On Mon, 06 Apr 2009 10:13:50 -0700, Paul Brandon writes:
>Actually, there's no conflict between being a rat running behaviorist 
>and believing that the CNS (central or conceptual as the case may be) 

What's that sound?  Is that B.F. spinning in his grave?

>is an integrated interactive system all of which is involved in any 
>given instance of behavior. And of course we talk a great deal 
>about verbally mediated behavior in all of its complexity. 

I may be mistaken but I think my original point has been lost:

(1)  Conditioning studies involving animals [insert your favorite species here]
may have questionable generalizability to humans because animals
do not have language or high level symbolic processing capability
that would mediate their learning.  Thus, the results of studies like
that described in the NY Times can only be speculatively extended
to humans. Especially if one is going to link human dementias to
the use of a drug that blocks the formation of long-term memories.

Examination of one of the relevant articles for the NYT story makes
it clear that the researchers are using the term memory in a very broad
way but relies heavily on associative learning mechanisms; see:
http://biology.plosjournals.org/perlserv/?request=get-document&doi=10.1371/journal.pbio.0060318&ct=1

(2) Perhaps long-term memories of conditioning are purely mediated
by PKMzeta in rodents and maybe it applies to some limited cases of 
conditioning in humans but it is unlikely to be plausible for many/most human 
long-term memories.  The classic "levels of processing" experimental 
result showing that processing a word's appearance results in poorer 
memory than processing its meaning is clearly a problem if one relies
upon a molecular mechanism or a single neuron mechanism (the problem
is similar to that for the classic "grandmother" or "yellow volkswagen" account
of pattern recognition).  At the very least, a neural circuit is needed and 
perhaps a more complex neural network is needed if one is going to rely 
on purely associative mechanisms in contrast to general rule and symbol 
architectures.  

-Mike Palij
New York University
[email protected]



---
To make changes to your subscription contact:

Bill Southerly ([email protected])

Reply via email to