Avraham Zalkin.

In some languages every statement must contain a specification of the type of evidence on which it is based: for example, whether the speaker saw it, or heard it, or inferred it from indirect evidence, or learnt it from someone else. This grammatical reference to information source is called 'evidentiality', and is one of the least described grammatical categories. Evidentiality systems differ in how complex they are: some distinguish just two terms (eyewitness and noneyewitness, or reported and everything else), while (...) others have six or even more terms. Evidentiality is a category in its own right, and not a subcategory of epistemic or some other modality, nor of tense-aspect. Every language has some way of referring to the source of information, but not every language has grammatical evidentiality. In English expressions such as I guess, they say, I hear that, the alleged are not obligatory and do not constitute a grammatical system. Similar expressions in other languages may provide historical sources for evidentials. True evidentials, by contrast, form a grammatical system. In the North Arawak language Tariana an expression such as "the dog bit the man" must be augmented by a grammatical suffix indicating whether the event was seen, or heard, or assumed, or reported. This book provides the first exhaustive cross-linguistic typological study of how languages deal with the marking of information source. Examples are drawn from over 500 languages from all over the world, several of them based on the author's original fieldwork. Professor Aikhenvald also considers the role evidentiality plays in human cognition, and the ways in which evidentiality influences human perception of the world.. This is an important book on an intriguing subject. It will interest anthropologists, cognitive psychologists and philosophers, as well as linguists. (shrink)

Knowledge can be expressed in language using a plethora of grammatical means. Four major groups of meanings related to knowledge are Evidentiality: grammatical expression of information source; Egophoricity: grammatical expression of access to knowledge; Mirativity: grammatical expression of expectation of knowledge; and Epistemic modality: grammatical expression of attitude to knowledge. The four groups of categories interact. Some develop overtones of the others. Evidentials stand apart from other means in many ways, including their correlations with speech genres and social environment. This (...) essay presents a framework which connects the expression of knowledge across the world's languages in a coherent way, showing their dependencies and complexities, and pathways of historical development in various scenarios, including language obsolescence. (shrink)