Centre for Cognitive Science (COGS)

Spring term 2009

  • Week 1 (January 13th) Dr. Barbara Webb (Institute for Perception, Action, and Behaviour, Edinburgh): Multimodal Control and Learning in Insects and Robots
  • Week 2 (January 20th) Dr. Andrew Philippides (Centre for Computational Neuroscience and Robotics, Sussex): Insect Visual Homing: Where Next for the Snapshot Model? - postponed to 10 March
  • Week 3 (January 27th) Dr. Mark Sprevak (King's College, Cambridge): Inference to the Hypothesis of Extended Cognition - Video available
  • Week 4 (February 3rd) Prof. Owen Holland (Computer Science & Electronic Engineering, Essex): Will a Conscious Machine Need a Simulation Engine?
  • Week 5 (February 10th) Prof. Zoltan Dienes, (Psychology, Sussex): Do Amnesics Learn Quickly and Happy People Slowly?
  • Week 6 (February 17th) Dr. Robert Clowes (Informatics, Sussex): On Shallow and Deep Interpretation: The Cognitive Role of Language
  • Week 7 (February 24th) Dr. Ron Chrisley (Informatics, Sussex): Phenomenal Directness and the Extended Mind: Why Your Beliefs Are Not in Your iPhone
  • Week 8 (March 3rd)Dr. Mike Wheeler (Philosophy, Stirling): Dwellers on the Threshold: Intelligent Architecture and the Architecture of Intelligence
  • Week 9 (March 10th) Dr. Andrew Philippides (Centre for Computational Neuroscience and Robotics, Sussex): Insect Visual Homing: Where Next for the Snapshot Model?
  • Week 10 (March 17th) Professor Margaret Boden (Informatics, Sussex):Collingwood and Computer Art


Week 1

Date: Tuesday 13th January 2009
Speaker:
 Dr. Barbara Webb (Institute for Perception, Action, and Behaviour, Edinburgh)

Multimodal Control and Learning in Insects and Robots

The robust behaviours, specialised sensors and small brains of insects have been a source of inspiration for efficient processing algorithms for sensorimotor control in robots. But insects are not just reactive systems. They are capable of much more complex behaviours - including learning, integration of multimodal cues, real-world navigation and flexible behavioural choice - than any existing autonomous robots. I will discuss our recent work in this area, including both experimental and modelling results on the interaction of olfaction and vision, place memory, and the learning of re-afference.


Week 2

Date: Tuesday, 20th January 2009
Speaker:
 Dr. Andrew Philippides (Centre for Computational Neuroscience and Robotics, Sussex)

Insect Visual Homing: Where Next for the Snapshot Model?

postponed to 10 March


Week 3

Date: Tuesday, 27th January 2009
Speaker:
 Dr. Mark Sprevak (King's College, Cambridge)

Inference to the Hypothesis of Extended Cognition

This paper considers the main arguments for and against the hypothesis of extended cognition (HEC). HEC says that our cognitive processes can, and often do, extend outside our heads to include objects in the environment like notebooks and computers. Arguments for and against HEC often take the form of inferences to the best explanation (IBEs): the truth or falsity of HEC is justified by its explanatory value to cognitive science. Advocates and critics of HEC have tended to assume that IBE must support either the truth or falsity of HEC. The truth value of HEC should be decided by whether HEC makes a positive or negative explanatory contribution to cognitive science. Little consideration has been given to a third option: that this form of explanatory-value reasoning may not even be a warranted form of inference in this case. In this paper, I argue that, on close analysis, use of IBE is not valid in these contexts, and that if HEC is to supported or criticised, it should be done in other ways.


Week 4

Date: Tuesday, 3rd February 2009
Speaker:
 Prof. Owen Holland (Computer Science & Electronic Engineering, Essex)

Will a Conscious Machine Need a Simulation Engine?

The point of departure of this talk is Kenneth Craik's observation in 'The Nature of Explanation' (1943): "If the organism carries a small-scale model of external reality and of its own possible actions within its head, it is able to try out various alternatives, conclude which is the best of them, react to future situations before they arise, utilize the knowledge of past events in dealing with the present and future, and in every way to react in a much fuller, safer, and more competent manner to the emergencies which face it." After reviewing some work in robotics that has attempted to use this framework as a basis for generating intelligent action, I will examine some possible links between various versions of the framework and some cognitive and phenomenal aspects of consciousness, particularly those involving the bodily self. I will then argue that, although a capacity for modelling or simulation may be desirable for an intelligent machine, it may be necessary for a conscious machine, and that a sophisticated version of Craik's scheme may even be sufficient.


Week 5

Date: Tuesday, 10th February 2009
Speaker:
 Prof. Zoltan Dienes (Psychology, Sussex)

Do Amnesics Learn Quickly and Happy People Slowly?

Implicit learning is an important process by which we learn about the structure of our environment so we can interact with it effectively. Much of implicit learning consists of associative learning. I develop a paradigm in order to measure the speed of implicit associative learning in a simple way, the simplest way I could. I explore two factors that theoretically should affect learning rate. I argue amnesia should, paradoxically, increase learning rate and provide evidence that it does. I argue that transient small changes in mood while performing a task should also affect learning rate (mood as information hypothesis), such that happy moods make one learn more slowly and sad moods increase learning rate. Thus far I have only tentative evidence for the latter claim, which will be presented.


Week 6

Date: Tuesday, 17th February 2009
Speaker:
 Dr. Robert Clowes (Informatics, Sussex)

On Shallow and Deep Interpretation: The Cognitive Role of Language

Over the last several years the philosopher Andy Clark (1998; 2006) is one of several -- Dennett (1991) and Carruthers (1996; 2002) are others -- who, against previous cognitive science orthodoxy, have argued that human language plays a central role in human thinking. Clark's ecological account of the cognitive role of language hinges on language playing a role in constraining, structuring and scaffolding non-linguistic cognition, rather than mirroring it. Clark's view hangs together well with current embodied and ecological approaches to mind.

Clark's account seems to hinge on what I have called shallow interpretation (Clowes, 2007): that is, words are used as ecological pivots around which other cognitive abilities turn. One of the interesting things about this account is that it posits a special role for language which appears contrary to other 'embodied' theories of thinking, specifically those posited by some cognitive linguists (Fauconnier & Turner, 2002; Lakoff & Johnson, 1999). The cognitive linguists see distinctly human abilities depending on what might be called 'deep interpretation': the use of unconscious metaphor systems facilitated in part by language. This talk will consider the tensions between these different approaches and how an encounter between them may shed new light on the explanation of our distinctly human cognitive abilities.

References

  • Carruthers, P. (1996). Language thought and consciousness: an essay in philosophical psychology. Cambridge: Cambridge University Press.
  • Carruthers, P. (2002). The cognitive function of language. Behavioral and Brain Sciences, 25(06), 657-674.
  • Clark, A. (1998). Magic Words: How Language Augments Human Computation. In P. Carruthers & J. Boucher (Eds.), Language and Thought. Interdisciplinary Themes (pp. 162 - 183). Oxford: Oxford University Press.
  • Clark, A. (2006). Material Symbols. Philosophical Psychology, 19(3), 291-307.
  • Clowes, R. W. (2007). The complex vehicles of human thought and the role of scaffolding, internalisation and semiotics in human representation. Paper presented at the Adaptation and Representation, virtual platform @ www.interdisciplines.org.
  • Dennett, D. C. (1991). Consciousness Explained: Penguin Books.
  • Fauconnier, G., & Turner, M. (2002). The Way we Think: Conceptual Blending and the Mind's Hidden Complexities. New York: Basic Books.
  • Lakoff, G., & Johnson, M. (1999). Philosophy in the Flesh. New York: Basic Books.

Week 7

Date: Tuesday, 24th February 2009
Speaker:
 Dr. Ron Chrisley (Informatics, Sussex)

Phenomenal Directness and the Extended Mind: Why Your Beliefs Are Not in Your iPhone

This talk will address a topic that has already been the subject of one COGS seminar this term (Mark Sprevak): Active Externalism. As put forward by Andy Clark and David Chalmers in their 1998 paper entitled "The Extended Mind", and as developed more recently in Clark's 2008 book "Supersizing the Mind", this is the view that cognitive processes at least sometimes extend beyond the biological body into the non-biological environment: "The causally active physical vehicles of content could be spread across the biological organism and the world".

Clark and Chalmers claim that Otto, a hypothetical Alzheimer's patient who uses (in a very specific way) a notebook to remember addresses, has cognitive processes that include the notebook as a component; specifically, the notebook is the vehicle for the content of Otto's beliefs concerning the addresses it contains. They conclude that: "In these cases, a belief is simply not in the head." An even more liberal version of their thesis, one encouraged by Chalmers' foreword to Clark's book, would have it that, for similar reasons, your beliefs are located in your iPhone.

I argue that even if, as I am disposed to believe, Active Externalism is true, we should resist what Clark and Chalmers say about cases such as Otto's; even more so concerning you and your iPhone. I show that a principled distinction can be made between two kinds of components causally related to cognitive processes: indirect components, that are causally involved in a cognitive process by virtue of having first been available as the object of an agent's awareness; and direct components, in which there is no such dependence on availability in awareness.

It seems most compatible with our intuitions that only direct components can be the minded constituents of a cognitive processes; specifically, only they can serve as the physical vehicles of the content of a subject's mental states. It follows that Otto's beliefs cannot be located in his notebook, since the notebook's role in Otto's cognitions is indirect. The same goes for you and your iPhone. Along the way I hope to clarify what is at issue in disputes over active externalism (for one thing, it isn't about the spatial location), and to reveal a weakness in one of the premises on which Clark and Chalmers base their conclusions: The Parity Principle.


Week 8

Date: Tuesday, 3rd March 2009
Speaker:
 Dr. Mike Wheeler (Philosophy, Stirling)

Dwellers on the Threshold: Intelligent Architecture and the Architecture of Intelligence

Increasingly, architects will be designing buildings that, via embedded computers, autonomously modify our spatial and cognitive environments in the light of what those buildings 'believe' about the needs, goals and desires of their users. In this talk I shall attempt to do three things.

  1. Via examples of actual buildings and architectural research projects, I shall sketch out an analysis of a number of different kinds of intelligent architecture, noting a sense in which the development of the field is recapitulating the recent history of artificial intelligence.
  2. Using this analysis I shall argue that although reflecting on intelligent architecture provides a way of developing a response to one common objection lodged against the extended mind (EM) hypothesis in the philosophy of cognitive science (what has been called the portability objection), the same considerations open up a different kind of objection to EM, one that places a severe restriction on cognitive extension, and thus on how one might reasonably conceive of the architecture of intelligence.
  3. Again using the analysis of intelligent architecture developed in (1), I shall turn to the issue of whether human cognitive life will be productively transformed by such buildings. Using concepts from Heideggerian phenomenology (in particular the notions of 'dwelling' and 'deseverance'), as well as some recent work in phenomenologically inspired philosophy of cognitive science, I shall articulate and interrogate the conceptions of what it is to lead a human cognitive life that are shaping current work in intelligent architecture.

 


Week 9

Date: Tuesday, 10th March 2009
Speaker:
 Dr. Andrew Philippides (Centre for Computational Neuroscience and Robotics, Sussex)

Insect Visual Homing: Where Next for the Snapshot Model?

Visual homing, the ability to get back to a nest or goal location using visual landmarks, is a vital capability for insects. Insects appear to achieve this behaviour through a process of image-matching in which the direction to nest or goal is recovered from the difference between the view from their current position and a view or 'snapshot' stored at the nest or goal position. Since it was first proposed, there have been many snapshot-type models which have demonstrated successful homing over a broad range of environments and robotic platforms.

However, we know that insects do not use vision simply to pinpoint a single goal location, but rather to navigate over large scales through complex environments. Moreover, they must learn the visual information for guiding these routes as rapidly as possible. In this talk, I will discuss the use of snapshot models in a natural environment and highlight issues that arise during large-scale navigation. I will then present the strategies that bumblebees use to learn their nest position during their learning flights. Learning flights are stereotypical manoeuvres performed by bees and wasps when they leave their nests and are known to scaffold visual learning. Finally, I will outline a probabilistic approach to place recognition inspired by recent advances in autonomous robotics that we are using to investigate insect visual homing and route learning in particular.

The talk previously scheduled for this week, by Dr. Anil Seth (Informatics, Sussex) has been postponed until 5 May.


Week 10

Date: Tuesday, 17th March 2009
Speaker: Professor Margaret Boden (Informatics, Sussex)

Collingwood and Computer Art

R. G. Collingwood's view of art was unusual in various ways, including his insistence that "art proper" involves the expression of some highly particular emotional experience in the artist's mind. Although most of his criteria of art are satisfied by one or more types of computer art, his aesthetic particularism is not. Or rather, it's not met by computer-generated art (although it could, conceivably, be satisfied by computer-assisted art). This isn't because it involves emotion, but because particularism runs contrary to the spirit of CG-art.


Series organized by Joel Parthemore (jep25@sussex.ac.uk)