Philosophical Issues in HumanEnvironment Interaction

1
Philosophical Issues in Human/Environment
Interaction
2
Classic Vs. Dynamic/Interactive Model
CLASSIC In response to stimulation from world,
perception creates representations that get
passed onto cognition. Agent then calculates
best course of action based on goals, and sends
command to action system. Action system then
triggers behavior.
WORLD
PERCEPTION
COGNITION
ACTION
DYNAMIC Constantly active perception/action
loops provide cognition with messy, partial
information. No detailed internal representation
s the world is its own best model.
WORLD
PERCEPTION
COGNITION
ACTION
3
Some Experimental Paradigms

• Ballard (1991) Animate Vision (AI 48(1)57-86).
• Ramachandrans Bag of Tricks alternative to
  Marrs Vision (1982). A Critique of Pure
  Vision (1994), co-authored with Churchland and
  Sejnowski.
• Changeblindness. No detailed internal
  representation.

4
Motivational Oscillatory Theory
Townsend, James T. (1992), 'Don't be fazed by
PHASER Beginning exploration of a cyclical
motivational system', Behavior Research Methods,
Instruments, Computers 24, pp. 219-227.
5
Epistemic Action
Kirsh Maglio (1994). On distinguishing epistemi
c from pragmatic action. Cog. Sci.18513-49. Prag
matic action aimed at achieving
some non-cognitive goal, like a full
stomach. Epistemic action aimed at changing
the world to make it easier to compute. E.g.,
Experienced Tetris players rotate zoids as soon
as they appear to visually see where they fit
best.
6
Cognitive Anthropology

• Hutchins, E. (1995). Cognition in the Wild. MIT
  Press.
• Navigation of ocean going vessels is a
  collaboration between individuals, cultural
  practices/institutions, and artifacts.
• Its the interaction among these that is
  responsible for successful navigation.
• Cognitive system as social/institutional/artifactu
  al individuals are merely parts of it.

7
The Hypothesis of Extended Cognition

• Clark Chalmers (1998). The extended mind.
  Analysis 5810-23.
• Bits of the environment that are (1) constantly
  present, (2) contain information that is directly
  available, and (3) automatically endorsed, and
  (4) has been consciously endorsed in the past,
  count as part of a persons mind.
• E.g., Alzheimers patient for whom a notebook and
  pen play the same role as memory.
• Argument Any part of the brain that played same
  role in thinking would count as part of the mind,
  and the skin is an arbitrary boundary.

8
Scaffolding

• Vygotskys idea (Thought and Language MIT Press,
  1997)
• External structures, especially language and
  social interaction, scaffold individual
  cognition at first, and then are internalized.
• Language, institutions, social interactions give
  individuals cognitive powers they would not
  otherwise have.
• Zone of Proximal Development children develop
  concepts by internalizing problem solving through
  social/linguistic interaction.
• See also Berk 1994. Why children talk to
  themselves. Scientific American 271(5)78-83.

9
Externalizing the Classical Model

• Clark (1997), McClelland Rumelhart (1986),
  Hutchins (1995).
• The GOFAI model of the mind passive data in
  stable memory processed by a serial processor
  (computer metaphor) - applies to humans plus
  chunks of the environment.
• The processor is the individual brain its a
  massively parallel pattern recognizer.
• The stable memory consists in structures in the
  environment books, maps, diagrams, etc.

10
Philosophical Puzzles

• Agency and responsibility If were all parts of
  one big cognitive system, how can individuals be
  responsible for their cognitive accomplishments?
  Does the self extend beyond the skin? (Clark 1997
  Being There (MIT Press) engages with such
  questions).
• Moral problems Is messing up an Alzheimers
  patients notebook the equivalent to lesioning a
  healthy persons brain?
• Bootstrapping problems How do these external,
  mind-enhancing structures get started?

11
More on Bootstrapping

• My focus of interest.
• How can humans invent or create artifacts,
  institutions, social structures that enhance
  their ability to think?
• Dont we have to be able to think the thoughts
  that such structures enable us to think ahead of
  time, in order to design/invent them?
• E.g., language how did it evolve? Why would we
  need an external medium to help us think? How
  could we tell ourselves something we didnt
  already know?