JGA%MIT-MC@sri-unix.UUCP (01/24/84)
From: John G. Aspinall <JGA @ MIT-MC> This "random walking" experiment is a good example of something that Gardner and others point out often: Many "psi" experiments hide their data with irrelevant noise. If this "psi adept" can influence the decay of radioactive substances, then you should measure the decay as directly as possible. Instead, this experiment attempts to mask any deviations (or lack thereof) with another noisy process - the random walk. Another well known "psi" experiment involves influencing the fall of dice. If our "adept" can exert force on dice why not put the dice on a sensitive balance and measure the force as directly as possible? "But NOOOO..." (as the late J. Belushi would say), they have to tumble the dice and introduce randomness. Statistics is an extremely useful discipline, but when noise is introduced by the experimenter, any scientist should hear alarm bells. John Aspinall.
mmt@dciem.UUCP (Martin Taylor) (01/30/84)
John Aspinall warns against interpretations of experiments in which the
experimenter introduces noise, presumably on the grounds that the
noise will inevitably mask what is expected to be a small effect.
He worries that the random data will be interpreted (perhaps inadvertently)
as real effects, and perhaps that a real effect may be masked.
There is another way to look at the question: the introduced noise
serves the same function as "dither" in a quantized measurement.
Dither can render audible signals that are smaller than the smallest
quantization step. It can also allow small forces to move a balanced
object that might otherwise be stuck by friction. Since there are both
physical and psychological possibilities of analogues to quantization
and friction in a psychokinesis experiment, the introduction of noise
is quite proper. Naturally, it must be given proper allowance in
the interpretation of the experiments.
--
Martin Taylor
{allegra,linus,ihnp4,uw-beaver,floyd,ubc-vision}!utzoo!dciem!mmtpalmer@uw-june (David Palmer) (02/03/84)
<this line is blank when the mailer isn't looking at it>
For those of you who asked, the address for 'The Skeptical
Inquirer" is:
Skeptical Inquirer
P.O. Box 229
Central Park Station
Buffalo NY 14215
and the subscription prices are
1 Year 16.50 (4 issues)
2 years 29.00
3 years 38.00
David Palmerjlg@lanl-a.UUCP (02/05/84)
As long as someone mentioned The Skeptical Inquirer, I would recommend that people look into some of the back issues of that journal for a more balanced presentation of some of the material in the recent NOVA program. In particular, the remote viewing experiments that were presented so positively have been very carefully criticised. The findings of PSI experimenters to date has yet to provide convincing evidence, biased presentations such as the recent NOVA program do no service to either the supporters of PSI or the critics. If there is something to PSI it will be found only when the mechanisms for self delusion are clearly understood and eliminated from the studies.