[Edited from Conclusion section:] We have looked at various challenging issues to do with getting connectionism to cope with high-level cognitive activities such a reasoning and natural language understanding. The issues are to do with various facets of generalization that are not commonly noted. We have been concerned in particular with the special forms these issues take in the arena of propositional attitude processing. The main problems we have looked at are: (1) The need to construct explicit representations of (...) generalizations, not just generalize correctly to individual cases; (2) The need to be able to match two or more complex short-term information structures, to enable rapid generalization from recent examples rather than from long-term memories; (3) The need to represent and reason with anomalous combinations of concepts; (4) The need to perform embedded reasoning. This presents special problems for systems using non-concatenative representations (as in mainstream connectionist approaches). We also touched on vague quantification in attitude report complements. Neither this topic nor that of analogies between short-term structures (point 2) has been adequately addressed in the symbolic framework, let alone in connectionism. -/- The opportunities and problems covered are put forward as things worth being optimistic about or pessimistic about, respectively. They are not put forward as decisive arguments for or against connectionism. The hope is that this chapter contributes to a greater understanding of the connectionist/symbolist gap by presenting some unusual issues and by throwing new light on some well known ones. (shrink)

Connectionism seems to avoid many of the problems of classical artificial intelligence, but has it avoided all of them? In this thesis I examine the problem that Intentionality, the directedness of thought to an object, raises for connectionism. As a preliminary approach, I consider the role of Intentionality in classical artificial intelligence from the programmer’s point of view. In this investigation, one problem I identify with classical artificial intelligence is that the Intentionality of the programmer seems to be (...) projected onto the system, rather than the programmer creating a system whereby Intentionality arises intrinsically within the system. -/- In considering the current practice of connectionism, the same problem with Intentionality reappears. The assignment of Intentional content to input or output nodes in a neural network likewise projects the Intentionality of the programmer onto the system, and that projection is often reinforced by the training process of the neural network. However, connectionism seems to have an advantage over classical artificial intelligence in this respect, in that there is a form of neural network in which the network itself organizes the output nodes. The challenge is to utilize these self-organizing networks without having the programmer project Intentionality onto the system in an act of interpretation. -/- I suggest that the way to overcome these problems in connectionism is to embody the neural network within a world, whether physical or virtual, and to allow the system to develop concepts purely empirically. In this manner, connectionism can make scientific observations concerning the nature of cognition without the risk of contamination from the programmer’s own Intentionality. (shrink)

Fodor and Pylyshyn's critique of connectionism has posed a challenge to connectionists: Adequately explain such nomological regularities as systematicity and productivity without postulating a "language of thought" (LOT). Some connectionists like Smolensky took the challenge very seriously, and attempted to meet it by developing models that were supposed to be non-classical. At the core of these attempts lies the claim that connectionist models can provide a representational system with a combinatorial syntax and processes sensitive to syntactic structure. They are (...) not implementation models because, it is claimed, the way they obtain syntax and structure sensitivity is not "concatenative," hence "radically different" from the way classicists handle them. In this paper, I offer an analysis of what it is to physically satisfy/realize a formal system. In this context, I examine the minimal truth-conditions of LOT Hypothesis. From my analysis it will follow that concatenative realization of formal systems is irrelevant to LOTH since the very notion of LOT is indifferent to such an implementation level issue as concatenation. I will conclude that to the extent to which they can explain the law-like cognitive regularities, a certain class of connectionist models proposed as radical alternatives to the classical LOT paradigm will in fact turn out to be LOT models, even though new and potentially very exciting ones. (shrink)

In his book The Sensory Order, Hayek anticipates many of the central ideas behind what we now call the connectionist paradigm, and develops on this basis a theory of the workings of the human mind that extends the thinking of Hume and Mach. He shows that the idea of neural networks is can be applied not only in psychology and neurology but also in the sphere of economics. For the mind, from the perspective of The Sensory Order, is a dynamic, (...) relational affair that is in many respects analogous to a market process. The mind, as Hayek puts it, is a 'continuous stream of impulses, the significance of each and every contribution of which is determined by the place in the pattern of channels through which they flow' , so that the flow of representative neural impulses can be compared 'to a stock of capital being nourished by inputs and giving a continuous stream of outputs'. (shrink)

This paper has a two-fold aim. First, it reinforces a version of the "syntactic argument" given in Aizawa (1994). This argument shows that connectionist networks do not provide a means of implementing representations without rules. Horgan and Tlenson have responded to the syntactic argument in their book and in another paper (Horgan & Tlenson, 1993), but their responses do not meet the challenge posed by my formulation of the syntactic argument. My second aim is to describe a kind of cognitive (...) architecture. This architecture might be called a computational architecture, but it is not a rules and representations architecture nor the representations without rules architecture that Horgan and Tlenson wish to endorse. (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)

This paper is devoted to some generalizations of explanatory potential of connectionist approaches to theoretical problems of the philosophy of mind. Are considered both strong, and weaknesses of neural network models. Connectionism has close methodological ties with modern neurosciences and neurophilosophy. And this fact strengthens its positions, in terms of empirical naturalistic approaches. However, at the same time this direction inherits weaknesses of computational approach, and in this case all system of anticomputational critical arguments becomes applicable to the connectionst (...) models of mind. The last developments in the field of deep learning gave rich empirical material for cognitive sciences. Multilayered networks, mathematical models of associative dynamics of learning, self-organizing neuronets and all that allow to explain the principles of human conceptual organizing and after this to emulate these processes in computer systems. At all engineering achievements of this technology there is a traditional criticism from representatives of cognitive psychology who cannot accept a thesis about learning ability of a neuronet on the basis of redistribution of scales. Process of learning of natural intelligence, according to cognitive models, happens due to attraction of knowledge broadcast in a symbolical form (mental representations, concepts) at the expense of the systems of output knowledge expressed in the propositional contents. Some philosophical aspects of «neural metaphor» in modern cognitive sciences create the problem field which demands comprehensive understanding, the first step towards which is taken in this work. (shrink)

This paper is a commentary on the target article by Dana H. Ballard, “Cortical connections and parallel processing: Structure and function”, in the same issue of the journal, pp. 67–120. -/- I raise some issues about the connectionist or neural-network implementation of information and information processing. Issues include the sharing of information by different parts of a connectionist/neural network, the copying of complex information from one place to another in a network, the possibility of connection weights not being synaptic weights, (...) and the possibility of fast communication within single neurons as complex systems in their own right. (shrink)

The paper is a commentary on the target article by Christine A. Skarda & Walter J. Freeman, “How brains make chaos in order to make sense of the world”, in the same issue of the journal, pp.161–195. -/- I confine my comments largely to some philosophical claims that Skarda & Freeman make and to the relationship of their model to connectionism. Some of the comments hinge on what symbols are and how they might sit in neural systems.

In cognitive science, there are many computational theories regarding the function of the mind; connectionism is one of them. Connectionist networks are intricate systems of simple units related to their environment. Some have thousands of units, but those with only a few units can also behave with surprising complexity and subtlety. This is because processing occurs in parallel as also interactively, in marked contrast with the serial processing to which this is accustomed. In the first section of this paper, (...) I intend to describe a simple network that illustrates several features of connectionist processing. Secondly, I would like to examine its relation with other areas in realm of cognitive science. Thirdly, I shall make an attempt to find out whether this theory contributes to the replacement of folk psychology. Lastly, I find that connectionist thus fails to account for the real nature of the mental states because of its not too clear attempt to reduce mental states to the machine states. The mechanistic theory of mind in all its hues faces the question as to how we can account for the qualitative content of our consciousness. It cannot ultimately tell us how the subjective experience is possible and how consciousness to be real feature of universe. (shrink)

Philosophy and Memory Traces defends two theories of autobiographical memory. One is a bewildering historical view of memories as dynamic patterns in fleeting animal spirits, nervous fluids which rummaged through the pores of brain and body. The other is new connectionism, in which memories are 'stored' only superpositionally, and reconstructed rather than reproduced. Both models, argues John Sutton, depart from static archival metaphors by employing distributed representation, which brings interference and confusion between memory traces. Both raise urgent issues about (...) control of the personal past, and about relations between self and body. Sutton demonstrates the role of bizarre body fluids in moral physiology, as philosophers from Descartes and Locke to Coleridge struggled to control their own innards and impose cognitive discipline on 'the phantasmal chaos of association'. Going on to defend connectionism against Fodor and critics of passive mental representations, he shows how problems of the self are implicated in cognitive science. (shrink)

Philosophers have argued that on the prevailing theory of mind, functionalism, the fact that mental states are multiply realizable or can be instantiated in a variety of different physical forms, at least in principle, shows that materialism or physical is probably false. A similar argument rejects the relevance to psychology of connectionism, which holds that mental states are embodied and and constituted by connectionist neural networks. These arguments, I argue, fall before reductios ad absurdam, proving too much -- they (...) apply as well to genes, which are multiply realizable, but the reduction of which to DNA is one the core cases of scientific reductive explanation, a reduction if anything is. -/- I suggest that psychology, like biology, be what I call "provincialized," abandon claims to universal validity, except as an idealization, and treat different classes of cognizers differently from an an explanatory perspective. This would permit specifies specific, or more precisely, provincial reductions of different psychologies that might be multiply realized if such reductions were available. Connectionism may be the foundation of a reduction of human psychology, and thus biology and connectionism retain their relevance to psychology, and physicalism or materialism is consistent with functionalism. (shrink)

Three decades ago, William Ramsey, Steven Stich & Joseph Garon put forward an argument in favor of the following conditional: if connectionist models that implement parallelly distributed processing represent faithfully human cognitive processing, eliminativism about propositional attitudes is true. The corollary of their argument (if it proves to be sound) is that there is no place for folk psychology in contemporary cognitive science. This understanding of connectionism as a hypothesis about cognitive architecture compatible with eliminativism is also endorsed by (...) Paul Churchland, a radical opponent of folk psychology and a prominent supporter of eliminative materialism. I aim to examine whether current connectionist models based on long-short term memory (LSTM) neural networks can back up these arguments in favor of eliminativism. Nonetheless, I will rather put my faith in the eliminativism of the limited domain. This position amounts to the claim that even though connectionist cognitive science has no need whatsoever for folk psychology qua theory, this does not entail the illegitimacy of folk psychology per se in other scientific domains, most notably in humanities, but only on condition that one regards folk psychology as mere heuristics. (shrink)

This draft preprint presents a nine step argument for “Connectionist Structuralism” (CS), an account of the ontology of abstract objects that is neither purely nominalist nor purely platonist. CS is a common, often implicit assumption in parts of the artificial intelligence literature, but such discussions have not presented formal accounts of the position or engaged with metaphysical issues that potentially undermine it. By making the position legible and presenting an initial case for it, we hope to support a constructive dialogue (...) between AI researchers and philosophers of metaphysics that helps both sides to refine the position. -/- CS proposes that each abstract object we can draw on in human analysis corresponds to a particular subset of an individual person’s brain structure whose functionality is isomorphic to a subset of the nodes and connections in a suitable connectionist network. In other words, abstract objects are physically realised, but in individual brains, rather than only in the referent objects (pure nominalism) or in metaphysical universals (pure platonism). -/- This paper’s minimum claim is that CS can account for all abstract object predicables regarding sensible properties, such as “is red” or “is a square”. Using evidence from cognitive neuroscience, machine learning, and evolutionary biology, as well as a fully traceable toy example, we describe how CS can support our core cognitive uses of such sensible properties and can account for our core phenomenal experiences of them. In the former, CS provides sufficient albeit imperfect inferential safety, whose limitations are argued to strengthen rather than weaken the case for CS as describing human behaviour. In the latter, four target phenomenal features are accounted for – abstract objects as feeling intangible, non-located, transparent, and unchanging - along with accounts for further phenomena such as semantic refinement, the Stroop effect, synaesthesia, semantic clarity, sensory overload, and satiation. -/- Our minimum claim concerns the day-to-day usage and felt experience of abstract objects, but we suggest also an extended claim in which CS can form the basis of a pragmatic sufficient logic. As such, the initial outline of a response is provided for five common objections to positions that seek to ground abstract objects without reference to metaphysically stand-alone universals: referential opacity; identity of indiscernibles; infinite regression; non-physical concepts; and necessary truths. These outlines lay the foundation for (but do not seek to formally demonstrate) an extended CS account that addresses other abstract objects, including issues relating to their use in mathematics and formal logic. -/- The CS account leads to a four-layer hierarchy of similarity for whether your “red” is the same as mine. Considering both semantic and phenomenal similarity, we conclude that our “reds” are likely non-identical but can be made close enough for practical purposes. Finally, we describe how future work could elaborate CS as a metaphysical project and how confidence in it could be tested through empirical research. (shrink)

Cognitive scientists used to deem reasoning either as a higher cognitive process based on the manipulation of abstract rules or as a higher cognitive process that is stochastic rather than involving abstract rules. I maintain that these different perspectives are closely intertwined with a theoretical and methodological endorsement of either cognitivism or connectionism. Cognitivism and connectionism represent two prevailing and opposed paradigms in cognitive science. I aim to extoll the virtues of connectionist models of enthymematic reasoning by following (...) means: via the phenomenon of creative enthymeme, viz. the inference where one cannot even articulate the missing premise, I introduce a connectionist mechanism of pattern recognition as underlying expertise; via Gestalt switch or Gestalt click, I demonstrate how differences in pattern recognition of an expert and a novice can be construed as qualitatively different, and not merely a matter of faster reasoning. (shrink)

Recent work in the methodology of connectionist explanation has I'ocrrsccl on the notion of levels of explanation. Specific issucs in conncctionisrn hcrc intersect with rvider areas of debate in the philosophy of psychology and thc philosophy of science generally. The issues I raise in this chapter, then, are not unique to cognitive science; but they arise in new and important contexts when connectionism is taken seriously as a model of cognition. The general questions are the relation between levels and (...) the status of levels which have no obvious relation to others. In speaking of levels, what is the connection, if there is one, between explanation and ontology? Which, if any, conccpt of reduction is applicable to connectionist systems? What kind of legitinrtcy can the constructs of common sense psychology, or of that vclsion ol intentional realism represented by classical symbol-systems n I, hirvc irr ir full-scale connectionist theory of mind? (shrink)

In Book I, part iv, section 2 of the Treatise, "Of scepticism with regard to the senses," Hume presents two different answers to the question of how we come to believe in the continued existence of unperceived objects. He rejects his first answer shortly after its formulation, and the remainder of the section articulates an alternative account of the development of the belief. The account that Hume adopts, however, is susceptible to a number of insurmountable objections, which motivates a reassessment (...) of his original proposal. This paper defends a version of Hume's initial explanation of the belief in continued existence and examines some of its philosophical implications. (shrink)

In artificial intelligence, recent research has demonstrated the remarkable potential of Deep Convolutional Neural Networks (DCNNs), which seem to exceed state-of-the-art performance in new domains weekly, especially on the sorts of very difficult perceptual discrimination tasks that skeptics thought would remain beyond the reach of artificial intelligence. However, it has proven difficult to explain why DCNNs perform so well. In philosophy of mind, empiricists have long suggested that complex cognition is based on information derived from sensory experience, often appealing to (...) a faculty of abstraction. Rationalists have frequently complained, however, that empiricists never adequately explained how this faculty of abstraction actually works. In this paper, I tie these two questions together, to the mutual benefit of both disciplines. I argue that the architectural features that distinguish DCNNs from earlier neural networks allow them to implement a form of hierarchical processing that I call “transformational abstraction”. Transformational abstraction iteratively converts sensory-based representations of category exemplars into new formats that are increasingly tolerant to “nuisance variation” in input. Reflecting upon the way that DCNNs leverage a combination of linear and non-linear processing to efficiently accomplish this feat allows us to understand how the brain is capable of bi-directional travel between exemplars and abstractions, addressing longstanding problems in empiricist philosophy of mind. I end by considering the prospects for future research on DCNNs, arguing that rather than simply implementing 80s connectionism with more brute-force computation, transformational abstraction counts as a qualitatively distinct form of processing ripe with philosophical and psychological significance, because it is significantly better suited to depict the generic mechanism responsible for this important kind of psychological processing in the brain. (shrink)

This paper aims to offer a new view of the role of connectionist models in the study of human cognition through the conceptualization of the history of connectionism – from the simplest perceptrons to convolutional neural nets based on deep learning techniques, as well as through the interpretation of criticism coming from symbolic cognitive science. Namely, the connectionist approach in cognitive science was the target of sharp criticism from the symbolists, which on several occasions caused its marginalization and almost (...) complete abandonment of its assumptions in the study of cognition. Criticisms have mostly pointed to its explanatory inadequacy as a theory of cognition or to its biological implausibility as a theory of implementation, and critics often focused on specific shortcomings of some connectionist models and argued that they apply to connectionism in general. In this paper, we want to show that both types of critique are based on the assumption that the only valid explanations in cognitive science are instances of homuncular functionalism and that by removing this assumption and by adopting an alternative methodology – exploratory mechanistic strategy, we can reject most objections to connectionism as irrelevant, explain the progress of connectionist models despite their shortcomings and sketch the trajectory of their future development. By adopting mechanistic explanations and by criticizing functionalism, we will reject the objections of explanatory inadequacy, by characterizing connectionist models as generic rather than concrete mechanisms, we will reject the objections of biological implausibility, and by attributing the exploratory character to connectionist models we will show that practice of generalizing current to general failures of connectionism is unjustified. (shrink)

1. Do models formulated in programming languages use explicit rules where connectionist models do not? 2. Are rules as found in programming languages hard, precise, and exceptionless, where connectionist rules are not? 3. Do connectionist models use rules operating on distributed representations where models formulated in programming languages do not? 4. Do connectionist models fail to use structure sensitive rules of the sort found in "classical" computer architectures? In this chapter we argue that the answer to each of these questions (...) is negative. (shrink)

The emerging contemporary natural philosophy provides a common ground for the integrative view of the natural, the artificial, and the human-social knowledge and practices. Learning process is central for acquiring, maintaining, and managing knowledge, both theoretical and practical. This paper explores the relationships between the present advances in understanding of learning in the sciences of the artificial (deep learning, robotics), natural sciences (neuroscience, cognitive science, biology), and philosophy (philosophy of computing, philosophy of mind, natural philosophy). The question is, what at (...) this stage of the development the inspiration from nature, specifically its computational models such as info-computation through morphological computing, can contribute to machine learning and artificial intelligence, and how much on the other hand models and experiments in machine learning and robotics can motivate, justify, and inform research in computational cognitive science, neurosciences, and computing nature. We propose that one contribution can be understanding of the mechanisms of ‘learning to learn’, as a step towards deep learning with symbolic layer of computation/information processing in a framework linking connectionism with symbolism. As all natural systems possessing intelligence are cognitive systems, we describe the evolutionary arguments for the necessity of learning to learn for a system to reach human-level intelligence through evolution and development. The paper thus presents a contribution to the epistemology of the contemporary philosophy of nature. (shrink)

Paul Cilliers has developed a novel post-structural approach to complexity that has influenced several writers contributing to the current complexity literature. Concomitantly however, Cilliers advocates for modelling complex systems using connectionist neural networks (rather than analytic, rule-based models). In this paper, I argue that it is dilemmic to simultaneously hold these two positions. Cilliers’ post-structural interpretation of complexity states that models of complex systems are always contextual and provisional; there is no exclusive model of complex systems. This sentiment however appears (...) at odds with Cilliers’ promotion of connectionist neural networks as the best way to model complex systems. The lesson is that those who currently follow Cilliers’ post-structural approach to complexity cannot also develop a preferred model of complex systems, and those who currently advocate for some preferred model of complex systems cannot adopt the post-structural approach to complexity without giving up the purported objectivity and/or superiority of their preferred model. (shrink)

Some contemporary philosophers suggest that we know just by introspection that folk psychological states exist. However, such an "armchair refutation" of eliminativism seems too easy. I first attack two strategems, inspired by Descartes, on how such a refutation might proceed. However, I concede that the Cartesian intuition that we have direct knowledge of representational states is very powerful. The rest of this paper then offers an error theory of how that intuition might really be mistaken. The idea is that introspection (...) does not detect any folk psychological states, but rather detects neural states that dispose us to certain sorts of linguistic behavior. Briefly, rather than detecting a mental appearance of the state-of-affairs that I am sitting near the fire, introspection detects a state disposing me to assert "It appears to me that I am sitting by the fire." The nature of this linguistic-dispositional state is then given a connectionist underpinning, where inputs from introspection lead to such assertive behavior. The Cartesian certainty we feel in self-ascribing such appearances is also explained. Even though I do not endorse such an error theory outright, I suggest that its tenability is enough to undercut introspection arguments against eliminativism. (shrink)

The systematicity argument only challenges connectionism if systematicity is a general property of cognition. I examine this thesis in terms of properties of concepts. First, I propose that Evans's Generality Constraint only applies to attributions of belief. Then I defend a variety of conceptual pluralism, arguing that concepts share two fundamental properties related to centrality and belief-attribution, and contending that there are two kinds of concepts that differ in their compositional properties. Finally, I rely on Dual Systems Theory and (...) on differences between animal and human cognition to suggest a scenario of two processing systems that work on different kinds of concepts, with only one of them supporting full systematicity. I sketch a non-classical systematicity argument that rules out classicism as the basis of one of those systems given that it would wrongly entail that both systems are fully systematic. (shrink)

This paper is a commentary on the target article by Lokendra Shastri & Venkat Ajjanagadde [S&A]: “From simple associations to systematic reasoning: A connectionist representation of rules, variables and dynamic bindings using temporal synchrony” in same issue of the journal, pp.417–451. -/- It puts S&A's temporal-synchrony binding method in a broader context, comments on notions of pointing and other ways of associating information - in both computers and connectionist systems - and mentions types of reasoning that are a challenge to (...) S&A's system (amongst many other connectionist systems, recognizing that S&A's is much more capable as regards reasoning than most other contemporary systems). (shrink)

This paper is a commentary on the target article by Michael Arbib, “Levels of modeling of mechanisms of visually guided behavior”, in the same issue of the journal, pp. 407–465. -/- I focus on the importance of the inclusion of an ability of a system to entertain, at a given time, multiple instantiations of a given schema (situation template, frame, script, action plan, etc.), and complications introduced into neural/connectionist network systems by such inclusion.

In this Master's dissertation, I try to show that the causal theory of memory, which is the only theory developed so far that at first view seems more plausible and that could be integrated with psychological explanations and investigations of memory, shows some conceptual and ontological problems that go beyond the internal inconsistencies that each version can present. On one hand, the memory phenomenon analyzed is very limited: in general it is reduced to the conscious act of remembering expressed in (...) a propositional format: the idea of an agent who can control his memory reports, in contrast to a passive subject who merely retrieve his encoded memories, is not even considered. Furthermore, the representational function of memory is the only function taken into account; and all the diachronic changes that could entail a dynamic conception of memory are ignored, as is the consideration of context in a broad sense of the term. On the other hand, causal theories make some implicit assumptions that are obviously questionable (e.g., the difference between facts and events) and leave some of their key concepts unexplained, like the nature of the causality, the isomorphism between the memory trace and the memory event, etc. Nonetheless, the principal point that we criticized is the idea of a correspondence between a past event or representation of it, the correspondent memory trace and the memory representation of it. Psychological models of memory, with the exception of the spatial analogy, question some of these assumptions and suggest that memory information is organized in a way other than that implied by the causal theory. But it is distributed connectionism that actually challenges the causal theory of memory, not only through some of its basic postulates, like the superpositional conception of memory traces, their mutual influence and constant state of change, but also because it is compatible with a more situated approach to memory phenomena as well as with the idea that memory in reality accomplishes a variety of functions that go beyond the representational one. The explicit and implicit ideas provided by distributed connectionism, together with the previous criticisms that we made to of causal theory, clearly suggest the possibility of developing another kind of philosophical theory of memory, different from existing theories. This task, nonetheless, implies: (a) A much deeper study of connectionism principles and memory models, as well as of the rival candidate, symbolism, plus the evaluation of these models in the light of psychological and neurophysiological empirical research, which unfortunately I have not dealt with in this work. (b) An analysis of the metaphysical nature of events and facts, in order to see if we should decide in favor of a realist or a non-realist account. It is also clear that this decision can’t be taken without an examination of perception studies (and not exclusively philosophical ones) which would be crucial for determining what is encoded in memory, and studies devoted to language, which, I suspect, plays an important role in the segmentation of reality. (c) A clarification of the notion of causality, because we wouldn’t like to deny any kind entail the complete disappearance of boundaries between memory and imagination. And even if I suspect that my future work will question the existence of absolute boundaries between these two capacities, the notion of causality can’t be completely erase from the theory of memory without erasing the memory phenomenon itself. But this causality, as we’ve already showed, can’t consist in the naïve conception adopted by the causal accounts of memory, nor can it be explained in counterfactual terms. If we think of the reconsolidation concept, of the interdependence between all the memory system, and of the deciding influence of context, it’s almost sure that more than one notion of causality will need to be invoked in order to explain memory. (d) Because I do not want to reduce the study of memory exclusively to cases of remembering, it’s essential to analyze how memory intervenes not only in imagination, but also in perception, reasoning and future projection, for example. In this respect, we could quote the interesting articles of Atance & O’Neill (2001) and Bucker & Carroll (2006), the first concerning the relationship between episodic memory and episodic future thinking, the second concerning the relationship between episodic memory and all kinds of self-projection, such as navigation and theory of mind, relations that would also have to be explored from a philosophical perspective, in particular to overcome the tendency to suppose that representation is the only function of memory. I will also mention the necessity of combining the notion of metamemory with that of memory to some extent; as I have already remarked, it is implausible to suppose that a subject who consciously remembers something is unable to exert any kind of control over his memory reports. (e) Finally, in order to integrate a situated conception of memory into a philosophical account, it will also be indispensable to take into consideration what is known as “everyday memory research” as well as studies in social and collective memory. (shrink)

Since the second half of the XXth century, researchers in cybernetics and AI, neural nets and connectionism, Artificial Life and new robotics have endeavoured to build different machines that could simulate functions of living organisms, such as adaptation and development, problem solving and learning. In this book these research programs are discussed, particularly as regard the epistemological issues of the behaviour modelling. One of the main novelty of this book consists of the fact that certain projects involving the building (...) of simulative machine models before the advent of cybernetics are been investigated for the first time, on the basis of little known, and sometimes completely forgotten or unpublished, texts and figures. These pre-cybernetics projects can be considered as steps toward the “discovery” of a modelling methodology that has been fully developed by those more recent research programs, and that shares some of their central goals and key methodological proposals. -/- More info in Springer link: http://www.springer.com/new+%26+forthcoming+titles+%28default%29/book/978-1-4020-0606-7 -/- This book is the English translation of La scoperta dell'artificiale, Dunod/Masson, Milan, 1998. (shrink)

The topic of this dissertation is the nature of linguistic competence, the capacity to understand and produce sentences of natural language. I defend the empiricist account of linguistic competence embedded in the connectionist cognitive science. This strand of cognitive science has been opposed to the traditional symbolic cognitive science, coupled with transformational-generative grammar, which was committed to nativism due to the view that human cognition, including language capacity, should be construed in terms of symbolic representations and hardwired rules. Similarly, linguistic (...) competence in this framework was regarded as being innate, rule-governed, domain-specific, and fundamentally different from performance, i.e., idiosyncrasies and factors governing linguistic behavior. I analyze state-of-the-art connectionist, deep learning models of natural language processing, most notably large language models, to see what they can tell us about linguistic competence. Deep learning is a statistical technique for the classification of patterns through which artificial intelligence researchers train artificial neural networks containing multiple layers that crunch a gargantuan amount of textual and/or visual data. I argue that these models suggest that linguistic competence should be construed as stochastic, pattern-based, and stemming from domain-general mechanisms. Moreover, I distinguish syntactic from semantic competence, and I show for each the ramifications of the endorsement of a connectionist research program as opposed to the traditional symbolic cognitive science and transformational-generative grammar. I provide a unifying front, consisting of usage-based theories, a construction grammar approach, and an embodied approach to cognition to show that the more multimodal and diverse models are in terms of architectural features and training data, the stronger the case is for the connectionist linguistic competence. I also propose to discard the competence vs. performance distinction as theoretically inferior so that a novel and integrative account of linguistic competence originating in connectionism and empiricism that I propose and defend in the dissertation could be put forward in scientific and philosophical literature. (shrink)

Among the cognitive capacities of evolved creatures is the capacity to represent. Theories in cognitive neuroscience typically explain our manifest representational capacities by positing internal representations, but there is little agreement about how these representations function, especially with the relatively recent proliferation of connectionist, dynamical, embodied, and enactive approaches to cognition. In this talk I sketch an account of the nature and function of representation in cognitive neuroscience that couples a realist construal of representational vehicles with a pragmatic account of (...) mental content. I call the resulting package a deflationary account of mental representation and I argue that it avoids the problems that afflict competing accounts. (shrink)

Book Description (Blurb): Cognitive Design for Artificial Minds explains the crucial role that human cognition research plays in the design and realization of artificial intelligence systems, illustrating the steps necessary for the design of artificial models of cognition. It bridges the gap between the theoretical, experimental and technological issues addressed in the context of AI of cognitive inspiration and computational cognitive science. -/- Beginning with an overview of the historical, methodological and technical issues in the field of Cognitively-Inspired Artificial Intelligence, (...) Lieto illustrates how the cognitive design approach has an important role to play in the development of intelligent AI technologies and plausible computational models of cognition. Introducing a unique perspective that draws upon Cybernetics and early AI principles, Lieto emphasizes the need for an equivalence between cognitive processes and implemented AI procedures, in order to realise biologically and cognitively inspired artificial minds. He also introduces the Minimal Cognitive Grid, a pragmatic method to rank the different degrees of biologically and cognitive accuracy of artificial systems in order project and predict their explanatory power with respect to the natural systems taken as source of inspiration. -/- Providing a comprehensive overview of cognitive design principles in constructing artificial minds, this text will be essential reading for students and researchers of artificial intelligence and cognitive science. (shrink)

The emerging contemporary natural philosophy provides a common ground for the integrative view of the natural, the artificial, and the human-social knowledge and practices. Learning process is central for acquiring, maintaining, and managing knowledge, both theoretical and practical. This paper explores the relationships between the present advances in understanding of learning in the sciences of the artificial, natural sciences, and philosophy. The question is, what at this stage of the development the inspiration from nature, specifically its computational models such as (...) info-computation through morphological computing, can contribute to machine learning and artificial intelligence, and how much on the other hand models and experiments in machine learning and robotics can motivate, justify, and inform research in computational cognitive science, neurosciences, and computing nature. We propose that one contribution can be understanding of the mechanisms of ‘learning to learn’, as a step towards deep learning with symbolic layer of computation/information processing in a framework linking connectionism with symbolism. As all natural systems possessing intelligence are cognitive systems, we describe the evolutionary arguments for the necessity of learning to learn for a system to reach humanlevel intelligence through evolution and development. The paper thus presents a contribution to the epistemology of the contemporary philosophy of nature. (shrink)

Many philosophers and psychologists have attempted to elucidate the nature of mental representation by appealing to notions like isomorphism or abstract structural resemblance. The ‘structural representations’ that these theorists champion are said to count as representations by virtue of functioning as internal models of distal systems. In his 2007 book, Representation Reconsidered, William Ramsey endorses the structural conception of mental representation, but uses it to develop a novel argument against representationalism, the widespread view that cognition essentially involves the manipulation of (...) mental representations. Ramsey argues that although theories within the ‘classical’ tradition of cognitive science once posited structural representations, these theories are being superseded by newer theories, within the tradition of connectionism and cognitive neuroscience, which rarely if ever appeal to structural representations. Instead, these theories seem to be explaining cognition by invoking so-called ‘receptor representations’, which, Ramsey claims, aren’t genuine representations at all—despite being called representations, these mechanisms function more as triggers or causal relays than as genuine stand-ins for distal systems. I argue that when the notions of structural and receptor representation are properly explicated, there turns out to be no distinction between them. There only appears to be a distinction between receptor and structural representations because the latter are tacitly conflated with the ‘mental models’ ostensibly involved in offline cognitive processes such as episodic memory and mental imagery. While structural representations might count as genuine representations, they aren’t distinctively mental representations, for they can be found in all sorts of non-intentional systems such as plants. Thus to explain the kinds of offline cognitive capacities that have motivated talk of mental models, we must develop richer conceptions of mental representation than those provided by the notions of structural and receptor representation. (shrink)

There are two versions of the language-of-thought hypothesis (LOT): Representational LOT (roughly, structured representation), introduced by Ockham, and computational LOT (roughly, symbolic computation) introduced by Fodor. Like many others, I oppose the latter but not the former. Quilty-Dunn et al. defend representational LOT, but they do not defend the strong computational LOT thesis central to the classical-connectionist debate.

Classical cognitivist and connectionist models posit a Cartesian disembodiment of mind assuming that brain events can adequately explain thought and related notions such as intellect. Instead, we argue for the bodily basis of thought and its continuity beyond the sensorimotor stage. Indeed, there are no eternally fixed representations of the external world in the "motor system," rather, it is under the guidance of both internal and external factors with important linkages to frontal, parietal, cerebellar, basal ganglionic, and cingulate gyrus areas (...) that subserve cognitive and motivational activities. Indeed, the motor system, including related structures, is a self-organizing dynamical system contextualized among musculoskeletal, environmental (e.g., gravity), and social forces. We do not simply inhabit our bodies; we literally use them to think with. (shrink)

The sensorimotor theory of vision and visual consciousness is often described as a radical alternative to the computational and connectionist orthodoxy in the study of visual perception. However, it is far from clear whether the theory represents a significant departure from orthodox approaches or whether it is an enrichment of it. In this study, I tackle this issue by focusing on the explanatory structure of the sensorimotor theory. I argue that the standard formulation of the theory subscribes to the same (...) theses of the dynamical hypothesis and that it affords covering-law explanations. This however exposes the theory to the mere description worry and generates a puzzle about the role of representations. I then argue that the sensorimotor theory is compatible with a mechanistic framework, and show how this can overcome the mere description worry and solve the problem of the explanatory role of representations. By doing so, it will be shown that the theory should be understood as an enrichment of the orthodoxy, rather than an alternative. (shrink)

During the last decades, many cognitive architectures (CAs) have been realized adopting different assumptions about the organization and the representation of their knowledge level. Some of them (e.g. SOAR [35]) adopt a classical symbolic approach, some (e.g. LEABRA[ 48]) are based on a purely connectionist model, while others (e.g. CLARION [59]) adopt a hybrid approach combining connectionist and symbolic representational levels. Additionally, some attempts (e.g. biSOAR) trying to extend the representational capacities of CAs by integrating diagrammatical representations and reasoning are (...) also available [34]. In this paper we propose a reflection on the role that Conceptual Spaces, a framework developed by Peter G¨ardenfors [24] more than fifteen years ago, can play in the current development of the Knowledge Level in Cognitive Systems and Architectures. In particular, we claim that Conceptual Spaces offer a lingua franca that allows to unify and generalize many aspects of the symbolic, sub-symbolic and diagrammatic approaches (by overcoming some of their typical problems) and to integrate them on a common ground. In doing so we extend and detail some of the arguments explored by G¨ardenfors [23] for defending the need of a conceptual, intermediate, representation level between the symbolic and the sub-symbolic one. In particular we focus on the advantages offered by Conceptual Spaces (w.r.t. symbolic and sub-symbolic approaches) in dealing with the problem of compositionality of representations based on typicality traits. Additionally, we argue that Conceptual Spaces could offer a unifying framework for interpreting many kinds of diagrammatic and analogical representations. As a consequence, their adoption could also favor the integration of diagrammatical representation and reasoning in CAs. (shrink)

Virtue ethics has many times been suggested as a promising recipe for the construction of artificial moral agents due to its emphasis on moral character and learning. However, given the complex nature of the theory, hardly any work has de facto attempted to implement the core tenets of virtue ethics in moral machines. The main goal of this paper is to demonstrate how virtue ethics can be taken all the way from theory to machine implementation. To achieve this goal, we (...) critically explore the possibilities and challenges for virtue ethics from a computational perspective. Drawing on previous conceptual and technical work, we outline a version of artificial virtue based on moral functionalism, connectionist bottom–up learning, and eudaimonic reward. We then describe how core features of the outlined theory can be interpreted in terms of functionality, which in turn informs the design of components necessary for virtuous cognition. Finally, we present a comprehensive framework for the technical development of artificial virtuous agents and discuss how they can be implemented in moral environments. (shrink)

A comprehensive introduction to the Language of Though Hypothesis (LOTH) accessible to general audiences. LOTH is an empirical thesis about thought and thinking. For their explication, it postulates a physically realized system of representations that have a combinatorial syntax (and semantics) such that operations on representations are causally sensitive only to the syntactic properties of representations. According to LOTH, thought is, roughly, the tokening of a representation that has a syntactic (constituent) structure with an appropriate semantics. Thinking thus consists in (...) syntactic operations defined over representations. Most of the arguments for LOTH derive their strength from their ability to explain certain empirical phenomena like productivity, systematicity of thought and thinking. (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)

Experts in Artificial Intelligence (AI) development predict that advances in the development of intelligent systems and agents will reshape vital areas in our society. Nevertheless, if such an advance is not made prudently and critically-reflexively, it can result in negative outcomes for humanity. For this reason, several researchers in the area have developed a robust, beneficial, and safe concept of AI for the preservation of humanity and the environment. Currently, several of the open problems in the field of AI research (...) arise from the difficulty of avoiding unwanted behaviors of intelligent agents and systems, and at the same time specifying what we really want such systems to do, especially when we look for the possibility of intelligent agents acting in several domains over the long term. It is of utmost importance that artificial intelligent agents have their values aligned with human values, given the fact that we cannot expect an AI to develop human moral values simply because of its intelligence, as discussed in the Orthogonality Thesis. Perhaps this difficulty comes from the way we are addressing the problem of expressing objectives, values, and ends, using representational cognitive methods. A solution to this problem would be the dynamic approach proposed by Dreyfus, whose phenomenological philosophy shows that the human experience of being-in-the-world in several aspects is not well represented by the symbolic or connectionist cognitive method, especially in regards to the question of learning values. A possible approach to this problem would be to use theoretical models such as SED (situated embodied dynamics) to address the values learning problem in AI. (shrink)

‘‘COGNITIVE ECOLOGY’’ is a fruitful model for Shakespearian studies, early modern literary and cultural history, and theatrical history more widely. Cognitive ecologies are the multidimensional contexts in which we remember, feel, think, sense, communicate, imagine, and act, often collaboratively, on the fly, and in rich ongoing interaction with our environments. Along with the anthropologist Edwin Hutchins,1 we use the term ‘‘cognitive ecology’’ to integrate a number of recent approaches to cultural cognition: we believe these approaches offer productive lines of engagement (...) with early modern literary and historical studies.2 The framework arises out of our work in extended mind and distributed cognition.3 The extended mind hypothesis arose from a post-connectionist philosophy of cognitive science. This approach was articulated in Andy Clark’s Being There: Putting Brain, Body, and World Together Again, and further developed by Susan Hurley and Mark Rowlands, among others.4 The distributed cognition approach arose independently, from work in cognitive anthropology, HCI (Human-Computer Interaction), the sociology of education and work, and science studies. The principles of distributed cognition were articulated in Hutchins’s ethnography of navigation, Cogni- tion in the Wild,5 and developed by theorists such as David Kirsh and Lucy Suchman.6 These models share an anti-individualist approach to cognition. In all these views, mental activities spread or smear across the boundaries of skull and skin to include parts of the social and material world. In remembering, decision making, and acting, whether individually or in small groups, our complex and structured activities involve many distinctive dimensions: neural, affective, kines-. (shrink)

Experts in Artificial Intelligence (AI) development predict that advances in the dvelopment of intelligent systems and agents will reshape vital areas in our society. Nevertheless, if such an advance isn't done with prudence, it can result in negative outcomes for humanity. For this reason, several researchers in the area are trying to develop a robust, beneficial, and safe concept of artificial intelligence. Currently, several of the open problems in the field of AI research arise from the difficulty of avoiding unwanted (...) behaviors of intelligent agents, and at the same time specifying what we want such systems to do. It is of utmost importance that artificial intelligent agents have their values aligned with human values, given the fact that we cannot expect an AI to develop our moral preferences simply because of its intelligence, as discussed in the Orthogonality Thesis. Perhaps this difficulty comes from the way we are addressing the problem of expressing objectives, values, and ends, using representational cognitive methods. A solution to this problem would be the dynamic cognitive approach proposed by Dreyfus, whose phenomenological philosophy defends that the human experience of being-in-the-world cannot be represented by the symbolic or connectionist cognitive methods. A possible approach to this problem would be to use theoretical models such as SED (situated embodied dynamics) to address the values learning problem in AI. (shrink)

The paper is dedicated to particular cases of interaction and mutual impact of philosophy and cognitive science. Thus, philosophical preconditions in the middle of the 20th century shaped the newly born cognitive science as mainly based on conceptual and propositional representations and syntactical inference. Further developments towards neural networks and statistical representations did not change the prejudice much: many still believe that network models must be complemented with some extra tools that would account for proper human cognitive traits. I address (...) some real implemented connectionist models that show how ‘new associationism ’ of the neural network approach may not only surpass Humean limitations, but, as well, realistically explain abstraction, inference and prediction. Then I stay on Predictive Processing theories in a little more detail to demonstrate that sophisticated statistical tools applied to a biologically realist ontology may not only provide solutions to scientific problems or integrate different cognitive paradigms but propose some philosophical insights either. To conclude, I touch on a certain parallelism of Predictive Processing and philosophical inferentialism as presented by Robert Brandom. (shrink)

This work addresses a broad range of questions which belong to four fields: computation theory, general philosophy of science, philosophy of cognitive science, and philosophy of mind. Dynamical system theory provides the framework for a unified treatment of these questions. ;The main goal of this dissertation is to propose a new view of the aims and methods of cognitive science--the dynamical approach . According to this view, the object of cognitive science is a particular set of dynamical systems, which I (...) call "cognitive systems". The goal of a cognitive study is to specify a dynamical model of a cognitive system, and then use this model to produce a detailed account of the specific cognitive abilities of that system. The dynamical approach does not limit a-priori the form of the dynamical models which cognitive science may consider. In particular, this approach is compatible with both computational and connectionist modeling, for both computational systems and connectionist networks are special types of dynamical systems. ;To substantiate these methodological claims about cognitive science, I deal first with two questions in two different fields: What is a computational system? What is a dynamical explanation of a deterministic process? ;Intuitively, a computational system is a deterministic system which evolves in discrete time steps, and which can be described in an effective way. In chapter 1, I give a formal definition of this concept which employs the notions of isomorphism between dynamical systems, and of Turing computable function. In chapter 2, I propose a more comprehensive analysis which is based on a natural generalization of the concept of Turing machine. ;The goal of chapter 3 is to develop a theory of the dynamical explanation of a deterministic process. By a "dynamical explanation" I mean the specification of a dynamical model of the system or process which we want to explain. I start from the analysis of a specific type of explanandum--dynamical phenomena--and I then use this analysis to shed light on the general form of a dynamical explanation. Finally, I analyze the structure of those theories which generate explanations of this form, namely dynamical theories. (shrink)

Can beliefs that are not consciously formulated serve as part of an agent's evidence for other beliefs? A common view says no, any belief that is psychologically immediate is also epistemically immediate. I argue that some unconscious beliefs can serve as evidence, but other unconscious beliefs cannot. Person-level beliefs can serve as evidence, but subpersonal beliefs cannot. I try to clarify the nature of the personal/subpersonal distinction and to show how my proposal illuminates various epistemological problems and provides a principled (...) framework for solving other problems. (shrink)

In contemporary literature, the fact that there is negative causation is the primary motivation for rejecting the physical connection view, and arguing for alternative accounts of causation. In this paper we insist that such a conclusion is too fast. We present two frameworks, which help the proponent of the physical connection view to resist the anti-connectionist conclusion. According to the first framework, there are positive causal claims, which co-refer with at least some negative causal claims. According to the second framework, (...) negative causal claims are generated from mapping and comparing different scenarios, which can fully be accounted for in purely positive terms. Since the positive causal claims evoked by both frameworks pose no obvious difficulties for the physical connection view, these frameworks make it possible for the connectionists to accommodate negative causal claims into their theory. Once these strategies are available, the connectionists become able to render all the arguments starting from the observation that there are negative causal claims in our causal discourse inconclusive with regard to the viability of the physical connection view. (shrink)

Ask most any cognitive scientist working today if a digital computational system could develop aesthetic sensibility and you will likely receive the optimistic reply that this remains an open empirical question. However, I attempt to show, while drawing upon the later Wittgenstein, that the correct answer is in fact available. And it is a negative a priori. It would seem, for example, that recent computational successes in generative AI and textual attribution, most notably those of Donald Foster (famed finder of (...) Ted Kazinski a.k.a. “the Unibomber”) speak favorably of the connectionist model's capacity to overcome the “aspect blindness” handicap in this domain. I argue however that such results are only achievable when rigid input‐to‐output parameters are given, and that this element is precisely what is absent in standard examples of aesthetic judgment. I thus conclude that while the connectionist model anticipated by Turing may provide the best approach for the AI project, its capacity for meeting its own sufficiency requirements is necessarily crippled by its inability to share in what can be generally referred to as the collective engagements of human solidarity. (shrink)