DEREK ZAHN wrote:
Richard Loosemore writes:
I am talking about distilling the essential facts uncovered by
cognitive science into a unified formalism.
Just imagine all of your favorite models and theories in Cog Sci,
integrated in such a way that they become an actual system
specification instead of a dog's dinner. Then imagine a lot of hard
work.....
I think your imagination is better than mine. Could you briefly name
two such essential facts and two of the models that should be integrated
(apparently Fodor's Language of Thought and Dennett's theory of
Consciousness don't qualify, which is fine, but I'd like to know which
ones do).
Sure.
Two essential facts, picked at random:
1) Variabillity influences category membership. Rips and Collins
(1993) performed an experiment to show that variability influences our
judgements of category membership. When asked about whether an unknown
object with a length 18 inches was likely to be a pizza or a ruler,
participants were much more likely to say that it was a pizza. Pizzas
vary greatly in size, whereas rulers do not but notice that pizzas and
rulers have the same average length. This is a fact about how the
system represents categories: it cannot simply encode averages for a
prototype, because clearly something more is used. This interacts with
other factors that might determined the architecture of concept
formation and representation, as well as the mechanisms deployed when
categories are discussed in an experimental context.
Rips, L.J., & Collins, A. (1993). Categories and resemblance. Journal of
Experimental Psychology: General, 122, 468-486.
2) The type of knowledge acquired about a new concept depends on the
type of task engaged in during learning. Chin-Parker and Ross (2004)
performed an experinent in which they got people to learn about two
categories of bugs by performing one of two tasks: either a
classification task in which the bugs had to be assigned to the
appropriate category, and an inference task in which people were shown
bugs lacking a feature and were asked to say which of two possible
features belonged in the missing place. Those people performing the
classification task focused on features that differed between the two
categories, whereas those performing the inference task focused more on
the relationships between features within the category. This is a fact
about the general sensitivity of the concept learning mechanisms to the
task engaged in, as well as about the particular circumstances of
learning picture-concepts that are in pre-chosen categories.
Chin-Parker, S., & Ross, B. H. (2004). Diagnosticity and prototypicality
in category learning: A comparison of inference learning and
classification learning. Journal of Experimental Psychology: Learning,
Memory and Cognition, 30, 216-226.
Models to be integrated:
1) Barsalou's situated simulation theory.
2) Chater and Oaksford's model of human reasoning.
I am not saying these two in prticular should be integrated, I am just
providing you with two examples, as per your request.
These are "models" in the cognitive science sense: local models of
particular phenomena.
Richard Loosemore.
-----
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change your options, please go to:
http://v2.listbox.com/member/?member_id=231415&user_secret=fabd7936
