Selected Newsgroup Message
From: "Sergio Navega" <snavega@ibm.net>
Subject: Re: Just a thought....
Date: 26 Dec 1998 00:00:00 GMT
Message-ID: <36876a7e.0@news3.ibm.net>
References: <xz5g2.51$pj.1364@nsw.nnrp.telstra.net>
<36813CB0.4F90AF46@sandpiper.net> <RSig2.120$qn.4208@nsw.nnrp.telstra.net>
<75u4ls$jei$1@bertrand.ccs.carleton.ca>
<OQLVb2EM#GA.406@nih2naaa.prod2.compuserve.com>
X-Notice: should be reported to postmaster@ibm.net
X-MimeOLE: Produced By Microsoft MimeOLE V4.71.1712.3
X-Complaints-To: postmaster@ibm.net
X-Trace: 28 Dec 1998 11:24:46 GMT, 129.37.182.100
Organization: SilWis
Newsgroups: comp.ai.philosophy
Josef <71351.356@compuserve.com> wrote in message ...
>I was just reading "The Number Sense; How the Mind Creates Mathematics" by
>Stanislas Dehaene, where the author states that there are plenty of studies
>that show animals do show some degree of "intelligence", in particular
>mathematical ability.
This is a very nice book. Dehaene is a known neuroscientist with a great
deal of important works.
> There is more research required. And in the latest
>National Geographic magazine their is a piece on Ravens, that also give
>examples of "intelligent" behavior. So, if their is a
universal grammar
>underlying intelligence then perhaps it must also be inter species. Unless,
>of course, each species exhibits a different form of intelligence and thus a
>different species grammar.
>
This is the most obvious conclusion, but it is not correct. Although we
often see in nature animal behavior that is indisputably innate, this
sort of thing is not "the rule" among humans. Suporters of innate knowledge
in humans use animal examples as evidence to conclude that humans must
also possess a great deal of knowledge from the "factory". Among famous
supporters, we can find Noam Chomsky and Jerry Fodor. Chomskyans often
use a gazelle as example: they start walking just after birth. So they
conclude that human babies learn language using a lot of "prebuilt"
knowledge (in fact, they conclude this mainly because they can't
explain how children learn language so fast, given insufficient
examples, unless they carry innate predispositions to syntactic
structures; this is getting strong opposition, as some models
using recurrent neural networks are able to learn language starting
from an empty condition).
The error these guys commit is in assuming that humans belong to the
same "class" of evolutionary machines as do gazelles. Since the
introduction of the Baldwin effect, about a century ago, we've
discovered that some animals start their lives with few learning
abilities (a cockroach is a typical example, their performance
comes basically from "evolutionary learning"), and on the other
side of the spectrum we find humans, where most of what a baby
does is learned (in fact, this condition is easy to understand
because babies are so defenseless).
In my vision, practically everything (except internal automatic
organ control and the initial level of sensory processing) is
learned in humans. This has profound consequences for AI, because
one can find few reasons to start with a system with lots of
"innate" knowledge. The problem is not that it's bad to have
prebuilt knowledge. The problem is that by focusing one's effort
on the preparation of this initial knowledge, it is often easy
to forget that the entity must be able to come up with this
knowledge *by itself*. Without this capacity, only one could get
from such an "AI" system is the electronic equivalent of an
encyclopedia (which means, a way to store the intelligence of
its designers).
Regards,
Sergio Navega.
From: rickert@cs.niu.edu (Neil Rickert)
Subject: Re: Just a thought....
Date: 28 Dec 1998 00:00:00 GMT
Message-ID: <768fig$gtj@ux.cs.niu.edu>
References: <OQLVb2EM#GA.406@nih2naaa.prod2.compuserve.com>
<36876a7e.0@news3.ibm.net>
Organization: Northern Illinois University
Newsgroups: comp.ai.philosophy
"Sergio Navega" <snavega@ibm.net> writes:
>
Suporters
of innate knowledge
>in humans use animal examples as evidence to conclude that humans must
>also possess a great deal of knowledge from the "factory". Among famous
>supporters, we can find Noam Chomsky and Jerry Fodor. Chomskyans often
>use a gazelle as example: they start walking just after birth. ...
>The error these guys commit is in assuming that humans belong to the
>same "class" of evolutionary machines as do gazelles. Since the
>introduction of the Baldwin effect, about a century ago, we've
>discovered that some animals start their lives with few learning
>abilities (a cockroach is a typical example, their performance
>comes basically from "evolutionary learning"), and on the other
>side of the spectrum we find humans, where most of what a baby
>does is learned (in fact, this condition is easy to understand
>because babies are so defenseless).
The Baldwin effect seems to be a weak one.
The example of gazelles is a poor one. Walking is relatively simple
for animals with 4 legs. A newborn gazelle is not nearly a
proficient at walking as it will be a few weeks later. In other
words, the gazelle has a lot of learning to do, even with respect to
something like walking.
From: "Sergio Navega" <snavega@ibm.net>
Subject: Re: Just a thought....
Date: 28 Dec 1998 00:00:00 GMT
Message-ID: <3687d328.0@news3.ibm.net>
References: <OQLVb2EM#GA.406@nih2naaa.prod2.compuserve.com>
<36876a7e.0@news3.ibm.net> <768fig$gtj@ux.cs.niu.edu>
X-Notice: should be reported to postmaster@ibm.net
X-MimeOLE: Produced By Microsoft MimeOLE V4.71.1712.3
X-Complaints-To: postmaster@ibm.net
X-Trace: 28 Dec 1998 18:51:20 GMT, 200.229.242.57
Organization: SilWis
Newsgroups: comp.ai.philosophy
Neil Rickert wrote in message <768fig$gtj@ux.cs.niu.edu>...
>"Sergio Navega" <snavega@ibm.net> writes:
>
>>
Suporters
of innate knowledge
>>in humans use animal examples as evidence to conclude that humans must
>>also possess a great deal of knowledge from the "factory". Among famous
>>supporters, we can find Noam Chomsky and Jerry Fodor. Chomskyans often
>>use a gazelle as example: they start walking just after birth. ...
>
>>The error these guys commit is in assuming that humans belong to the
>>same "class" of evolutionary machines as do gazelles. Since the
>>introduction of the Baldwin effect, about a century ago, we've
>>discovered that some animals start their lives with few learning
>>abilities (a cockroach is a typical example, their performance
>>comes basically from "evolutionary learning"), and on the other
>>side of the spectrum we find humans, where most of what a baby
>>does is learned (in fact, this condition is easy to understand
>>because babies are so defenseless).
>
>The Baldwin effect seems to be a weak one.
>
I'm surprised to read this. I think that without something like
the Baldwin effect we would be today as we were 10 thousand
years ago. Note that I put language (and in special, written
language) as one very important component of transmission of
adaptations that runs in parallel with genetic transmission.
Under this view, Baldwin effect is much more faster and effective
than the cycle combination/mutation/selection of traditional evolution.
I'm not surprised to see the Baldwin effect as the main factor
pushing our technical and scientific development.
>The example of gazelles is a poor one. Walking is relatively simple
>for animals with 4 legs. A newborn gazelle is not nearly a
>proficient at walking as it will be a few weeks later. In other
>words, the gazelle has a lot of learning to do, even with respect to
>something like walking.
>
There is indeed a difference between walking using 4 and 2 legs, and
there's no doubt that gazelles learn things, but even taking this
into consideration, human babies are significantly slower in their
sensorimotor development than baby gazelles.
In a few days Gazelles are running fast while it takes some months
for human babies to crawl. For me, this is an indication that
nature is trading initial performance for greater learning abilities.
It is interesting to investigate why we can't have one organism with
good initial performance *and* good learning abilities.
Regards,
Sergio Navega.
From: rickert@cs.niu.edu (Neil Rickert)
Subject: Re: Just a thought....
Date: 28 Dec 1998 00:00:00 GMT
Message-ID: <768s75$h7t@ux.cs.niu.edu>
References: <768fig$gtj@ux.cs.niu.edu> <3687d328.0@news3.ibm.net>
Organization: Northern Illinois University
Newsgroups: comp.ai.philosophy
"Sergio Navega" <snavega@ibm.net> writes:
>Neil Rickert wrote in message <768fig$gtj@ux.cs.niu.edu>...
>>The Baldwin effect seems to be a weak one.
>I'm surprised to read this. I think that without something like
>the Baldwin effect we would be today as we were 10 thousand
>years ago.
I suggest that biologically, we are about where we were 10000 year
ago. Where we differ is in what is conveyed by cultural means.
>In a few days Gazelles are running fast while it takes some months
>for human babies to crawl. For me, this is an indication that
>nature is trading initial performance for greater learning abilities.
>It is interesting to investigate why we can't have one organism with
>good initial performance *and* good learning abilities.
I suggest that they are mutually incompatible.
From: "Sergio Navega" <snavega@ibm.net>
Subject: Re: Just a thought....
Date: 29 Dec 1998 00:00:00 GMT
Message-ID: <3688c6a4.0@news3.ibm.net>
References: <768fig$gtj@ux.cs.niu.edu> <3687d328.0@news3.ibm.net>
<768s75$h7t@ux.cs.niu.edu>
X-Notice: should be reported to postmaster@ibm.net
X-MimeOLE: Produced By Microsoft MimeOLE V4.71.1712.3
X-Complaints-To: postmaster@ibm.net
X-Trace: 29 Dec 1998 12:10:12 GMT, 129.37.182.237
Organization: SilWis
Newsgroups: comp.ai.philosophy
Neil Rickert wrote in message <768s75$h7t@ux.cs.niu.edu>...
>"Sergio Navega" <snavega@ibm.net> writes:
>>Neil Rickert wrote in message <768fig$gtj@ux.cs.niu.edu>...
>
>>>The Baldwin effect seems to be a weak one.
>
>>I'm surprised to read this. I think that without something like
>>the Baldwin effect we would be today as we were 10 thousand
>>years ago.
>
>I suggest that biologically, we are about where we were 10000 year
>ago. Where we differ is in what is conveyed by cultural means.
>
I agree, physically we are pretty much the same. So in terms of
biological change, Baldwin effect has been negligible (so far; if
genetic engineering proceeds the way it is going, soon we can
have some surprises). But the effect of culture over our mind
grew considerably and Baldwin effect seems to be concerned about
the overall effects of the environment over one species.
>>In a few days Gazelles are running fast while it takes some months
>>for human babies to crawl. For me, this is an indication that
>>nature is trading initial performance for greater learning abilities.
>
>>It is interesting to investigate why we can't have one organism with
>>good initial performance *and* good learning abilities.
>
>I suggest that they are mutually incompatible.
>
That's my impression too, and this makes me wonder that learning
is an ability that demands extreme flexibility. I guess this can be
further developed as another argument against innateness in humans.
Regards,
Sergio Navega.
Back to Menu of Messages Sergio's Homepage
Any Comments? snavega@attglobal.net