Flynn effect - IQ increase 3 pts/decade

Hal (hfinney@shell.portal.com)
Sat, 14 Oct 1995 18:06:49 -0700


I just received my November Scientific American, and found a fascinating
article about the "Flynn effect" (page 12). "In the early 1980s, while
studying intelligence testing in the U.S. military, Flynn found that
recruits who were merely average when compared with their contemporaries
were above average when compared with recruites in a previous generation
who had taken exactly the same test.... Flynn found that scores on
virtually every type of IQ test - administered to military recruits and
to students of all ages - had risen roughly three points per decade since
they were first instituted in the U.S. Flynn learned that 20 other
countries for which sufficient data are available... showed similar
increases.

"The gains ranged from 10 points per generation, or 30 years, in Sweden
and Denmark to 20 points per generation in Israel and Belgium. The
upward surges tended to be greatest for tests that minimize cultural or
educational advantages by probing the ability to recognize abstract
patterns or solve other non-verbal problems. Flynn has recently analyzed
scores from Raven's Progressive Matrices, which is considered to be one
of the least 'culturally loaded' IQ tests. The birth dates of those
examined span a century, ranging from 1877 to 1977. Flynn concluded that
someone scoring in the 90th percentile 100 years ago would be in the
fifth percentile today."

That means that someone who was better on this test than 90% of people a
century ago would be worse than 95% of people today!

The article goes on to say that elderly people score poorly on IQ tests
not because they've gotten dumber as they get old, but because everyone
did so poorly on those tests when they were young.

Several explanations are rejected, among them that people today are more
experienced at test taking (fewer people take tests today than in some
past days), that it is a matter of more education (education hours have
fallen in some countries), that television is making people smarter (the
phenomenon long predates television), and that better nutrition and
medicine have improved health (studies have not found much correlation
between nutrition and IQ).

Flynn himself is apparently too old and hence stupid to believe his own data.
"In fact, he even finds the notion that his generation is significantly
more intelligent than that of his parents ludicrous - and yet that is the
implication of his own research. 'You can see why I'm baffled,' he says
with a sigh." Well, maybe I am being a bit harsh on him. Yet I don't
see why the idea is ludicrous.

I find this result fascinating for several reasons. First, it is
contrary to the conventional wisdom (hence poor Dr. Flynn's confusion)
and that is always good. Plus, this particular bit of conventional
wisdom (that people are getting dumber) is particularly pernicious and
anti-extropian. I would argue that not only are people smarter but they
are wiser and kinder as well. The whole 20th century is like an
awakening, a realization that the comforting lies of the past will not
fit any more. Civil rights, Vietnam, feminimism are just the most recent
manifestations of the ethical progress we have seen. It doesn't surprise
me that people are thinking more these days - they didn't have to do so
back when everything was taken care of for them and everyone knew his place.

The general interconnectedness of the world has played a part as well,
as has the increasing access to information. 100 years ago a lot of
people were farmers living in little towns and never moving far from
them. I can just imagine how flustered they would be when faced with
Raven's Progressive Matrices. Where's the plow? Those bumpkins were
lucky if they could hold the right end of the pencil. Today, people
have to be ready for anything. They are smart, alert, and flexible.
It's not surprising that the occasional farm boy who could outperform
90% of his contemporaries would be left in the dust by young people
today.

And it makes sense to me that the biggest difference is on the abstract
tests. Schooling in the old days consisted of rote memorization and
learning of irrelevant data which could only dull the mind. But they
probably did great on the synonym, math and memory tests. The abstract
tests would show the least effect from the shorter school hours and
simpler textbooks the schools use today.

Apparently famous reputed racist Arthur Jensen is going to have a chapter
on the Flynn effect in a new book on IQ. I hope this phenomenon becomes
more widely known. In its own way it represents exactly the kind of
thing that I think leads to the effect - a surprising result that is
contrary to the easy and unthinking answers which people were contented
with in the past. People have to think, faced with this effect. We've
had a lot of surprises this century, and after a while thinking kind of
gets to be a habit.

Another interesting question is what the implications are for the future.
Even without smart drugs people may be expected to get smarter in each
generation. Hopefully this result will deal another blow to the
negativists with their illusions of deterioration. It is an amazing
effect and I hope we hear a lot more about it.

Hal Finney