Re: death of the mind.

From: Glen M. Sizemore (gmsizemore2_at_yahoo.com)
Date: 08/29/04 

Date: Sun, 29 Aug 2004 09:07:54 -0400

SN: The trouble with behaviorists is that they refuse to develop their
> > theories past the operant conditioning level. We all know that we
> > generalize things. We are able to acquire abstract representations
> > of invariant features of kitchens, houses, cities, libraries, etc.
> > Our behavior is influenced by these abstract representations, up
> > to a point where no theory based solely on behavioristic constructs
> > will satisfactorily explain. The price they pay for this radical
> > vision of science is the sterile state of their explanatory theories.
> >
> > GS: You don't know anything about the basic facts of the science, so you
> do
> > not know how behaviorists use it to explain complex human behavior.
> > Generalization is part of stimulus control. What variables affect
> > generalization Sergio? You are ignorant, and you are arrogant.
>
> SN: And so once more one avoids dealing with the main issue: how are
> we able to acquire the invariant features of kitchens, houses, cities,
> libraries, language, reasoning heuristics, and a host of other
> phenomena?
>
> GS: First be clear on the OBSERVATIONS you would like explained in terms
of
> basic behavioral processes.

SN: When I say "invariant features of kitchens" I am certainly not talking
about direct observations. But I'm talking about an understandable
abstraction.

GS: I'll ask you again: what observations are you asking me to explain.

> SN: How can the explanation of such things be correlated
> with neurobiology?
>
> GS: By hard work, and by not looking for metaphors (i.e., things that
can't
> be found because they don't exist).

SN: Without metaphors and abstract models one can hardly enter into
such complex domains as the understanding of brains. Hard core
neuroscientists talk about memory, []

GS: The term "memory" is not a metaphor. It is a label applied anytime an
animal's behavior can be shown to have been influenced by some feature of
the world that was present, but is not present at the time the animal
engages in the behavior said to "show memory." In that sense "memory" can be
considered an intervening variable (i.e., something that names an empirical
relation). It is when memory (and "memories") is used as a hypothetical
construct that causes trouble.

SN: []retinotopic maps,[]

GS: This is not a metaphor. I have no problem with retinotopic maps.

SN: activations of populations of neurons, etc. These may be metaphors, but
I doubt one would make any progress without using them.

GS: None of what you mentioned are metaphors. You don't seem to know what
you are talking about. Surprise, surprise!

>
> SN: How can these explanations be coherent with
> evolutionary biology,[]
>
> GS: How can they not be (except that "coherent with" is not really proper
> usage)? The basic mechanisms of learning result in the modification of
> behavior during an organism's ontogeny, but the processes themselves exist
> because of natural selection.

SN: These explanations must be *compatible* with evolutionary biology.
Memory, for instance, is a concept that fits well with evolutionary
biology.

GS: Do you even read what I write before you comment?
>
> SN: []computer science,
>
> GS: Why should behavioral phenomena "fit in" with computer science. You
can
> use computers to test mathematical hypotheses about behavior, I suppose,
but
> computers aren't animals and probably aren't like them in very many ways.

SN: You are confusing computers (the machines we have in front of us) with
*computation*, which is the main theme of what one studies in computer
science. Have you heard of Computational Neuroscience? They talk about
memory, representation, maps, information.

GS: Yes, the last class I took as a graduate student was a seminar that
covered Randy Gallistel's book. Needless to say, I found what Randy had to
say quite stupid, even though he's a very smart guy. What a waste.

>
> SN: []linguistics,
>
> GS: Some of what linguists point to are facts about languages. The level
of
> explanation for some of that is cultural - not strictly behavioral in the
> sense I mean it. As to psycholinguistics, behaviorism demolishes it by
> showing that the phenomena are functions of reinforcement contingencies.

SN: Linguistics have important links with evolutionary biology. Linguistics
have also important links with computer science. Cognitive science is
able to contribute with all these disciplines, which increases our overall
understanding of all these phenomena.

GS: I am aware that this is your claim, as well as the claim of many others.
I see nothing to back it up. Is filling journal pages necessarily a measure
of the quality of the "science?" I think not. But other than that, I see
nothing; no technology, little experimental control of their subject matter,
etc.

>
> SN: mathematics,
>
> GS: By pointing to the behavior of mathematicians.

SN: And by disregarding all that can be learned with topics such as
information theory,[]

GS: I haven't ignored it, it just hasn't proven useful to me in trying to
understand behavior. I know in your little made up world where messages are
sent to and from homunculi it is considered important, but I'm talking about
real science.

SN: computational learning theory,[]

GS: Ditto.

SN: statistics,

GS: Four semesters. Nothing wrong, by the way, with descriptive statistics.
And I've published inferential statistics when forced to - which was most of
the time that I worked in a physiology/pharmacology department. And unlike
you, I understand what the use of null-hypothesis statistical testing has
done to mainstream psychology. I showed you how statistical significance can
be obtained when very few subjects are actually changed in any way by an
independent variable, and I urged you to consult a statistician to verify my
assertion. I have pointed out to you that real mathematical science makes
predictions and tries to falsify that prediction. This is the opposite of
what mainstream psychology does when it strives to disprove the null
hypothesis. And I have pointed out how one may obtain significance simply by
adding more subjects. This guarantees that much will be made of trivial
effects.

SN: probability, etc.

GS: Now, "probability" is a notion that I have thought a lot about.

SN: Have you heard of Bayesian modeling?

GS: Yes.

SN: Cognitive science have important links with all such topics.

GS: Oooooh daddy.

SN: Behaviorism?

GS: Yes?

"Sergio Navega" <snavega@intelliwise.com> wrote in message
news:41312bc6$1_5@news.athenanews.com...
> "Glen M. Sizemore" <gmsizemore2@yahoo.com> escreveu na mensagem

Relevant Pages

  • Re: Texas: Creationists appointed to review curriculum
    ... Mechanical does not equal science. ... computers without any need for the knowledge provided by science? ... Computers are math and electronics. ... Babbage's Difference Engine 2. ...
    (talk.origins)
  • Re: self-extracting diskette image to hard disk? (or, IBM keeping inexpensive supercomputers awa
    ... >> also have brought in huge amounts of money, ... > I know what IBM was doing, ... >> claiming not only that memory would be many times larger than existing ... >> selling these super computers for $3000. ...
    (comp.os.os2.misc)
  • Re: Breaking Large Composite Numbers...???
    ... with somewhere between 1 and 10 Tbytes of memory.. ... supercomputers that have more than your 10 TiB of memory. ... because those aren't computers, they are clusters. ... precision calculations on hundred thousands of cores. ...
    (sci.crypt)
  • Re: Physicists dont deliver
    ... scientists the means of constructing the technology that makes you ... I was thinking more in terms of CDs, computers, modern cars, electronics, ... Do you need your subscription to Popular Science renewed? ... dave hillstrom mhm15x4 zrbj ...
    (sci.physics)
  • Re: 386sx/25mhz compatibility
    ... I'd need to add hard drives also, ... a better computer out of the garbage. ... Pentium, mid-2001, suddenly better computers started appearing. ... later, I got a 50MHz 486 with 16megs of memory, for ten dollars. ...
    (comp.os.linux.hardware)