Neural networks and neuroscience

[James M. Bower]

I would point out that Michael Arbib's request:

>>  "Models A and B have shown the role of brain regions C and D in functions E
>>  and F - see specific references G and H".

does not necessarily involve "neural networks" in the strict sense at all.
My understanding is that the original question raised by Michael involved
the value of "neural network" research to brain science.

There are many models of brain function that have no relationship to what
are generally accepted as "neural network" forms (connectionist models,
backprop, etc).

Further, I would claim that there are very few examples where "neural
network" type models have had much to say at all about neurobiology.  A
quick look at the NN component of the table of contents of the NIPS
proceedings (NIPS being the long running gold standard for neural
network/connectionist research) should make clear the lack of connection
(or real interest) of most NN practisioners in real neurobiology.
Similarly, a survey of the usual traffic on this mailing list reveals the
same.

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++


With respect to Richard Granger's post,  the statement:

(Various neuro-biological results)  was shown in models to lead
to an unexpected function of not just remembering odors but organizing
those memories hierarchically and producing successively finer-grained
recognition of an odor over iterative (theta) cycles of operation
(Ambros-Ingerson et al., Science, 247: 1344-1348, 1990)

Was not, in fact, unexpected, as the model was specifically designed to do
just this.


Jim Bower