In a message dated 99-04-22 00:32:33 EDT, you write:
This might appear pedantic, but the story used in L021367 is inaccurate. I
also have this "thinking" challenge when I see metaphors used that
inherently lack accuracy. I also think it is tenuous to use complex
systems to explain another complex system. First, the one to one mapping
is unlikely, if not improbable. Secondly, I thought the use of a metaphor
to communicate is to show a general similarity where one novel, therefore
more cognitively demanding model, is expressed, compared to, a well known,
therefore less cogntively demanding model. Where the matching of the two
models is not at the detailed level of analysis but more on the gist level
of cognitive comparison. Hence the easy helps to explain the hard.
Below is partial explanation of the current knowledge of ants navigational
principles and the previous post follows. It is an example of where people
can get totally lost in their comparisons when using complex systems to
express another complex system. The extant information can be inaccurate.
The principle of keep it simple comes to my mind.
The ants navigational system is not based on ground level environmental
cues. Learning theorists have studied the navigational principles of the
ants path because it does appear erratic but highly sophisticated,
efficient. The learning theorists were mainly interested in ants, not
because they have nothing better to do, but because of this incredibly
sophisticated pattern. The ants pattern, to them, didn't represent the
classic stimulus response learning model. It did however turn out to be a
stimulus response pattern of another kind, where the feedback mechanism
was found, and not at ground level where most people would think.
What first caught their attention was how ants would have to "remember" an
extraordinary amount of information to get back home. In a standard path
taken, the environmental cues seemed far too great to recall. To perform
at that level of accuracy was astounding. What they found was the ant has
this ability to use the sun to navigate. The ant calculates this global
positioning and alters its course based on its relative distance from the
hole, and not on ground level environmental cues. When the ant is then
taken manually off course, and placed a great distance from its hole,
while environmental cues have been changed, it can still navigate back to
the nest.
The ground level cues were mainly based on foraging behavior, a pattern of
covering ground.
Using "animal" models to express human behavior is walking on thin ice.
Animal perception and cognition are at times extremely different, well at
least I hope so in some cases...lol.
Glen
[Host's Note: Glen wrote "ques" and I changed these to "cues" in the above
msg. ...Rick]
--- Prev msg Quote ---
Herbert Simon, in his monograph "Sciences of the Artificial," uses the
example of the path taken by an ant attempting to get back to its home.
Its path -- an erratic, almost random-looking one -- cannot be understood
from looking within the ant (assuming one could do so.) Within the ant
are simply a goal (to get home) and some capabilities (leg strength,
etc.), none of which help us much in understanding why the ant is taking
this particular path. No, to understand this particular path one must
look primarily at the environment outside the ant -- what terrain, what
obstructions lie in the ant's straight-line path toward home? Put
together an understanding of the "task environment" and the ant's few
relevant capabilities, then you'll understand why it turned left here
(obstruction too steep to crawl over), went right there (downhill slope),
etc.
--Learning-org -- Hosted by Rick Karash <rkarash@karash.com> Public Dialog on Learning Organizations -- <http://www.learning-org.com>