In Connectionism and the Philosophy of Psychology, Horgan and Tienson articulate and defend a new view of cognition.

For the last 20 years or so, philosophers of mind have been using the term ‘qualia’, which is frequently glossed as standing for the “what-it-is-like” of experience. The examples of what-it-is-like that are typically given are feelings of pain or itches, and color and sound sensations. This suggests an identification of the experiential what-it-islike with such states. More recently, philosophers have begun speaking of the “phenomenology“ of experience, which they have also glossed as “what-it-is-like”. Many say, for example, that any (...) acceptable materialism—or any acceptable account of the relation of mind and body—must “respect the phenomenology.”1 Typically, no examples beyond those mentioned in the first paragraph are offered. This suggests that the picture of the phenomenology that ”must be respected” is the what-it-is-like of bodily sensations, of sensations that occur in perception, and perhaps of certain analogous nonperceptual states, such as imaginings and image-like rememberings. According to the suggested picture, all there is to phenomenology is such states; intentional mental states—as such— have no phenomenology; there is nothing that it is like to undergo them. Although beliefs and desires are intentionally directed—i.e., they have aboutness—these mental states allegedly are not inherently phenomenal. On this view, there is nothing that it is like to be.. (shrink)

Consciousness and intentionality help to define the mental qua mental. Consciousness and intentionality, insist some philosophers, although perhaps often co‐occurring, are mutually independent or separable. Consciousness and intentionality, insist others, are interdependent or inseparable. This chapter discusses an important aspect of inseparatism: the relation between phenomenal character and intentional content. The contemporary philosophers and theorists have developed inseparatist or nearly inseparatist theses in various ways. The chapter mentions some of this work, and the philosophers responsible for it. It discusses two (...) implications of thesis C‐Ins. First, since phenomenally conscious states are mental, every phenomenally conscious state also is intentional. Second, since phenomenally intentional content is determined by phenomenal character alone, such content is entirely constituted by features internal or intrinsic to phenomenology. Finally, the chapter sketches the questions about or challenges to inseparatism and explains something, again briefly, about how the inseparatist might reply to each. (shrink)

David Marr provided an influential account of levels of description in classical cognitive science. In this paper we contrast Marr'ent with some alternatives that are suggested by the recent emergence of connectionism. Marr's account is interesting and important both because of the levels of description it distinguishes, and because of the way his presentation reflects some of the most basic, foundational, assumptions of classical AI-style cognitive science. Thus, by focusing on levels of description, one can sharpen foundational differences between classicism (...) and potential non-classical conceptions of mentality that might emerge under the rubric of connectionism. (shrink)

Contemporary cognitive science has two principal branches: the classical computational approach (sometimes called classicism) and connectionism. Rules are fundamental to theorizing about the basic units of processing in both classicism and connectionism. But we will be concerned primarily with rules that apply to representations and that determine transitions from one cognitive/representational state to the next. Such rules are fundamental to classicism, since, according to classicism, cognitive processes simply are rule‐governed cognitive state transitions. Rules that apply to representations are not a (...) built‐in feature of connectionist architecture, so the role of such rules in connectionism is less clear. We will argue that they are neither necessary nor (in general) desirable in connectionism. We discuss the status of rules first in classicism, then in connectionism. (shrink)

Connectionism was explicitly put forward as an alternative to classical cognitive science. The questions arise: how exactly does connectionism differ from classical cognitive science, and how is it potentially better? The classical “rules and representations” conception of cognition is that cognitive transitions are determined by exceptionless rules that apply to the syntactic structure of symbols. Many philosophers have seen connectionism as a basis for denying structured symbols. We, on the other hand, argue that cognition is too rich and flexible to (...) be simulable by the exceptionless representation-level rules that classicism requires. However, this very richness of cognition requires syntactically structured representations—what philosophers call a language of thought. The natural mathematical characterization of neural networks comes from the theory of dynamical systems. We propose that the mathematics of dynamical systems, not the mathematics of algorithms, holds the key to how cognitive structure and cognitive processes can be realized in the physical structure and processes of a network. (shrink)

David Marr provided an influential account of levels of description in classical cognitive science. In this paper we contrast Marr'ent with some alternatives that are suggested by the recent emergence of connectionism. Marr's account is interesting and important both because of the levels of description it distinguishes, and because of the way his presentation reflects some of the most basic, foundational, assumptions of classical AI-style cognitive science . Thus, by focusing on levels of description, one can sharpen foundational differences between (...) classicism and potential non-classical conceptions of mentality that might emerge under the rubric of connectionism. (shrink)

What van Gelder calls the dynamical hypothesis is only a special case of what we here dub the general dynamical hypothesis. His terminology makes it easy to overlook important alternative dynamical approaches in cognitive science. Connectionist models typically conform to the general dynamical hypothesis, but not to van Gelder's.

"A third of the papers in this volume originated at the 1987 Spindel Conference ... at Memphis State University"--Pref.

In the first post World War II identity theories (e.g., Place 1956, Smart 1962), mind brain identities were held to be contingent. However, in work beginning in the late 1960's, Saul Kripke (1971, 1980) convinced the philosophical community that true identity statements involving names and natural kind terms are necessarily true and furthermore, that many such necessary identities can only be known a posteriori. Kripke also offered an explanation of the a posteriori nature of ordinary theoretical identities such as that (...) water = H2O. We identify the kinds and substances involved in theoretical identities by certain of their contingent properties. What we discover when we discover a theoretical identity is the underlying nature of the kind that we identify by those contingent properties. Now, of course, it was being a posteriori, not being contingent, that mattered to the identity theorists anyway, so the necessity of identity is not, in itself, damaging to mind brain identity theories. However, Kripke also argued persuasively that the alleged mind brain identities could not be treated in the same way as ordinary theoretical identities. We "identify" pain by feeling it, and surely how it feels is an essential property of pain, not a contingent property. Thus, a mind body identity theory must provide a different explanation of why its identities are a posteriori. A new wave of materialists has appeared on the scene with a new strategy for explaining [1] the a posteriori nature of its alleged identities. The strategy is to locate the explanation for the a posteriori nature of mind body identities, not on the side of the world, but on the side of the mind -in different ways of thinking about or imagining, or in different concepts. Thus, on this new view, there is only one property—this brain process type, which is identical with this pain.. (shrink)

David Marr provided a useful framework for theorizing about cognition within classical, AI-style cognitive science, in terms of three levels of description: the levels of (i) cognitive function, (ii) algorithm and (iii) physical implementation. We generalize this framework: (i) cognitive state transitions, (ii) mathematical/functional design and (iii) physical implementation or realization. Specifying the middle, design level to be the theory of dynamical systems yields a nonclassical, alternative framework that suits (but is not committed to) connectionism. We consider how a brain's (...) (or a network's) being a dynamical system might be the key both to its realizing various essential features of cognition — productivity, systematicity, structure-sensitive processing, syntax — and also to a non-classical solution of (frame-type) problems plaguing classical cognitive science. (shrink)

Human cognition is rich, varied, and complex. In this Chapter we argue that because of the richness of human cognition (and human mental life generally), there must be a syntax of cognitive states, but because of this very richness, cognitive processes cannot be describable by exceptionless rules. The argument for syntax, in Section 1, has to do with being able to get around in any number of possible environments in a complex world. Since nature did not know where in the (...) world humans would find themselves—nor within pretty broad limits what the world would be like —nature had to provide them with a means of “representing” a great deal of information about any of indefinitely many locations. We see no way that this could be done except by way of syntax— that is, by a systematic way of producing new, appropriate representations as needed. We discuss what being systematic must amount to, and what, in consequence, syntax should mean. We hold that syntax does not require a part/whole relationship. The argument for the claim that human cognitive processes cannot be described by exceptionless rules, in Section 2, appeals to the fact that there is no limit to the factors one might.. (shrink)

Terence Horgan, George Graham and John Tienson argue that some intentional content is constitutively determined by phenomenology alone. We argue that this would require a certain kind of covariation of phenomenal states and intentional states that is not established by Horgan, Tienson and Graham’s arguments. We make the case that there is inadequate reason to think phenomenology determines perceptual belief, and that there is reason to doubt that phenomenology determines any species of non-perceptual intentionality. We also (...) raise worries about the capacity of phenomenology to map onto intentionality in a way that would be appropriate for any determiner of content/fixer of truth conditions. (shrink)

We have a familiar idea of levels of description or levels of theory in science: microphysics, atomic physics, chemistry, biology, psychology, and the various social sciences. It is clear that philosophers - such as Terry Horgan - who want to be nonreductive materialists with regard to the mental must hold that this is not mere description; there must be genuine higher-level causes, and hence, genuine higher-level properties, in particular mental properties and causes. But there appears to be a deep (...) problem concerning mental causes. The (micro-) physical world is causally closed. Mental states are - or depend on or are realized by - physical states. It seems, then, that the physical state on which a mental state depends will be responsible for any alleged effects of the mental state. There will be no room for mental causation. And if properties exist insofar at they have a causal role, there will be no room for mental properties either. Many philosophers - Horgan included - have seen this problem of the "causalexclusion" ofthe mental as a specialcase of a general problem:the exclusion of higher-level causes by the causal closure of microphysics. Suppose one higher-level state, H1 leads to another higher-level state, H2. H1 is realized by some base level state, B1, which leads to a base-level state, B2, which in turn realizes H2. All of the casual work, so to speak, takes place at the base level. There is no room for any genuine causal connection between H1, as such, and H2, as such. I argue that there is no problem about higher-order causation in general. There are genuine, unsurprising higher-level causes and properties. A ball roles, for example, or breaks a window. If there is a problem of exclusion regarding putative mental causes, it is not an instance of a general exclusion problem, but is sui generis, and mental causation remains mysterious. (shrink)

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

Terence Horgan and John Tienson claim that folk psychological laws are different in kind from basic physical laws in at least two ways: first, physical laws do not possess the kind of ceteris paribus qualifications possessed by folk psychological laws, which means the two types of laws have different logical forms; and second, applied physical laws are best thought of as being about an idealized world and folk psychological laws about the actual world. I argue that Horgan (...) and Tienson have not made a persuasive case for either of the preceding views. (shrink)

A review of Terry Horgan and John Tienson's book.

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)