Andrew Clark

1. Throughout the paper, and especially in the section called "LISP vs. DST", I worried that there was not enough focus on EXPLANATION. For the real question, it seems to me, is not whether some dynamical system can implement human cognition, but whether the dynamical description of the system is more explanatorily potent than a computational/representational one. Thus we know, for example, that a purely physical specification can fix a system capable of computing any LISP function. But from this it doesn't follow that the physical description is the one we need to understand the power of the system considered as an information processing device. In the same way, I don't think your demonstration that bifurcating attractor sets can yield the same behavior as a LISP program goes any way towards showing that we should not PREFER the LISP description. To reduce symbolic stories to a subset of DST (as hinted in that section) requires MORE than showing this kind of equivalence: it requires showing that there is explanatory gain, or at the very least, no explanatory loss, at that level. I append an extract from a recent paper of mine that touches on these issues, in case it helps clarify what I am after here

Of course, words aren’t magic. Neither are sextants, compasses, maps, slide rules and all the other paraphenelia which have accreted around the basic biological brains of homo sapiens. In the case of these other tools and props, however, it is transparently clear that they function so as to either carry out or to facilitate computational operations important to various human projects. The slide rule transforms complex mathematical problems (ones that would baffle or tax the unaided subject) into simple tasks of perceptual recognition. The map provides geographical information in a format well-suited to aid complex planning and strategic military operations. The compass gathers and displays a kind of information that (most) unaided human subjects do not seem to command. These various tools and props thus act to generate information, or to store it, or to transform it, or some combination of the three. In so doing, they impact our individual and collective problem- solving capacities in much the same dramatic ways as various software packages impact the performance of a simple pc.

There is a definite challenge in the air regarding the pivotal notion of internal representation. This challenge is explicit in, e.g., van Gelder, 1995; Beer, 1995; Thelen & Smith, 1994; Wheeler, 1994; and elsewhere. We think it is a challenge that can be met and that (importantly) can be met by arguing from within a general framework that accepts many of the basic premises of the work (in new robotics and in dynamical systems theory) that motivates such scepticism in the first place. Our strategy will be as follows. We begin (Section 1) by offering an account (an example and something close to a definition) of what we shall term Minimal Robust Representationalism (MRR). Sections 2 & 3 address some likely worries and questions about this notion. We end (Section 4) by making explicit the conditions under which, on our account, a science (e.g., robot- ics) may claim to be addressing cognitive phenomena

Kleinberg (1999) describes a novel procedure for efficient search in a dense hyper-linked environment, such as the world wide web. The procedure exploits information implicit in the links between pages so as to identify patterns of connectivity indicative of “authorative sources”. At a more general level, the trick is to use this second-order link-structure information to rapidly and cheaply identify the knowledge- structures most likely to be relevant given a specific input. I shall argue that Kleinberg’s procedure is suggestive of a new, viable, and neuroscientifically plausible solution to at least (one incarnation of) the so-called “Frame Problem” in cognitive science viz the problem of explaining global abductive inference. More accurately, I shall argue that.

Recent work in "embodied, embedded" cognitive science links mental contents to large-scale distributed effects: dynamic patterns implicating elements of (what are traditionally seen as) sensing, reasoning and acting. Central to this approach is an idea of biological cognition as profoundly "action-oriented" - geared not to the creation of rich, passive inner models of the world, but to the cheap and efficient production of real-world action in real-world context. A case in point is Hurley's (1998) account of the profound role of motor output in fixing the contents of conscious visual awareness - an account that also emphasizes distributed vehicles and long-range dynamical loops. Such stories can seem dramatically opposed to accounts, such as Milner and Goodale (1995), that stress relatively local mechanisms and that posit firm divisions between processes of visual awareness and of visuomotor action. But such accounts, I argue, can be deeply complimentary and together illustrate an important lesson. The lesson is that cognition may be embodied and action-oriented in two distinct - but complimentary - ways. There is a way of being embodied and action-oriented that implies being closely geared to the fine-grained control of low level effectors (hands, arms, legs and so on). And there is a way of being embodied and action-oriented that implies being closely geared to gross motor intentions, current goals, and schematic motor plans. Human cognition, I suggest, is embodied and action-oriented in both these ways. But the neural systems involved, and the size and scope of the key dynamic loops, may be quite different in each case

In _Action in Perception _Alva No develops and presents a sensorimotor account of vision and of visual consciousness. According to such an account seeing (and indeed perceiving more generally) is analysed as a kind of skilful bodily activity. Such a view is consistent with the emerging emphasis, in both philosophy and cognitive science, on the critical role of embodiment in the construction of intelligent agency. I shall argue, however, that the full sensorimotor model faces three important challenges. The first is to negotiate a path between two prima facie unsatisfactory readings of the central claim that conscious perceptual experience is constituted by knowledge of patterns of sensorimotor dependence. The second is to convince us that the sensorimotor contribution, in such cases, is actually constitutive of perceptual experience rather than merely causally implicated in the origination of such experience.² And the third is to respond to the important challenge raised by what I will dub 'sensorimotor summarizing' models of the relation between conscious experience and richly detailed sensorimotor routines. According to such models³ conscious perceptual experience only rather indirectly reflects the rich detail of our actual sensorimotor engagements, which are instead lightly sampled as a coarse guide, optimized for planning and reasoning, and geared and filtered according to current needs and purposes.

[opening paragraph]: Nicholas Humphrey offers a refreshingly progressive recipe for laying wide the doors of sensation: for understanding the peculiar features of qualitative or sensational experience in terms of the physical or functional facts about brains, bodies and environments. The key move in the treatment is the promotion of a kind of co- ordinated, double-sided tweaking: a careful restatement, with some amendments, of each side of the elusive identity statement ‘sensational property x = brain state y'. Only after such restatements herd the two kinds of facts into a roughly common arena, Humphrey believes, can some kind of identity story be revealed as coherent, plausible, and explanatorily potent

[Andy Clark] What is the relation between perceptual experience and the suite of sensorimotor skills that enable us to act in the very world we perceive? The relation, according to 'sensorimotor models' (O'Regan and Noë 2001, Noë 2004) is tight indeed. Perceptual experience, on these accounts, is enacted via skilled sensorimotor activity, and gains its content and character courtesy of our knowledge of the relations between (typically) movement and sensory stimulation. I shall argue that this formulation is too extreme, and that it fails to accommodate the substantial firewalls, dis-integrations, and special-purpose streamings that form the massed strata of human cognition. In particular, such strong sensorimotor models threaten to obscure the computationally potent insensitivity of key information-processing events to the full subtleties of embodied cycles of sensing and moving. /// [Naomi Eilan] The strong sensorimotor account of perception gives self-induced movements two constitutive roles in explaining visual consciousness. The first says that self-induced movements are vehicles of visual awareness, and for this reason consciousness 'does not happen in the brain only'. The second says that the phenomenal nature of visual experiences is consists in the action-directing content of vision. In response I suggest, first, that the sense in which visual awareness is active should be explained by appeal to the role of attention in visual consciousness, rather than self-induced movements; and second, that the sense in which perceptual consciousness does not happen in the brain only should be explained by appeal to the relational nature of perceptual consciousness, appeal to which also shows why links with action cannot exhaust phenomenal content

Some regularities enjoy only an attenuated existence in a body of training data. These are regularities whose statistical visibility depends on some systematic recoding of the data. The space of possible recodings is, however, infinitely large type-2 problems. they are standardly solved! This presents a puzzle. How, given the statistical intractability of these type-2 cases, does nature turn the trick? One answer, which we do not pursue, is to suppose that evolution gifts us with exactly the right set of recoding biases so as to reduce specific type-2 problems to (tractable) type-1 mappings. Such a heavy-duty nativism is no doubt sometimes plausible. But we believe there are other, more general mechanisms also at work. Such mechanisms provide general (not task-specific) strategies for managing problems of type-2 complexity. Several such mechanisms are investigated. At the heart of each is a fundamental ploy representational redescription language and culture – may themselves be viewed as adaptations enabling this representation/computation trade-off to be pursued on an even grander scale

Strange inversions occur when things work in ways that turn received wisdom upside down. Hume offered a strangely inverted story about causation, and Darwin, about apparent design. Dennett suggests that a strange inversion also occurs when we project our own reactive complexes outward, painting our world with elusive properties like cuteness, sweetness, blueness, sexiness, funniness, and more. Such properties strike us as experiential causes, but they are really effects—a kind of shorthand for whole sets of reactive dispositions rooted in the nuts and bolts of human information processing. Understanding the nature and origins of that strange inversion, Dennett believes, is thus key to understanding the nature and origins of human experience itself. This paper examines this claim, paying special attention to recent formulations that link that strange inversion to the emerging vision of the brain as a Bayesian estimator, constantly seeking to predict the unfolding sensory barrage.

The meta-problem of consciousness (Chalmers, 2018) is the problem of explaining the behaviours and verbal reports that we associate with the so-called 'hard problem of consciousness'. These may include reports of puzzlement, of the attractiveness of dualism, of explanatory gaps, and the like. We present and defend a solution to the meta-problem. Our solution takes as its starting point the emerging picture of the brain as a hierarchical inference engine. We show why such a device, operating under familiar forms of adaptive pressure, may come to represent some of its mid-level inferences as especially certain. These mid-level states confidently re-code raw sensory stimulation in ways that (they are able to realize) fall short of fully determining how properties and states of affairs are arranged in the distal world. This drives a wedge between experience and the world. Advanced agents then represent these mid-level inferences as irreducibly special, becoming increasingly puzzled as a result.