IUBio Biosequences .. Software .. Molbio soft .. Network News .. FTP

the literary mind

Sergio Navega snavega at ibm.net
Sat Feb 13 06:57:43 EST 1999


AlMaur777 wrote in message <19990212213831.17630.00000615 at ng-fz1.aol.com>...
>I'm a journalist working on an article and book proposal on the idea of the
>"literary mind," as outlined by Mark Turner in his book of the same name.
>Turner's theory comes from research into how the brain creates and
recognizes
>categories then combines them in small spacial stories to make sense of
>incoming sensory data relative to the body.
>Roger Schank, formerly of Yales artificial intelligence laboratory wrote a
book
>called _Tell Me A Story_ saying outright "We think in stories." He
developed a
>theory of "scripts," to explain this. We have a restaurant script, a
grocery
>store script, a new person script. We only remember details that vary in
many
>of these scripts. You remember the grocery store trip where you slipped on
a
>wet floor or they changed where they put the coffee.
>Psychologist John Brunner says we need a cultural psychology to study how
>narrative creates personal, family, social, and national meaning.
>Turner and others base some of their speculations on Gerard Edelman's ideas
>about neruonal group firing (which accounts for categories that receive
instant
>recognition).
>Any comments on this to me privately or on the list would be appreciated,
since
>the neurological connections so to speak, are an important part of this
whole
>idea.
>I am particularly interested in any research that might confirm or
seriously
>challenge any of these ideas. But ideas for further research are also
welcome.
>Allan Maurer
>AlMaur777 at aol.com
>

Dear Allan,

I'm also a great fan of Schank's work. I guess he touched on
very interesting subjects, even being far from any neurobiological
evidence to support it. One of my current concerns is to find
some of these evidences to make a "bridge" between theories such
as Schank's. Other author I think you should keep one eye on is
George Lakoff (I can cite others if you will).

One of the things I like most in Schank is his Conceptual Dependency
formalism to represent semantic structure in natural language.
This is one of the most neglected areas in NL Processing.
I suspect Schank's work will be recognized later as having
"guessed" important things, maybe something comparable to
Donald Hebb's educated guesses in 1949.

You mentioned Edelman's neuronal group theory. I think you may
benefit a lot from reading William Calvin's "The Cerebral Code".
Calvin also defends something on the line of groups of neurons,
but with a particular emphasis on darwinian methods (thoughts
would be the result of a "competition" among several fronts
in which only one thread would be the winner).

All this is at the center of what I believe will be the next
boom in cognitive neuroscience and cognitive psychology.
Although research in the behavior of individual neurons
is still very active, we're seeing some good results
derived from populations of neurons. The basic idea is that
"information" is stored and manipulated in brains through
synchronous oscillations in ensembles of neurons. Calvin's
speculations use this paradigm to suggest how we construct
high-level aspects of our cognition such as metaphors.
I guess you agree with me on the importance of metaphors
when we think about Schank-like "stories".

There are some labs doing serious work on the way information
could be represented and manipulated by this series of
oscillations. A recent work shows how odors are coded in
groups of neurons in the locust's olfactory system.

The greatest difficulty of this research is in obtaining
experimental data. Monitoring the firings of a single neuron
is already a difficult task, imagine doing the same for
thousands of them.

However, what I like most in the resonating ensemble theory
is its straightforward connection with cognitive phenomena.
I wouldn't be surprised if something really plausible
emerged from these studies (well, I'm betting on that...)

Regards,
Sergio Navega.









More information about the Neur-sci mailing list

Send comments to us at biosci-help [At] net.bio.net