From: Ben Goertzel (email@example.com)
Date: Tue Aug 22 2006 - 19:16:35 MDT
On 8/22/06, Michael Vassar <firstname.lastname@example.org> wrote:
> Unfortunately, as far as I know, zero people who have attained high scores
> on open-ended IQ tests have also achieved dramatic real world achievements.
> By contrast, a significant number of people who score very well at a young
> age on traditional IQ tests have done so.
Sure... but there are many more examples of the latter to draw from...
Anyway, I don't really find the analysis of human psychometrics all
that fascinating. Intelligence is both multidimensional and difficult
IMO, rather than focusing on **measuring** intelligence, the
educational and employment and research systems in our society should
focus creating environments in which people can flexibly and
creatively explore the potential of the various dimensions of their
As a mildly amusing aside, last spring my 9 year old daughter (in
spite of being surprisingly non-nerdy for my family) did very well on
a test and placed into the local "Magnet Elementary School Program for
the Highly Gifted" starting next month. What amuses me is the
nomenclature. These days, what used to be called the "honors track"
at the local schools is now called the "Gifted and Talented" track,
whereas what used to be called the "Gifted Program" is now a program
for the "Highly Gifted." It's not too long before "Gifted" becomes
repurposed to mean "Normal" and another new word needs to be invented
for what used to be called "smart" ;-p
Now, this Program for Highly Gifted is focused on teaching kids in a
way that encourages them to think and create rather than on learning
by rote. But does it really benefit us to offer this kind of
opportunity only to the kids who score almost-perfectly on some test?
Rather than focusing on refining these tests (and this is certainly
not sour grapes: my family always does extremely well on such tests),
I'd rather see energy put into figuring out how to make the school
system as a whole focus more on thinking and creating and less on
Yeah, better education costs money. But obviously, properly-educated
people do a better job of pushing toward a positive Singularity...
I won't turn this into a really long off-topic rant on the educational
system (or, oops, maybe I already did). But I just wanted to make the
point that intelligence assessment is really not the most useful thing
to be thinking about, if we want to talk about human intelligence and
the Singularity. We're just not going to be able to make a test that
will pick the N "smartest" people so we can fund them to create the
Singularity for us. Intelligence and humanity are too
multidimensional for that. If we want to maximize the odds of
positive Singularity we should create an environment in which the
diversity of natural intelligence is generally fostered.
(Of course, none of that really matters if I'm right about Novamente,
and the project is properly funded and flourishes and we create a
superhuman AGI on a time-scale much faster than any education reform
can plausibly happen. But if I should get tired of the currently
frustrating pace of fundraising for Novamente, and take up something
more entertaining like music or mathematics or proving theorems about
Friendly Ai instead, then perhaps these comments about education will
become relevant ;-)
-- Ben G
(though none of the extreme
> outliers have, unless you count the SATI pre-recentering and Eliezer) This
> despite the fact that open-ended IQ test score results seem to agree fairly
> well with the age adjusted very high IQs achieved by young high scorers.
> >From: "Ben Goertzel" <email@example.com>
> >Reply-To: firstname.lastname@example.org
> >To: email@example.com
> >Subject: Re: A study comparing 150 IQ+ persons to 180 IQ+ persons
> >Date: Tue, 22 Aug 2006 12:24:09 -0400
> >And, my conjecture would be that performance on an open-ended IQ test
> >would correlate slightly better with *dramatic* real-world
> >intellectual achievement than performance on a timed IQ test.
> >(There are of course many jobs where highly rapid intellectual
> >performance is important, and performance on the timed IQ test would
> >presumably predict performance at those better.)
> >On 8/22/06, Ben Goertzel <firstname.lastname@example.org> wrote:
> >> > Ben: You keep mentioning rapidity. There are a good number of high IQ
> >> > tests that permit very long time intervals to testers, and even some
> >> > open-ended ones, y'know.
> >>OK, but these open-ended IQ tests are not the ones for which scores
> >>are typically reported. I suppose we do not have relevant statistics
> >>on these.
> >>-- Ben
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:57 MDT