Questions on the Nature of memory, personality, etc.

Robert M?rtin robertmaertin at gmx.de
Thu May 20 14:56:17 EST 2004


@ Glen

Hi


> GS: Just a brief interruption; what is a computation? Is any phenomenon that
> can be given any kind of quantitative treatment the "result of computation?"
> Does the Moon "compute" its orbit, for example?
No. If you are looking at the moon and its noisy signal hits your
retinae... then the computation starts.
You now... noise reduction, signal enhancement, covert attentional
effects, edge detection, feature extraction...
that stuff.
No need to get philosophical here. Please read on.

> 
> RM: And I'm not talking about abstract symbol
> manipulating theories from cognitive psychology. I'm talking about
> underlying data-processing structures (nets of neurons of varying
> properties).
> 
> 
> 
> GS: Well..presumably you're talking about EPSPs, IPSPs, action potentials,
> neurotransmitter secretions, G-proteins, etc. etc. etc. etc. etc. etc.
> 
> 
When I thought of computation, I was thinking of neural networks. BUt
as this terms usually refers to classical neural networks that are
used to learn "anything", I did not use the term "neural network".
Obviously the brain computes, we don't need to discuss about that, do
we ? Actionpotential rush in, converge, diverge, sum up, inhibit
surrounding neurons, oscillate, backpropagate... WHATEVER. I call it
computation because thats what it is. Please read on.


> RM: Pretty straight forward...
> What do we do now ? Besides the fact that we are in need of new
> terminology we are right inside the Black Box. We can now make
> assumptions classical behaviourism does not offer the tools for. (Does
> it ?)
> 
> 
> GS: No, it doesn't. But neither does cognitive "science." There is nothing
> new about talking about physiology.

Ahem...
Novelty is not the point here. If we want to explain how "physiology
mediates behaviour" (remember?)
should we leave out the physiology here ? Please keep you comments
till you read the last paragraph :)


> RM: I think that we are past a point were we can take a term like "spatial
> attention", get its definition and look for (or propose) underlying
> mechanisms.
> 
> 
> 
> GS: Sorry, I can't figure out what you are saying here.

"spatial attention" was just an example. I think that when we work at
the junction between behaviour and physiology we can create terms like
"spatial attention" or "response selection" to describe putative parts
of a behavioural task.
We have to accurately describe what signal processing and control
tasks may be involved in a certain behaviour.
When we have done so, we may look for neural correlates of these
mechanisms using known techniques.
But the point is that the terms mentioned above are not as ill-defined
as usual cognitive science terms (e.g. "semantic processing") and I
want to convince you that an extension of the behaviouist vocabulary
is justifiable.
Yes, I know you don't see any cognition so far BUT :
keep on reading I'll come back to that in the end :)

 



> GS: I have no problem with anything you say in the above section, and saying
> that is not inconsistent with anything I have said. You are talking about
> the physiological mediation of conditioned responses. I don't see any
> cognition anywhere. I see Pavlovian conditioning and models of cerebellar
> function. What do you see? And, "no," behavioral terms are not physiological
> terms. But neither are those of cognitive "science."

> RM: Once such a theory has been created, it may be falsified,
> extended, cursed or implemented.
> "Cognition" and "Speech processing" are such HUGE terms. But if you
> stick with smaller bits... we might just find something :)
> 
> GS: And then "cognition" disappears. Which was my point.

> GS: In this post, nothing. I am basically agreeing with you here (except for
> "computation" - which I find to be a pretty vacuous term as currently
> employed). I don't see anything "cognitive" anywhere. I see behavioral
> neurobiology.


Here comes the last paragraph I have told you to wait for :

Now you and I may be at a point to understand one another.
As you stated above, as soon as we use behavioural neurobiology,
computational and signal processing theories we actually come to a
point where it seems to be possible to explain and assemble all the
"little bits" (remember : one piece was spatial attention for example)
that show us how physiology mediates behaviour.
> GS: And then "cognition" disappears. Which was my point.
Yes... it disappears. But only because the terms used by cognitive
science were ill fated from the beginning.
Most of them are too boastful and ill-defined to ever become
elucidated. The classical computational metaphor of mind cCMM (the
brain is a symbol manipulating device) has failed. Good old fashioned
AI, based upon the cCMM does not produce results. Yes cognitive
science (seen in such a manner) is dead.

As you know, I am studying cognitive science right now. And I already
wrote that (almost) everybody at our institute is well aware of
everything that has been said above.
So... why go on ?

The reason is both strange and simple. If we take the bold term
"cognition" and define it as the set of all the "small pieces" we can
try to define and elucidate using the above methods... then we are
doing baby-steps in the right direction :
explaining how physiology ecxtracts & refines environmental
information and produces appropriate behaviour.
But... on second thought I don't care if it's called "cognitive
science" whatnot... as long as I am not forced to do
textbook-cognitive science.
I only think that it does not sound too bad. 
The department of the institute of cognitive science here in
Osnabrueck that I love and respect most is the department of
"Neurobiopsychology" http://www.cogsci.uni-osnabrueck.de/~NBP/ (under
consctuction, but still quite nice).
I think "Neurobiopsychology" is a nice name... 

best regards
Robert



More information about the Neur-sci mailing list