Artificial Sight

Matt Jones jonesmat at physiology.wisc.edu
Mon Jul 23 12:57:56 EST 2001


"Glen M. Sizemore" <gmsizemore at triad.rr.com> wrote in message news:<QqF67.129068$ih.25099455 at typhoon.southeast.rr.com>...
> >The eye does not "send messages" and the brain
>  does not "interpret
> > anything."
> >
> 
> 
> MJ: Speak for yourself.
> 
> -MY- eye sends messages, and -MY- brain
> interprets them just fine.
> 
> If this didn't happen, I wouldn't be able to see.
> What's the
> alternative way of thinking about it? Magic?
> 
> 
> GS: The fact that you can ask this question in this
> way shows how pernicious is the philosophy that
> stimuli are transferred into the body (as "messages"
> no less! - how quaint!) where they are seen or heard
> or "interpreted." In my opinion (and, admittedly, the
> minority opinion) this philosophy is the worst thing
> that ever happened to psychology, philosophy,
> behavioral neurobiology, and many other fields
> concerned with behavior. 


Glen,

I normally hate to say any argument is "semantic". In this case,
however, I believe that the disagreement can be shown to reside in how
we are using the word "message".

Apparently, you use "message" to indicate something like a "purposeful
communication". I assume that's why you make the statement:

>It never fails to amaze me
> that people are so willing, when they are discussing
> the reduction of behavior to neurobiological
> processes, to describe the brain in the same terms as
> is used to describe the behavior of the whole
> organism. 

If I interpret this correctly, you are opposed to thinking of
information processing in the brain as a dialog among various
homunculi, who follow the same principles and actions that the whole
organism does. I would certainly be opposed to this too, because it is
a non-explanation, leading to the usual infinite loop of non-answers.

However, I (and I think, the story covered in the original post) was
using the word "message" to refer to -any- signal passing over a
communication channel that carries information. The signals passing
over the optic nerve to the thalamus and cortex are certainly
"messages" in the information theoretic sense. One can take the
analytical machinery of Information Theory, point it at these signals,
and "decode" what the initial visual stimulus was (see for example, J
Neurosci 1999 19:8036-8042 Reconstruction of natural scenes from
ensemble responses in the lateral geniculate nucleus. Stanley GB, Li
FF, Dan Y.).  This message can be quantified, and its accuracy in
carrying information about the visual stimulus can be evaluated.
Absolutely -no- purposefulness, teleological cause or homunculus needs
to be invoked. Frankly, I think it is -useful- to anthropomorphize.
But that pernicious opinion is beside the point.

Next, we'll need to clarify what we each meant by "interpret". I
haven't thought about this as much, but i guess i would probably lean
toward a definition something like this: "Interpretation" is that
response which brings about actions that are appropriate to the
message in such a way as to maintain the function of the organism.

So, why not come up with different terms that are less likely to be
confused as implying some sort of conscious intent? I think because
the quantitative tools for understanding information transmission were
originally developed to describe human communication channels (i.e.
telegraph and telephone signals) in which intent is definitely
present. However, if one looks at the original papers by Shannon, this
intent disappears. The signal source is treated as a random sequence
generator throughout Information Theory, and the information content
of the message is a purely statistical description of what was sent
and what was received. The important point being that it is
quantitative, has a fair degree of predictive utility, and is based on
mathematically well-formulated principles.

So I stick to my original statements: My eyes send messages which my
brain interprets. If this didn't happen, I wouldn't be able to see.
What's the alternative? The only alternative is that seeing takes
place -without- the transmission of information between the eyes and
the brain.

Cheers,

Matt




More information about the Neur-sci mailing list