[net.physics] Question about Helmut Schmidt's PSI experiment

gwyn%brl-vld@sri-unix.UUCP (01/24/84)

From:      Doug Gwyn (VLD/VMB) <gwyn@brl-vld>

ESP researchers seem to constantly claim that some experimental
outcome shows a trend that "cannot be explained as a chance fluctuation".

The most hilarious example I can recall went something like this:
An "ESP card deck" experiment is run on about 1000 subjects.
Those showing above-average results are retained for further testing
and the others dismissed.
This process is repeated until one "clearly psychic" individual
emerges; he has had a remarkable run of 10 above-average tests
in a row (some of which are considerably above average).
The subject is then subjected to carefully controlled testing by
skeptics, and he fails to consistently perform above average.
This "proves" that PSI is sensitive to the empathy of researchers
involved, since skeptics broke the spell.

lew@ihuxr.UUCP (01/29/84)

An experiment by Helmut Schmidt depicted on the NOVA ESP show purported
to show that a "random walking" light, triggered by radioactivity, ended
up with a greater displacement from zero than could be explained by chance,
when a subject tried to bias the direction of displacement through mental
influence.

I saw the show again with a friend of mine who has a VCR, so I got a chance
to look at the graphically depicted results at leisure (by stop action.)
The number of steps was 6000 and the "unexplainable" displacements were
roughly +120 and -120. The narrator calls them deviations of 2%. 

Unless I'm interpreting the graph wrong, there is nothing unusual at all
about this amount of displacement. The following table shows twenty
final displacements obtained with a C simulation:

82      98      -38     34      70      0       -68     190     -8      124
52      -106    -70     32      -78     -52     20      54      -188    14

... as you can see, final displacements of +120 and -120, would seem to be
readily obtainable by chance. The EXPECTED rms deviation is 77. Let me affirm
that the above table was the first result obtained by invoking my program
with "psi 6000 20" as I composed this article. It seeds rand() with getpid().

Since Dean Radin was involved in the followup to this experiment, perhaps
he would like to comment on this. By the way, Dean did appear in the show,
but not nearly long enough to make a fool of himself :-)  ... (They flashed
a still of him.)

		Lew Mammel, Jr. ihnp4!ihuxr!lew