CIA-RDP96-00792R000701020004-4

84
Approved For Release 2000/08/15 : CIA-RDP96-00792R000701020004-4
85
Unexpected Common Patterning of Subject Data. When all 10
subjects' (exploratory and replication) data were examined, 16 of
190 possible comparisons (8.4%) were significant at the .05 level,
whereas 7/190 comparisons of simulation data were significant at the
. 05 level. A graphical analysis of the data revealed an unexpected
common patterning in significantly intercorrelated subject data:
The subjects' intercorrelated data form a pattern that can be
roughly described as a "U" curve. No pattern is visible in inter-
correlated simulation data.
:
Blind Matching. The experimenter's data yielded a rank of 1
and the second exploratory subject's data yielded a rank of 2.
Among the eight replication subjects, 3 yielded ranks of 1 (exact
binomial probability of 3 out of 8 with rank of 1 = . 005, where
p = .
Discussion
The first hypothesis, that subjects temporally pattern RNG
data with which they interact, received weak confirmation. Two of
eight participants showed significant correlation of the temporal
patterning of significant run scores between two separate experi-
mental replications, and a third subject showed a marginal correla-
tion.
Matched simulation data showed chance outcomes.
The second hypothesis, that such patterns would be idiosyn-
cratic enough to allow blind matching by objective means was strong-
ly supported. Three of eight subjects' data (38%) were correctly
matched from within 20 decoy data sets,
What was most unexpected was the similarity of patterning
among individuals. The pattern formed can be described as a "U"
curve. As a group, the intercorrelated subject data showed a pat-
tern of significant hitting near the beginning of the game, followed
by significant missing in the game's middle, followed by recovery to
chance at the game's end. Had a summary measure alone been used,
it would be concluded that "no effect" was present in this analysis:
Consistent, periodic hitting and missing cancel each other out. This
conclusion, if based on individual data sets instead of an ex-post-
facto grouping, would be the same: Half of the 20 data sets showed
a periodic hitting/missing pattern, with the hitting at the beginning
of the game and the missing in the game's middle, followed by a re-
covery to chance levels. Of the 12 simulation data sets that showed
significant intercorrelations, no patterning can be seen.
Random fluctuations should statistically average out, and
there should be no differences between temporal epochs (if there
is no effect of the subject on the data). The present data indicate
that, indeed, subjects do influence random data idiosyncratically,
but the present subject population showed a strong common effect.
This pattern is one that has a long history in standard psychologi-
cal tasks (e.g., the "U" curve associated with learning temporal
sequences of items and attributed to the motivational effects of
"primacy" and "recency"). Replication of the present work is in
progress, and future work will explore other parameters of random
data for evidence of idiosyncratic patterning.
AN EXPERIMENTAL EVALUATION OF A BELIEF IN PSI
Kenneth Reed (Dept. of Psychology, University of Edinburgh,
7 George Square, Edinburgh EH8 9JZ, Scotland)
Belief in psi has been a topic of investigation over recent
years, especially by skeptics of parapsychology. The results
achieved, however, have been inconclusive, and many of the stud-
ies may have been biased. This study was therefore an attempt to
investigate belief in psi taking account of these biases.
The subjects involved in this study were second-year psy-
chology undergraduates at Edinburgh University, a class composed
of 117 students (83 females and 34 males). They were all initially
approached during their practical classes and asked to participate
in the study, when they were told that the purpose of the study
was to collect data to establish population norms for attitudes about
unusual phenomena. Each subject was then handed a questionnaire
which was set up along the following lines.
Section A: Subjects had to reply yes or no, depending on
whether they had experienced any of the following: a dream that
later came true; thought about someone just before they telephoned;
successfully tried to score in a definite way in a card game; been
consistently lucky; been consistently unlucky; known that a letter
was about to arrive. If a "yes" answer was given to any of these
experiences the subjects had to rate that experience on a five-point
scale, 1 (definitely psychic) to 5 (definitely coincidence/normal ex-
planation ) .
Section B: Subjects had to rate a list of phenomena, which
fell into the following groups: "real" -- ball lightning, acupuncture,
dreams, meteorites, hypnosis; "psi" -- ESP, psychic healing, telepa-
thy, mind over matter, clairvoyance, poltergeists; "anomalous" --
Bigfoot, Atlantis, UFOs, spontaneous combustion, Abominable Snow-
man, Loch Ness monster; "false" -- alchemy, phrenology, flat earth,
astrology. The phenomena were rated on a seven-point scale, 1
(not real) to 7 (real).
Section C: Subjects chose a passage which was either for or
Approved For Release 2000/08/15 : CIA-RDP96-00792R000701020004-4

86
87
Approved For Release 2000/08/15 : CIA-RDP96-00792R000701020004-4
against parapsychology, and said how much they agreed with it on
a three-point scale, 1 (strong agreement) to 3 (mild agreement).
Section D: Subjects had to rate the importance of religion
and science on a four-point scale, 1 (major influence) to 4 (no in-
fluence).
The subjects were asked to fill out the questionnaire and re-
turn it the following week. Those subjects who were participating
in the Differential Psychology course (88 students) were approached
a few weeks later and handed a questionnaire (Rotter's Locus of
Control, labeled "I-E Scale"). At a later stage permission was
granted for other personality data (from an Eysenck Personality
Questionnaire, or EPQ test) to be made available to the author for
the purpose of this study. Seventy-six subjects returned the
original "Attitude Survey" questionnaire, 35 of whom also returned
the Locus of Control scale. EPQ data were available for 37 subjects.
There are no sex differences in the number of experiences
reported, although males had a lower (i.e., more psychic) mean
evaluation for all experiences than females (2.7 to 3.7, p < 0.01).
There were no personality differences between subjects who rated
the experiences as coincidental and subjects who rated the experi-
ences as psychic, although subjects reporting fewer experiences
were more neurotic than those who reported more experiences than
average (p < 0.05). Subjects who rated the phenomena highly
(i.e., coincidence explanation) did not differ in personality from
other subjects. Subjects who rated the experiences at the other
extreme (i.e., psychic explanation) appear to be more introverted
and neurotic than other subjects (0.1 > p > 0.05). However, due
to the small number and the level of significance the result should
only be considered suggestive rather than conclusive. The mean
scores for each of the groups of phenomena were as expected,
i.e., real (5.91) > psi (4.41) > anomalous (3.51) > false (2.52).
Subjects who had a higher psi rating (i.e., rated psi as
valid) were more inclined to believe in the validity of anomalous
phenomena (p < 0.01). These subjects also had higher ratings for
the real group of phenomena (p < 0.0005). Males were also more
likely to rate the real and anomalous groups higher than did females
(p < 0.05). Subjects with a high psi rating reported more ex-
periences than subjects with a low psi rating, although there were
no significant differences in their evaluations of these experiences.
There were no personality differences between these two groups,
or groups with more extreme psi ratings.
Sixty-five subjects agreed with Passage A (positive to para-
psychology) and 10 agreed with Passage B. Seventeen subjects
expressed a strong agreement (16 for Passage A and 1 for Passage
B). There were no personality differences between those subjects
who chose Passage A and those who chose Passage B. Similarly,
there were no personality differences between those subjects who
expressed strong agreement and those who didn't. Subjects who
chose Passage A had a higher rating for the psi group (p < 0.0005),
the anomalous group (p < 0.001), and also the real group (p < 0.05).
The mean reported ratings were 2.4 (important /minor influ-
ence) for science and 3.0 (minor influence) for religion. Those
subjects who reported science or religion as being very important
in their lives did not differ in personality from those subjects who
did not report such an importance. There were no differences in
the passage chosen or the group ratings for those subjects who re-
ported a strong importance for science or religion compared with
the rest of the subjects.
From the first part of the study we can see that a belief in
psi is not the result of a specific personality pattern or the conse-
quence of some illogical cognitive process. The idea was therefore
put forward that the previous studies were actually measuring fac-
tors relevant to strength rather than content of belief. Consequent-
ly, the second part of this study was an attempt to measure differ-
ences due to extreme attitudes. However, the results achieved were
inconclusive with regard to this specific question, but this may have
been because of the size of the study rather than for any other
reason.
It therefore seems that the idea that subjects displaying
strong attitudes have a characteristic personality /cognitive profile
is worthy of further investigation. Such a study would have to
take into account the biases that could affect the results, while at
the same time have a large enough number of subjects to be able to
measure extremes. Since, in this case, the content of the belief
may be largely irrelevant, it should be possible to test this notion
using different contents. We would then be able to determine if the
personality /cognitive results do actually relate to one's holding an
extreme attitude or if they are dependent on the particular content
of that attitude. Either way, the results would shed more light onto
the debate about the validity of a belief in psi.
EXPERIMENTAL INVESTIGATION OF BIOLOGICALLY INDUCED
MAGNETIC ANOMALIES*
G. Egely and G. Vertesy (Central Research Institute for Physics,
H-1525 Budapest 114, Box 49, Hungary) **
While electrical phenomena in biological processes are relatively
À
*Abstracted from the full report by Debra H. Weiner.
** Presented in absentia by John Beloff.
Approved For Release 2000/08/15 : CIA-RDP96-00792R000701020004-4
:
.. ....
: