Re: Actually, Psychological Bulletin is a mainstream publication

From: justin corwin (outlawpoet@gmail.com)
Date: Sun Jan 01 2006 - 10:33:46 MST


On 12/31/05, Damien Broderick <thespike@satx.rr.com> wrote:
> One would hope so. Bearing in mind that we know how to train swimmers and
> runners to perform very well, but still they sometimes fail to win gold
> medals, even after they've just won three earlier in the day.

It's worth noting that in general, Gold Medal Performance is not
actually much higher than 'bad day' performance, in the olympics. It
depends on how high a competition you're in, and other things, aside
from the occasional world performance(which sometimes doesn't mean
you'll get a gold, if the man in the lane next to you also has such a
day)

But that is somewhat something of a side-issue. We don't need the
remote-viewers to be the best in the world, just above random.

> But this is exactly the practical consideration to which I drew attention.
> Millions or hundred of millions of bits, all generated (ideally) in the
> last couple of days before the draw. Pity we need humans in the loop. A psi
> machine would be handy. Maybe an AI will do it.

Well, let's see what we need. If we divide the lotto into single digit
guesses, then for the Mega Millions we have 12 digits to get. each
digit can be 0-9. If you really can get 1 extra correct guess out of a
hundred, and the guesses are otherwise randomly distributed, then
after ten thousand guesses, the correct digit will have an extra 100
points on an average of 1000, which should put it above the noise
floor of perfectly random guesses. Given human inability to actually
output good random noise, we might need much more than that, another
order of magnitude, perhaps. So if we assume 100000 guesses for each
digit, then we need 1.2 million guesses to win the mega millions
lotto. Now there are other considerations, of course, depending on the
output data, you may be able to hedge your bets by buying the tickets
that correspond to the top few candidate digits, lowering the accuracy
requirements.

I participated in a psychic experiment, and managed 1000 guesses in
two hours before getting too bored to continue. I would assume
sustained speed would be somewhat below that, so a good day's work
might be... 5000 guesses? that means that winning the lotto is 240
person-days of work. Assuming self-scoring psychics, that works out to
$216,666 per person-day for the current top US Jackpot of 52 million.
It would take five psychics under two months to do it, 20, twelve
days. A single psychic would be working full days for months and
months.

> Largely, yes. That, and the appalling cultural pollution by crackpots and
> halfwits and scoundrels.

If it's "largely" aversive reflex, why is it that attempted
replications by other parapsychologists even, like of Schmidt's random
number experiments, tend to fail?

"The main fact that emerges from this data is that 71 experiments gave
a result supporting Schmidt's findings and 261 experiments failed to
do so" (Hansel 1989: 185).

These are the kinds of results that I tend to be thinking of when
people mention psi, not my normal skeptical "this is a new and creepy
effect".

--
Justin Corwin
outlawpoet@hell.com
http://outlawpoet.blogspot.com
http://www.adaptiveai.com


This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:54 MDT