Skip to Content

Department of Cognitive Science

Mechanisms and Levels of Explanation in Cognitive Science Workshop

Macquarie University, Sydney

16th and 17th of december 2010

This conference is co-hosted by the Department of Philosophy and the Macquarie Centre for Cognitive Science.

Locations

Meeting Room 248, Macquarie Graduate School of Management (MGSM) Conference Centre, Macquarie University, 99 Talavera Rd, Macquarie Park. This is building E12A on the campus map.

Keynote Speaker

  • William Bechtel (UC San Diego)

Other Speakers

  • Lise Marie Andersen (Macquarie)
  • Jakob Hohwy (Monash)
  • Brian Keeley (Pitzer College)
  • Paul Griffiths (Sydney)
  • Patrick McGivern (Wollongong)
  • Peter Menzies (Macquarie)
  • Dominic Murphy (Sydney)
  • Jon Opie (Adelaide)
  • Karola Stotz (Sydney)
  • John Sutton (Macquarie)
  • Kellie Williamson (Macquarie)

Registration       

Free. But there is space for only a limited number of participants. To register interest in the workshop please email kellie.williamson@mq.edu.au

PROGRAMME

Thursday 16 December 2010

9.15am-9.30am: Welcome and Preliminaries
9.30am-11.00am: Keynote Speaker: Bill Bechtel (UCSD): "Dynamic Modules"
11.00am-11.30am: Morning Tea
11.30am-12.30pm: Jon Opie (Adelaide): "Is There a Cognitive level?'
12.30pm-1.00pm: Lise Marie Andersen (Macquarie): "Top-Down Causation
and Mechanistic Levels"
1.00pm-2.00pm: Lunch
2:00pm-3.00pm: Paul Griffiths (Sydney & Essex): "Reductive Explanation
and Explanatory Force"
3.00pm-4.00pm: Patrick McGivern (Wollongong): "Levels in Physics and
in Cognitive Science"
4.00pm-4.30pm: Afternoon Tea
4.30pm-5.30pm: Karola Stotz (Sydney): "How (not) to be a Reductionist
in a Non-reducible Universe"

Friday 17 December 2010

9.30am-10.30am: John Sutton (Macquarie): "Going Down and Out:
Mechanisms, Plasticity, and Scaffolding in the Sciences of Memory"
10.30am-11.00am: Kellie Williamson (Macquarie): "From Molecules to
Poetry Groups: A Thoroughly Mechanistic Account"
11.00am-11.30am: Morning Tea
11.30am-12.30pm: Dominic Murphy (Sydney): "Spanning Organism and Environment"
12.30pm-2.00pm: Lunch
2.00pm-3.00pm: Brian Keeley (Pitzer): "Sensory Mechanisms"
3.00pm-4.00pm: Jakob Hohwy (Monash): "The Free Energy Principle: a
Unifying Explanatory Mechanism for the Mind"
4.00pm-4.30pm: Afternoon Tea
4.30pm-5.30pm: Peter Menzies (Macquarie): "Mechanisms and Modularity:
An Interventionist Approach"

NOTES

  1. The workshop is open to the public but space at the workshop will be limited to 55 people. There is now space for only 4 more people unless some people drop out. If you would like to attend the workshop, please contact Kellie Williamson (kellie.williamson@mq.edu.au) as soon as possible.
  2. If you have already notified Kellie of your intention to attend, there is no need to contact her again. However, if you have said that you would attend and have now changed your mind, please inform Kellie of your change of mind.
  3. For information about public transport to the Macquarie University campus, see the website http://www.mq.edu.au/transport/
  4. There is free parking in the MGSM car park for those participating in workshops in the MGSM facilities. Announce that you are attending the workshop at the gates to the Talavera Rd entrance to the MGSM.
  5. Please direct all queries to Kellie Williamson at the email address above.

ABSTRACTS

(1) Bill Bechtel: Dynamic Modules
The notion of a module is invoked in a wide range of disciplines from molecular and evolutionary biology to cognitive science. The specific conception of what a module is differs between fields and theorists, but a common element is that modules are relatively stable parts of mechanisms that regularly function in the same way so that they can be investigated under a variety of conditions. As such, they are the building blocks of mechanisms: mechanistic research strategies seek to
characterize the modules that perform operations within in the mechanism in regular, constant terms. This puts the various research strategies that investigate mechanisms in tension with the emerging systems' perspectives in numerous disciplines that emphasize the complex dynamics exhibited in such systems. The dynamical activity within networks is generally viewed as altering the operation of individual constituents within the system. Integrating dynamic and mechanistic explanatory perspectives (in what Abrahamsen and I call dynamic mechanistic explanations) requires understanding modules as
dynamically changing. I identify a process of scientific discovery in which modules conceived statically provide a first approximation that is further elaborated into accounts that understand them dynamically.

(2) Jon Opie: Is There a Cognitive Level?
I will explore a number of senses in which there might be a cognitive level of explanation. All but the mechanistic approach to levels pioneered by William Bechteland Carl Craver will be found wanting. But this approach to levels call for a serious rethink of our ontological commitments.

(3) Lise Marie Andersen: Top-Down Causation and Mechanistic Levels
Craver and Bechtel advocate a mechanistic conception of levels and argue that in light of this conception top-down causation becomes impossible. They then go on to argue that standard cases of top-down causation can be explained as cases of mechanistically mediated effects. I discuss and question their approach to the issue of top-down causation.

(4) Paul Griffiths: Reductive Explanation and Explanatory Force
J.J.C Smart (1959) argued that biology is a form of engineering. Biology uses physics and chemistry to explain the working of specific mechanisms, albeit naturally occurring ones. Marcel Weber (2005) has christened this 'explanatory heteronomy': the force of biological explanations comes from physical laws. He describes this as a form of reductionism. I examine his assumptions about how the force of an explanation 'distributes' over its components and argue that the force
of typical biological explanations derives from their distinctively biological initial and boundary conditions, not the laws that operate within them.

(5) Patrick McGivern: Levels in Physics and in Cognitive Science
While work on levels of explanation in the cognitive sciences has focused on the role of mechanisms in understanding relationships between different levels, recent work in philosophy of physics has emphasized the role of limit relations between theories in answering similar questions. Since cognitive systems are physical systems, we might expect the account of levels in physics to generate an account of levels in cognitive science as well. However, the two ways of understanding levels, reduction, emergence, and related concepts seem quite different, and it's not clear how the one could have any significant contact with the other. In this paper, I'll examine the prospects for bridging the gap between levels in physics and levels in cognitive science.

(6) Karola Stotz: How (not) to be a reductionist in a non-reducible universe
Part-whole reductions are constitutive, not causal relations that attempt to explain the whole in terms of its intrinsic and fundamental parts and their relations. However, very rarely in biology, and often even in physics and chemistry, are these explanations achieved. What co-specifies the behavior of the components of a system and their relationship is the contingent system-level context, which cannot be reduced to parts that are either intrinsic or fundamental to the system to be explained. Open systems, in virtue of being integrated into more complex systems, can never be explained by their initial
intrinsic constraints alone. Such downward effects from external boundary conditions or constraints have been thought to contradict Kim's Principle of Completeness or Causal Closure of Physics. This arrogant metaphysical claim, however, rests on a mistaken God's Eye view rather than scientifically verifiable facts; there is no complete description of the physical world. Already in physics we see the emergence of some fundamental principles, such as organization or collective behavior, that are independent and transcendent of the laws of quantum mechanics, and which govern most of real-world phenomena.
Therefore the above claim needs to be countered by the Causal Incompletenss Principle. Using some examples from postgenomic biology I argue that a reductionist investigative strategy needs to complemented with an antireductionist, integrative explanatory strategy.

(7) John Sutton: Going Down and Out: Mechanisms, Plasticity, and Scaffolding in the Science of Memory
William Bechtel rightly criticizes certain psychological models of memory which are insufficiently dynamic or constructivist, and seeks a positive account which avoids both ruthless molecular reductionism and merely phenomenal decomposition. This talk puts some friendly questions to his account. I query the way Bechtel deploys disputes between systems theories and process theories in the psychology of memory, then suggest that the plasticity and interactivity of neural systems may pose more of a threat to the strategy of localization and decomposition than he acknowledges. I challenge the assumption that the radically constructive nature of memory is in tension with memory's reliability, and suggest that social or other external scaffolding often complements dynamic and unstable neural memory processes. Bechtel agrees that in mechanistic explanation we need to look around and up as well as down at the components: using examples from the study of cognitive reserve in buffering against memory impairments, I argue that the sciences of memory require yet more attention than he recommends to patterns in the contextual situations within which memory mechanisms operate.

(8) Kellie Williamson: From Molecules to Poetry Groups: A Thoroughly Mechanistic Account
This talk is a response to two very different reductionists: John Bickle and Rob Rupert. It culminates in a defence and extension of mechanism across a variety of disciplines concerned with cognition. I examine Bickle's use of data on Long-Term Potentiation and argue that the processes underpinning memory consolidation are mechanistic.
Moving on from cellular and molecular neuroscience, I argue that in some situations groups of individuals form distributed cognitive systems that can be understood mechanistically. By characterising groups as organised wholes it is possible to overcome Rupert's objection to positing group mental states. The upshot is that mechanisms and mechanistic explanation can be found not just in the biological and neurosciences but also in the social sciences.

(9) Dominic Murphy: Spanning Organism and Environment
Appeals to levels of explanation work well as a strategy in cases where our concern is with processes going on within a system. We bet that processes at one level realize or constitute processes at other levels. But in many cases we need to relate different sorts of causal processes where that bet is not plausible, especially in cases where explanations combine environmental factors with others that unfold within an organism. I shall argue that for most cases in the cognitive sciences levels talk is probably not much use, and look at some alternative ways of thinking, including appeals to mechanisms and to manipulationst theories.

(10) Brian Keeley: Sensory Mechanisms
One place where understanding mechanisms comes in very handy is sensory neurobiology. Here, much effort is spent trying to figure exactly how organisms are engaged with their worlds, especially when dealing with non-human or other unusual senses (pheromones, electroreception, magnetoreception, etc.) How exactly is the process of transduction—the conversion of physical properties external to the nervous system into electrical potentials in the brain—carried out? In this paper, I will consider a number of problem cases that arise in this. For example, how should different mechanisms in different organisms be compared? How should sensory mechanisms be analyzed in order to determine whether we are looking at a case of convergent
evolution? How much do we need to know about the physical realization of a sensory mechanism to declare that we've determined the presence of a sense? Finally, what philosophical understanding of the relationship between higher and lower levels of scientific explanation are required to make sense of such cases?

(11) Jakob Hohwy: The Free Energy Principle: A Unifying Explanatory Mechanism for the Mind
The free energy principle proposes that the brain is fundamentally an instrument for minimising surprising inputs from the environment. I introduce the principle and provide examples of empirical research supporting it. The principle is supposed to have extreme explanatory reach and unifying power: it allows us to understand the nature of perception, agency and attention, and in doing so it demonstrates how they are all aspects of the same type of mechanism. I discuss how the free energy principle dictates a replication of the same type of neuronal mechanism throughout levels of the cortical hierarchy and how
this uniform architecture can play different explanatory roles.

(12) Peter Menzies: Mechanisms and Modularity: An Interventionist Approach
The paper critically examines Carl Craver's account of mechanism, arguing that it does not satisfactorily capture the causal structure of mechanism. The paper uses an interventionist approach to causation to offer a definition of a mechanism in which a modularity requirement plays an important role. The paper considers how this account of mechanism illuminates aspects of mechanistic explanation in cognitive psychology.