The last 5 years have seen a series of remarkable achievements in deep-neural-network-based artificial intelligence research, and some modellers have argued that their performance compares favourably to human cognition. Critics, however, have argued that processing in deep neural networks is unlike human cognition for four reasons: they are (i) data-hungry, (ii) brittle, and (iii) inscrutable black boxes that merely (iv) reward-hack rather than learn real solutions to problems. This article rebuts these criticisms by exposing comparative bias within them, in the (...) process extracting some more general lessons that may also be useful for future debates. (shrink)

Though it''s difficult to agree on the exact date of their union, logic and artificial intelligence (AI) were married by the late 1950s, and, at least during their honeymoon, were happily united. What connubial permutation do logic and AI find themselves in now? Are they still (happily) married? Are they divorced? Or are they only separated, both still keeping alive the promise of a future in which the old magic is rekindled? This paper is an attempt to answer these questions (...) via a review of six books. Encapsulated, our answer is that (i) logic and AI, despite tabloidish reports to the contrary, still enjoy matrimonial bliss, and (ii) only their future robotic offspring (as opposed to the children of connectionist AI) will mark real progress in the attempt to understand cognition. (shrink)

The dominant scientific and philosophical view of the mind – according to which, put starkly, cognition is computation – is refuted herein, via specification and defense of the following new argument: Computation is reversible; cognition isn't; ergo, cognition isn't computation. After presenting a sustained dialectic arising from this defense, we conclude with a brief preview of the view we would put in place of the cognition-is-computation doctrine.

Humans grasp discrete infinities within several cognitive domains, such as in language, thought, social cognition and tool-making. It is sometimes suggested that any such generative ability is based on a computational system processing hierarchical and recursive mental representations. One view concerning such generativity has been that each of the mind’s modules defining a cognitive domain implements its own recursive computational system. In this paper recent evidence to the contrary is reviewed and it is proposed that there is only one supramodal (...) computational system with recursion in the human mind. A recursion thesis is defined, according to which the hominin cognitive evolution is constituted by a recent punctuated genetic mutation that installed the general, supramodal capacity for recursion into the human nervous system on top of the existing, evolutionarily older cognitive structures, and it is argued on the basis of empirical evidence and theoretical considerations that the recursion thesis constitutes a plausible research program for cognitive science. (shrink)

In this paper we argue that the insistence by Fodor et. al. that the Language of Thought hypothesis must be true rests on mistakes about the kinds of explanations that must be provided of cognitive phenomena. After examining the canonical arguments for the LOT, we identify a weak version of the LOT hypothesis which we think accounts for some of the intuitions that there must be a LOT. We then consider what kinds of explanation cognitive phenomena require, and conclude that (...) three main confusions lead to the invalid inference of the truth of a stronger LOT hypothesis from the weak and trivial version. These confusions concern the relationship between syntax and semantics, the nature of higher-level causation in cognitive science, and differing roles of explanations invoking intrinsic structures of minds on the one hand, and aetiological or evolutionary accounts of their properties on the other. (shrink)

The nature of complex concepts has important implications for the computational modelling of the mind, as well as for the cognitive science of concepts. This paper outlines the way in which RVC – a Relational View of Concepts – accommodates a range of complex concepts, cases which have been argued to be non-compositional. RVC attempts to integrate a number of psychological, linguistic and psycholinguistic considerations with the situation-theoretic view that information-carrying relations hold only relative to background situations. The central tenet (...) of RVC is that the content of concepts varies systematically with perspective. The analysis of complex concepts indicates that compositionality too should be considered to be sensitive to perspective. Such a view accords with concepts and mental states being situated and the implications for theories of concepts and for computational models of the mind are discussed. (shrink)

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)

Advanced AI systems are rapidly making their way into medical research and practice, and, arguably, it is only a matter of time before they will surpass human practitioners in terms of accuracy, reliability, and knowledge. If this is true, practitioners will have a prima facie epistemic and professional obligation to align their medical verdicts with those of advanced AI systems. However, in light of their complexity, these AI systems will often function as black boxes: the details of their contents, calculations, (...) and procedures cannot be meaningfully understood by human practitioners. When AI systems reach this level of complexity, we can also speak of black-box medicine. In this paper, we want to argue that black-box medicine conflicts with core ideals of patient-centered medicine. In particular, we claim, black-box medicine is not conducive for supporting informed decision-making based on shared information, shared deliberation, and shared mind between practitioner and patient. (shrink)

Interpreting the Miyashita et al. experiments in terms of a cellassembly representation does not adequately explain the performance of Miyashita's monkeys on novel stimuli. We will argue that the latter observations point to acompositionalrepresentation and suggest a dynamics involving rapid and reversible binding of distinct activity patterns.

Recently some philosophers have urged that connectionist artificial intelligence is (potentially) eliminative for the propositional attitudes of folk psychology. At the same time, however, these philosophers have also insisted that since philosophy of science has failed to provide criteria distinguishing ontologically retentive from eliminative theory changes, the resulting eliminativism is not principled. Application of some resources developed within the semantic view of scientific theories, particularly recent formal work on the theory reduction relation, reveals these philosophers to be wrong in this (...) second contention, yet by and large correct in the first. (shrink)

Müller misconstrues autonomy to mean strict locality of brain function, something quite different from the functional autonomy that linguists claim. Similarly, he misperceives the interaction of learned and innate components hypothesized in current generative models. Evidence from sign languages, Creole languages, and neurological studies of rare forms of aphasia also argues against his conclusions.

The notion of a ‘symbol’ plays an important role in the disciplines of Philosophy, Psychology, Computer Science, and Cognitive Science. However, there is comparatively little agreement on how this notion is to be understood, either between disciplines, or even within particular disciplines. This paper does not attempt to defend some putatively ‘correct’ version of the concept of a ‘symbol.’ Rather, some terminological conventions are suggested, some constraints are proposed and a taxonomy of the kinds of issue that give rise to (...) disagreement is articulated. The goal here is to provide something like a ‘geography’ of the various notions of ‘symbol’ that have appeared in the various literatures, so as to highlight the key issues and to permit the focusing of attention upon the important dimensions. In particular, the relationship between ‘tokens’ and ‘symbols’ is addressed. The issue of designation is discussed in some detail. The distinction between simple and complex symbols is clarified and an apparently necessary condition for a system to be potentially symbol, or token bearing, is introduced. (shrink)

Connectionist research first emerged in the 1940s. The first phase of connectionism attracted a certain amount of media attention, but scant philosophical interest. The phase came to an abrupt halt, due to the efforts of Minsky and Papert (1969), when they argued for the intrinsic limitations of the approach. In the mid-1980s connectionism saw a resurgence. This marked the beginning of the second phase of connectionist research. This phase did attract considerable philosophical attention. It was of philosophical interest, as it (...) offered a way of counteracting the conceptual ties to the philosophical traditions of atomism, rationalism, logic, nativism, rule realism and a concern with the role symbols play in human cognitive functioning, which was prevalent as a consequence of artificial intelligence research. The surge in philosophical interest waned, possibly in part due to the efforts of some traditionalists and the so-called black box problem. Most recently, what may be thought of as a third phase of connectionist research, based on so-called deep learning methods, is beginning to show some signs of again exciting philosophical interest. (shrink)

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

For most philosophers of psychology and cognitive science, inquiry into human cognitive activity begins at the level of intrapersonal processes. A central question is whether these processes are sufficiently autonomous from more basic neurophysiological processes to be investigated in their own terms, or whether all explanations must be in neurophysiological terms. Some philosophers have insisted on the relative autonomy of the cognitive level. One currently quite popular view, eliminative materialism, however, holds that the explanations that have been advanced at the (...) intrapersonal cognitive level are misguided, and must be replaced by accounts developed from an understanding of the underlying neurophysiological processes. (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.

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)

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)

This paper provides a glimpse of some different theoretical frameworks and empirical methods in the author's search for theories and practices that might improve the utility and usability of computer artifacts. The essay touches on some problematic aspects of currently accepted theories and techniques in the cognitive sciences, especially in their application to the field of human-computer interaction, and mentions some alternative conceptions based on a cultural-historical approach. The intent is to widen the nature of the debate about appropriate frameworks (...) for discussing human activities especially when we discuss design activities in the context of computer systems development. The paper concludes with some suggestions for more fruitful research directions that involve the active participation of those for whom the research is ostensibly being done, and a greater emphasis on understanding how artifact design and use are inextricably intertwined. (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)

The traditional model of human cognition (TMHC) postulates an ontological and/or structural gap between conscious and unconscious mental representations. By and large, it sees higher-level mental processes as commonly conceptual or symbolic in nature and therefore conscious, whereas unconscious, lower-level representations are conceived as non-conceptual or sub-symbolic. However, experimental evidence belies this model, suggesting that higher-level mental processes can be, and often are, carried out in a wholly unconscious way and/or without conceptual representations, and that these can be processed unconsciously. (...) This entails that the TMHC, as well as the theories on mental representation it motivates and that in turn support it, is wrong. (shrink)

p style="text-indent: 0cm; line-height: normal" class="MsoBodyTextIndent3"span style="font-size: 11pt"In this paper, I argue that starting with the organelles that constitute a cellmdash;and continuing up the hierarchy of components in processes and subsystems of an organismmdash;there exist clear instances of emergent biological phenomena that can be considered ldquo;livingrdquo; entities.spannbsp; /spanThese components and their attending processes are living emergent phenomena because of the way in which the components are organized to maintain homeostasis of the organism at the various levels in the hierarchy.spannbsp; /spanI (...) call this view the emhomeostatic organization view/em of biological phenomena, and, as is shown, it comports well with the standard philosophical accounts of nomological emergence and representational emergence.spannbsp; /spanTo proffer HOV, I describe properties of biological entities that include internal-hierarchical data exchange, data selectivity, informational integration, and environmental-organismic information exchange.spannbsp; /spanFurther, a distinction is drawn between particularized homeostasis and generalized homeostasis, and I argue that because the various processes and subsystems of an organism are functioning properly in their internal environments, the organism is able to exist as a hierarchically-organized entity in some environment external to it.spannbsp; /spanStated simply: that components of biological phenomena are organized to perform some function resulting in homeostasis marks them out to be living emergent entities distinguishable, in description and in reality, from the very physico-chemical processes of which they are composed./span/p p style="text-indent: 0cm; line-height: normal" class="MsoBodyTextIndent3"span style="font-size: 11pt"nbsp;/span/p. (shrink)

Block explains the conflation of phenomenal consciousness and access consciousness by appeal to the ambiguity of the term “consciousness.” However, the nature of ambiguity is not at all clear, and the thesis that “consciousness” is ambiguous between phenomenal consciousness and access consciousness is far from obvious. Moreover, the conflation can be explained without supposing that the term is ambiguous. Block's argument can thus be strengthened by avoiding controversial issues in the semantics of “consciousness.”.

Fodor and Pylyshyn (1988) have argued that the cognitive architecture is not Connectionist. Their argument takes the following form: (1) the cognitive architecture is Classical; (2) Classicalism and Connectionism are incompatible; (3) therefore the cognitive architecture is not Connectionist. In this essay I argue that Fodor and Pylyshyn's defenses of (1) and (2) are inadequate. Their argument for (1), based on their claim that Classicalism best explains the systematicity of cognitive capacities, is an invalid instance of inference to the best (...) explanation. And their argument for (2) turns out to be question-begging. The upshot is that, while Fodor and Pylyshyn have presented Connectionists with the important empirical challenge of explaining systematicity, they have failed to provide sufficient reason for inferring that the cognitive architecture is Classical and not Connectionist. (shrink)

The appropriate methodology for psychological research depends on whether one is studying mental algorithms or their implementation. Mental algorithms are abstract specifications of the steps taken by procedures that run in the mind. Implementational issues concern the speed and reliability of these procedures. The algorithmic level can be explored only by studying across-task variation. This contrasts with psychology's dominant methodology of looking for within-task generalities, which is appropriate only for studying implementational issues.The implementation-algorithm distinction is related to a number of (...) other “levels” considered in cognitive science. Its realization in Anderson's ACT theory of cognition is discussed. Research at the algorithmic level is more promising because it is hard to make further fundamental scientific progress at the implementational level with the methodologies available. Protocol data, which are appropriate only for algorithm-level theories, provide a richer source than data at the implementational level. Research at the algorithmic level will also yield more insight into fundamental properties of human knowledge because it is the level at which significant learning transitions are defined.The best way to study the algorithmic level is to look for differential learning outcomes in pedagogical experiments that manipulate instructional experience. This provides control and prediction in realistically complex learning situations. The intelligent tutoring paradigm provides a particularly fruitful way to implement such experiments.The implications of this analysis for the issue of modularity of mind, the status of language, research on human/computer interaction, and connectionist models are also examined. (shrink)