CIA-RDP96-00787R000500390001-6

...
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
Paper for presentation to the Annual Parapsychological
Association Convention, Moraga, California
August 1979
-
Precognitive Remote Perception:
A critical overview of the experimental program
B. J. Dunne* and J. P. Bisaha
Midwest Parapsychological Research Institute
Evanston, Illinois
..
Running head:
Precognitive Remote Perception
* After June 15, 1979:
School of Engineering/Applied Science,
Princeton University.
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

..
·
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
.
ABSTRACT
A three year experimental program in precognitive re-
mote perception (PRP) provides the data base for a critical anal-
ysis of this mode of ESP. The program consisted of a total of
forty formal trials with nineteen untrained percipients, and pro-
duced a total of eighty-two percipient transcripts of randomly
selected geographical locations where an agent was situated, spa-
tially and temporally remote from the percipeint (s). These eighty-
two transcripts were blind rank ordered against photographs of the
target locations in seven separate series, by a total of one hun-
dred and fifty-seven independent judges. Of the one hundred and
fifty-seven transcript rankings, eighty-four (53.5%) were correct-
ly ranked as one. Various comparisions were made, using Norris'
(1972) and Solfvin, Kelly and Burdick's (1978) statistical tech-
niques for evaluating free-response data. The implications and
problems of the protocol, evaluative methods, and the human factor
in PRP experiments are examined from the standpoint of establish-
ing the fundamental characteristics of this mode of information
transfer, and devising more effective future experiments.
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
L
Introduction
In the spring of 1976 a successful eight trial experiment in
precognitive remote perception (PRP) was conducted by the authors
(Bisaha & Dunne, 1977a; Dunne & Bisaha, 1979b), following a proto-
col first suggested by Puthoff and Targ (1975). That experiment
was the first in an experimental program which, to date, has con-
sisted of a total of 40 formal trials in seven experimental series,
with a total of 19 different percipients. Since several of the
series involved more than one percipient per trial, at this time
we have accumulated a data bank of 82 transcripts which have been
evaluated by independent judges, and 80 of which have been analyzed
in accordance with Solfvin, Kelly and Burdick's (1978) method of
analyzing preferential-ranking data. (Table 2.) In addition, over
30 informal trials have been carried out which have provided con-
siderable anecdotal evidence and insights for future research, even
though they have not contributed any formal data.
The purpose of this paper is to review the results of this ex-
perimental program, to discuss some of the problems and implications
which have emerged from it, and to make some suggestions for consid-
eration in future research in remote perception and other free-re- ,
sponse experimentation.
We have chosen to use the nomenclature of precognitive remote
perception at this point, in preference to precognitive remote view-
ing, since its generality avoids the categorization of this anoma-
lous process as a visual one. It is possible that even the word
"perception" will prove inappropriate once the process is understood
better, however, at this stage of our knowledge it is necessary to
find a description term which is suitably ambiguous, without extend-
ing beyond the prevailing paradigm.
In brief, the PRP experimental procedure, or portocol, requires
one or more percipients to describe, by free-response verbal or writ-
ten narrative or drawing, a remote, unknown target location where an
agent will be situated at a future time, with no available channels
for communication via known sensory modes between agent and percip-
ient, and no means of deducing the target by logical process. The
target is not selected, and therefore is unknown to anyone, includ-
ing the agent or the experimenter remaining with the percipient,
until after the percipient has completed his description.
(See
Table 1 for a sample protocol.)
Experimental Program
Protocol #1.
(Bisaha & Dunne, 1977a; Dunne & Bisaha, 1979b.)
Two inexperienced, volunteer, female percipients were tested
individually, P, participating in 6 trials and P, in two trials.
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

2
-
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
In all eight trials B.D. was the agent, or outbound experimenter,
and J.B. remained with the percipient. Percipients were asked to
spend 15 minutes describing aloud, into a tape recorder, their im-
pressions of where the agent would be between 35 and 50 minutes
later, and to draw these impressions, if possible. The target was
selected randomly from a group of 10 locations which had been ran-
domly selected from a pool of over 100, five minutes after the per-
cipient had concluded her narrative. The contents of the target
pool were unknown to anyone involved in the experiment, including
the two experimenters. (Table 2.)
The original judging procedure consisted of having three sepa-
rate judges blind rank order the eight transcripts of the percipi-
ents' narratives, three with accompanying drawings, against photo-
graphs and descriptive notes taken by the agent at the time of the
trials, on a scale of 1 to 8 (best to worst match). The results of
these rankings were analyzed by Morris' (1972) method for evaluating
preferentially matched free-response material. The sum of the ranks
assigned was 12 in two cases (p=10-4) and 15 in the third case
(p =. 0005). (All p-values cited in this paper are one-tailed. )
These transcripts were subsequently re-judged by three sets of eight
independent judges, each judge ranking a single transcript against
the eight targets, and the results analyzed by the Solfvin, Kelly
and Burdick (1978) technique. The resultant sums of ranks were
20 (p =. 008) , 21 (p =. 012), and 23 (p =. 027). (Table 3.)
Protocol #2
(Bisaha & Dunne, 1977b; Bisaha & Dunne, 1979; Dunne
& Bisaha, 1978.)
In the fall of 1976, a second series of PRP trials was con-
ducted following the same protocol and using the same target pool,
with the exception that the seven volunteer percipients were tested
in four different pairs while both percipients in each trial were
spatially separated from each other, in three instances by a dis-
tance of over ten miles. Seven trials of this sort were performed
with B.D. as the agent, yielding data in the form of one set of seven
target photographs and notes, and fourteen transcripts, two of which
corresponded to each target. The transcripts were randomly divided
into two sets, Group A and Group B, so that each set contained one
description for each of the seven targets.
Each set of transcripts was judged as if a separate experiment,
following the original procedure in Protocol #1. Two judges blind
rank ordered the Group A transcripts against the targets with sums
of ranks of 15 (p =. 01) and 13 (p =. 005) . Two other judges blind rank
ordered the Group B transcripts against the targets, with sums of
ranks of 15 (p =. 01) and 14 (p =. 005). In addition, a fifth judge had
matched both sets with sums of ranks for Group A of 18 (p =. 04) and
Group B of 19 (p =. 10). These transcripts have since been ranked by
four sets of seven independent judges each (two sets of judges for
each of the two groups of transcripts). The results of these eval-
uations, using Solfvin, Kelly and Burdick's (1978) method, were:
Group A sums of ranks of 15 (p =. 008) and 18 (p =. 036) ; Group B, sums
of ranks of 17 (p =. 023) and 12 (p =. 001). (Table 4.)
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

3
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
This design afforded us the opportunity to compare and note
the diversity in individual narrative styles of two percipients
describing the same target. In spite of this diversity, two judges
who were asked to rank order the Group B transcripts against the
Group A transcripts were able to match them with significant results
(rank sums of 12 (p =. 001) and 14 (p =. 005), using Morris' (1972)
table.)
.
Protocol #3
(Bisaha & Dunne, 1977b; Dunne & Bisaha, 1979a; Bisaha
=
& Dunne, 1979.)
In August of 1976, a series of five trials was conducted be-
tween eastern Europe and Wisconsin, with an approximate spatial dis-
tance of 5,000 miles and a temporal differential of approximately
24 hours separating the participants. In all five trials the agent
was J. B. and the percipient was B.D. The agent was on an extended
trip with an itinerary which was undetermined at the time of his
departure, precluding the possibility of compiling a target pool.
Since neither agent nor percipient had ever been in that part of the
world (Russia and Czechoslovakia) and had little or no familiarity
with its topography or geography and had no means of knowing where
the agents tour would place him at any given day or time, it was
agreed that the target would be wherever the agent happened to find
himself between 3:00 and 3:15 P.M. (local European time) and the
percipient would attempt to describe this location between 8:30 and
8:45 A.M. on the previous day.
Upon the agent's return, his photographs and notes were given,
along with the randomized transcripts of the percipient's narratives,
to three judges for rank ordering and analysis by Morris' (1972)
method.
The resultant rank sums were 9 (p =. 05), 11 (p <. 20), and
15 (p <. 20). Re-evaluation with four sets of five independent judges
each, and Solfvin, Burdick and Kelly's (1978) technique, yielded
rank sums of 9 (p =. 041), 11 (p =. 139) , 6 (p =. 002) , and 7 (p =. 007) .
(Table 5.)
Protocol #4
(Bisaha, Dunne & Blauvelt, 1979)
In June of 1977, two carefully controlled trials were carried
out under the supervision of CBS-TV, and were aired on national tele-
vision in a 15-minute segment of "CBS News Magazine" on January 5,
1978. Two experienced percipients, E.W. and D.F., were selected on
the basis of past successful PRP performance, and B.D. acted as a-
gent. In Trial #1, the agent was flown to an unknown destination,
which turned out to be Columbus, Indiana, and a target site was ran-
domly selected from a pool of 10 potential targets, prepared by an
employee of CBS, unconnected with this experiment. The agent visited
the site four hours after the percipient had described the target.
In Trial #2, the target was Rockefeller Chapel in Chicago, also
chosen by random process from an unknown target pool, and visited
by the agent an hour after the percipient described it.
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
Two trials were insufficient for analysis by the usual method,
however, two sets of ten independent judges each were asked to rank
each description against photographs of the ten sites which had com-
prised each target pool. The rank sum for Trial #1 was 27, and the
rank sum for Trial #2 was 10 (all ten judges correctly matched the
transcript to the correct target). (Table 6.)
Protocol #5
In September of 1978, B.D. had occasion to travel in the far
west under circumstances similar to those of Protocol #3, leaving
the last six days of the trip totally unplanned and unknown. We
took advantage of this situation to conduct a series of six multiple-
percipient, multiple-mode, long distance PRP trials. Seven volun-
teers, two of whom the agent had never personally met, from parapsy-
chology laboratories in various parts of the country, served as per-
cipients. Two followed the precognitive mode of the protocol, four
followed a retrocognitive mode (describing the target several hours
after the agent's visit), and one attempted to describe the target
simultaneously with the agent's visit. Two percipients were located
in the Chicago area, the others were in Princeton, N. J .; Brooklyn,
N. Y. ; Menlo Park, Cal .; Durham, N.C .; and the last spent part of
the period in London, England and part in San Antonio, Tex. It was
agreed that the target would be wherever the agent happened to find
herself at noon (Central Daylight Time) each day for six consecutive
days. Each set of transcripts was sent, along with a set of target
photographs and notes to a laboratory other than the one which had
generated that set (with the exception of one of the Chicago per-
cipients, whose transcripts were judged in Chicago) .
The results of these judgings provided rank sums of 6 (with
only five transcripts) (p <. 001), 13 (p =. 036), 14 (p =. 061) , 17 (p <. 145) ,
19 (p <. 145) , 22 (p <. 145) , 24 (p <145) .
(Table 7.)
Protocol #6
This series was conducted in two parts with a person (M. K. )
who had never before participated in a PRP experiment serving as
agent, and the authors serving as percipients. Three trials were
conducted in the spring of 1978, with B.D. as percipient, between
Acapulco, Mexico and Chicago, and three were conducted in the spring
of 1979, with J.B. as percipient, between Florida and Chicago. All
six trials followed the same precognitive protocol, with the percip-
ient attempting to describe the location where the agent would happen
to find herself at a future time. Since all the trials had occurred
in a similar climate at the same time of year and day, we combined
these transcripts and had them judged as a series. Six independent
judges assigned a rank sum of 16 (p =. 145). (Table 8)
Protocol #7
In the spring of 1979, a series of six long distance trials
was conducted between Florida and Chicago, with M.K. as agent and
two inexperienced percipients, J.B. and N.S. Since J.B. was only
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
able to complete three trials, those three were included as part
of Protocol #6, and the results of this series was based only on
the transcripts generated by N.S. The target selection process
was similar to Protocols #3, 5, and 6, with the target being some-
thing in the vicinity of where the agent happened to be at a given
time on six consecutive days. Five of these trials were conducted
precognitively, and one, due to unavoidable circumstances, took
place retrocognitively. In addition, in one trial the agent for-
got to "send" at the prearranged time, creating a control trial,
where the percipient was describing impressions of a non-existent
target. The sum of ranks of all six trials, evaluated by six in-
dependent judges, was 15 (p =. 097). Omitting the "control" trial,
which we had ranked twice and which was assigned a rank of 6 by
both judges, the rank sum was 9 (p =. 006) .
The results of the four
precognitive trials alone was 8 (p =. 007) .
(Table 9.)
Miscellaneous Trials and Anecdotal Evidence
In addition to the 40 trials described above, three formal
trials have been carried out under the conditions of Protocol #1,
but are insufficient for evaluation by the accepted methods. Ar-
rangements have been made to conduct three or four additional trials
in the near future, at the same time of year when the three exist-
ing trials took place, and to combine these for evaluation as a
series.
Of the formal trials conducted to date, six have been discarded;
two because they failed to produce any percipient narratives, three
because the designated targets were non-existent and the agent re-
turned to the laboratory instead of selecting an alternative target
(as in trial #6 of Protocol #1), and one because several interrup-
tions broke the percipient's concentration and she was unable to
maintain her flow of imagery. All other formal trials to date have
been reported above and elsewhere.
Over 30 informal trials have been conducted under a variety of
conditions, including a series of 16 consecutive trials between
Chicago and Russia, while an acquaintance was travelling in that
country. Most of these informal trials were evaluated on the basis
of subsequent exchange of information between agent and percipient
and did not involve target photographs or independent judging.
Nevertheless, we were able to observe many instances of extremely
accurate correspondence as well as some interesting serendipitous
effects which will be taken into consideration in the following
sections, along with the formal data.
General Observations
At this point in the experimental program we have collected
sufficient data to support the hypothesis that some non-sensory mode
of information transfer can function under the conditions of the PRP
protocol. Of a total of 157 transcript judgings, 84 (53.5%) , have
resulted in ranks of 1. Perhaps the success of this design might
be attributable to the fact that it comes closer to simulating
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

Approved For Release 2000/08/07 : CJA-RDP96-00787R000500390001-6
spontaneous psi experiences than most formal experimental designs
because: 1) it utilizes real experiences with real targets in a
naturalistic environment; and 2) the agent/percipient relationship
is closer and less formal as both are active participants in a
shared task within a shared belief system. Under these circum-
stances, the effects of experimenter expectations are likely to be
magnified as a result of being openly communicated to, and shared
by, percipients. However, it also is apparent that this communica-
tion channel is not completely reliable insofar as the clarity and
specificity of the transmitted information, or the "signal-to-noise
ratio" are concerned.
The question at this point is where do we go from here. We
can continue to carry out more confirmatory experiments of this type,
and indeed we are doing so, but these are not likely to tell us more
than that such a communication channel exists under these conditions
and seems to be relatively reliable. We are still far from explain-
ing what the phenomenon is, how it works, or why it fails on some
occasions. In some respects, the experimental procedure is some-
what analogous to the children's game of "Telephone," where a whis-
pered message is passed from one person to another, and after several
transmissions of this kind, becomes distorted, often beyond recog-
nition. We too are dealing with a chain of communications - from
target to agent to percipient to transcript to experimenter to judge -
and every link in that chain is vulnerable to distortion and bias.
At each link in the chain, information is being received, interpreted,
and transmitted by individuals with varying cognitive styles; and
at each transfer point a different mode of perception is being em-
ployed. When the quantification of the fidelity of information trans-
fer finally is attempted at the end of the process, as in the present
statistical procedure, much of the abstract and impressionistic com-
ponent may be overlooked. Then too, it is clear that much of the
unusual and plentiful anecdotal evidence, which seems to provide
empirical confirmation of the process being investigated, is not
taken into account by the judging process and may even work to the
detriment of the final statistical outcome.
One class of such unused evidence is that derived from time in-
tervals other than those prescribed by the protocol. For example,
during the series from the far west (Protocol #5), on one occasion,
several hours before the formal trial was to take place, the agent
unexpectedly found herself riding a very wobbly bicycle which she
found at a gas station where she and her companions had interrupted
their trip. The target site that day was a gambling casino. One
percipient's description, obtained retrocognitively, made no mention
of and contained only vague symbolic resemblances to the casino.
By
the formal judging criteria, this transcript might easily be con-
sidered a miss. However, one part of the transcript read: "I have
an image of (the agent) on a bicycle, now she's on it. She teetered
and tottered a little, but apparently she's OK. She's going down,
well - I have the impression that she's gotten the bike from maybe
like a roadside stand or something like that." This kind of dis-
placed information has been acquired on numerous occasions, and while
such events impress the experimenters as significant evidence of PRP,
they invariably distort the narrative and lower the probability of
.

Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
7
.
A similar effect is displayed in descriptions which contains
elements of more than one target; this has also occurred on several
occasions, notably in those series in which the agent was on a trip
and no feedback was available to percipients until well after the
series was completed and when the efforts were on successive days.
(Protocols #3, 5, 6, and 7.) In such cases, it is clear that infor-
mation is being transmitted, yet the formal results are negatively
affected by it, rather than enhanced.
It is beyond the scope of this paper to examine in detail each
link in this communication chain; nevertheless, they can, and should,
be recognized and taken into consideration in any attempt to formu-
late a model of the PRP process :
1.
Agent/target. First, the agent selects a target via a ran-
dom process which is possibly vulnerable to some psi influence. Then,
at the target, he is involved in some process of perception and rep-
resentation. While it is not completely clear whether the agent is
actually "sending" the information telepathically, or merely serving
as a beacon for the percipient (although the evidence of the single
control trial in Protocol #7 appears to support the "sending" hypo-
thesis), in either case he acquires information regarding the target
via sensory input, which he then translates into an extrasensory
transmission, whether consciously or unconsciously.
.
2.
Percipient/agent. The percipient is attempting to pick up
an extrasensory signal, either from the agent or the target, or pos-
sibly both, against a background of internal and external noise, to
interpret it in terms of his own cognitive patterns, and to trans-
late it into a more conventional communication. At the same time
the percipient may be extrasensorily influencing the agent's percep-
tions of the target. (At the time the envelope was opened, which
contained the designated target in trial #2 of Protocol #4, the agent
experienced a rush of excitement and a sense of certainty that the
trial was successful, and felt a strong desire to go inside the
chapel as well as to view the exterior; indeed, the percipient had
described the interior as well as the exterior of the building.
Up
to this point we had assumed that information was being transmitted
only from agent to percipient, however, in this case it seemed that
not only was the percipient perceiving forward in time, but the agent
was perceiving backward in time in a similar mode.
This experience
has recurred several times since then.)
3. Experimenter/percipient. The experimenter, in his instruc-
tions to the percipient and through the environment he creates for
the experiment, is in a position to exercise considerable influence
on the percipient's performance. It is his words, actions, and at-
titude which mold the percipient's understanding of his task, his
belief in his ability to-accomplish it, and the mood, or emotional
climate, of the trial.
:
4.
Percipient/narrative. The translation of the received sig-
nal into language or drawing involves a subjective interpretation of
the original signal, which can easily be biased by attempts to define,
rather than describe the impressions received. It is important to
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
8
realize that the percipient's interpretation of the signal emerges
in the form of language only after being filtered through a cog-
nitive structure predisposed to categorizing sensory input through
the use of memory, expectations, imagination, etc. Still another
potential distortion lies in the fact that the percipient's free-
response verbal description is presented to the judge in written
form, where nuances of emphasis, tone of voice, pauses, and 'so
forth, are lost.
5. Transcript/experimenter/judge. It is inevitable that an
experimenter will examine the transcript to seek out similarities
between the narrative and the target before passing the transcript
on to the judge. His unofficial evaluation could possibly bias his
expectations for the outcome, which in turn may influence the choice
of judges, the judge's attitude toward the task, or even the judge's
decision by some extrasensory influence.
6.
Judge/transcript. Once again there is a process of inter-
pretation going on, vulnerable to subjective opinion and perspective,
The judge reads the transcript, which is a written version of a per-
cipient's verbal impressions of the original target (links 1-4 above) ,
and attempts to match it to his perception of a photograph of that
original target. (This is why we include the agent's notes with the
photographs; it helps the judge to get a better idea of the agent's
perspective of the target. )
Keeping all these points in mind, it seems clear that if we
are to utilize the PRP protocol for further investigation of the
nature and process of psi phenomena, we must attempt to strengthen
some of the weaknesses in the technique itself, by examining and
evaluating the perception and communication links described above
and finding ways to minimize the distortions of the signal occurring
at these points, and possibly by finding means to evaluate the
quality of the transmission at each link.
Methodological Criticisms
Two methodological criticisms have been directed against the
RP protocol and other free-response experiments. The first is the
issue of target selection and the possibility of psi influence in
the selection of the random number yielding the target. . This pos-
sibility cannot be categorically refuted by any protocol, no matter
how elaborate, but we have taken deliberate care to avoid any logi-
cal deduction of the target. We have employed four different methods
of target selection in our experiments, and it appears that the method
employed had little influence on the results. In Protocols #1 and
#2, where the target was selected through two processes of random
selection (10 envelopes from a pool of 100, and 1 envelope from a
pool of 10), the contents of the envelopes were unknown to anyone
associated with the experiment. In the CBS trials (Protocol #4) ,
once again the contents were completely unknown to any of the par-
ticipants, and in one case even the city in which the target was
located was unknown. In these two trials random selection was care-
fully controlled through the use of electronic random number
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
.

9
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
generators, using in one instance a nuclear decay noise source,
and in another a hand calculator chip. In Protocols #3, #5, #6
and #7 (as well as trial #6 in Protocol #1) there was no target
pool, and the agent personally selected the target from his im-
mediate enviroment, which was itself unknown until the actual time
of the trial, and percipients were completely unaware of even the
general location of the agent. The method utilized in these trials
also permitted the agent to select targets which were as distinc-
tively different from each other as possible, thereby reducing the
possibility of confusing the judges with a target pool containing
several similar sites, as had been the situation in series utilizing
more traditional methods of random target selection.
Since each
judge ranked only one transcript, any deduction he might make as to
the order of the targets would provide him no information regarding
which of those targets corresponded to the transcript he was ranking.
We have tentatively concluded, on the basis of the manipulation of
these variables, particularly trial #6 of Protocol #1, where the
designated target was unavoidably aborted at the last moment, that
the method of target selection is not a critical component of the
process being studied, so long as the target is selected in some
random fashion and cannot be deduced by the percipient through logical
process.
The second criticism, the possibility of sensory cues, has been
raised in an article by Perci Diaconis in Science (1978), in a letter
to Nature by D. Marks and R. Kammann (1978), and by J. E. Kennedy in
an article in J.A.S.P.R. (1979). These critics hypothesize that the
apparent success of remote viewing experiments could be attributed
to an "artifact of statements" in the transcripts which provide ex-
traneous cues to the judges, about the weather on the day of the trial,
for example. While the criticisms of the Science and Nature articles
were not expressly directed at our work, Professor Kennedy has extend-
ed this censure to include our work as well, suggesting that the photo-
graphs taken on the days of the trials and the transcripts of those
days "might have contained cues about the weather on the day of the
trial." This criticism is invalid with regard to these experiment for
two reasons :
first, all trials in a given series took place at approx-
imately the same time of day and any variations in weather conditions
which might have existed were undetectable from the agent's photographs
or notes; and second, all transcirpts were carefully screened before
they were given to the judges eliminating references to weather, the
order of the trials (i.e., remarks such as "yesterday's trial" or "this
is the first (last) day," etc. ), or any other potential identifying
cues.
Judging Strategies and Problems in Quantitative Evaluation
In the nature of the PRP experiment, the data evaluation procedure
is not an integral part of the testing process, but involves post facto
comparisons by individuals who have not participated actively in the
experimental test. Results take the form of relative overall accuracy,
indicated by an assigned rank, rather than an absolute score, as would
obtain in a binary choice design. The distinction between a hit and
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
a miss is therefore blurred, causing the experimental outcome to
be especially sensitive to the personal characteristics of the judges.
For these reasons, there is room to question whether the traditional
judging procedures are appropriate for establishing the fidelity in-
dex of PRP experiments.
The procedure originally followed in the first three PRP experi-
ments was Morris' (1972) method for evaluating preferentially ranked
free-response material. It was pointed out that this method was in-
appropriate to the PRP protocol for two reasons. First, in the origi-
nal experiments, a single judge was asked to rank order the entire
series, which introduced the problem that once he had matched a par-
ticular transcript to a given target he was not as likely to give
full consideration to that target when making subsequent matches.
Second, the judging takes place under closed-deck conditions; that is,
the composition of the target pool is fixed, consisting of those tar-
gets consituting the given series of trials. These two factors were
corrected by having independent judges each match only one transcript
against the target pool, and by switching to Solfvin, Burdick and
Kelly's (1978) evaluative procedure, which, while similar to Morris',
is a more conservative measure as well as being more appropriate for
closed deck experiments. (Note: in the tables giving the results
of the experiments described in this paper, we have included the
p-valves from both statistical tables, for comparison. )
In spite of these corrections, there are still several short-
comings in the preferential ranking technique itself, some of which
have been pointed out in previous sections of this paper. For example,
the judgements of correspondence made are inherently subjective, and
are as much a measure of the individual judge's ability to discern,
interpret, and evaluate the informational content of the transcripts,
as they are an evaluation of the percipient's ability to obtain in-
formation via a non-sensory communication channel and translate it
into traditional communication symbolism. The existence of such sub-
jective bias was suggested by the judging results of the two CBS tran-
scripts (Protocol #4.) Both transcripts had been evaluated by ten in-
dependent judges, and while one had been correctly matched by every
judge, the second produced a variety of different ranks, ranging from
one to four. Even allowing for the similarities in the target pool
in this case, which might have posed a handicap to the judges, there
was still little conformity in the judges' opinions.
In an attempt to gauge the extent of judges' subjective bias,
we had the 27 transcripts of Protocols #1, 2, and 3 re-judged several
times by a number of different judges. The transcripts were all matched
against the target photos and notes of their particular series, the
same choices as had been presented to the original judges. These new
ranks were compared with the original ranks assigned these transcripts
by both non-independent and independent judges, resulting in a minimum
of five ranks assigned each transcript by separate individuals. (Tables
3 - 9.) Of the 27 transcripts, only three were consistently ranked as
1 by all judges, and three others were never ranked higher than 2. Five
Other transcripts were ranked as 1 or 2 by all judges with a single ex-
ception. In all, 13 transcripts received a mean rank score of 2 or less,
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

11
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
However, the fourteen remaining transcripts received a wide range
of ranks, demonstrating a broad diversity in the judges' opinions,
and suggesting that the final positive outcomes of these experiments
were, at least as far as these 14 transcripts were concerned, large-
ly due to chance. That is, after several series of judgings had been
performed on a given series of trials, if we added the highest ranks
assigned by any judge to each transcript, the resultant sume of
ranks for that total series would have been a non-significang figure.
This subjective variability in judges' ranks is especially insid-
ious when applied to the least correlated transcripts. Often, once
a judge has selected his first, and perhaps second choice, he will
tend to be less precise in the ordering of the remaining targets, and
the choice of whether to assign a 4 or an 8 might be purely arbitrary.
However, that extra four points, carelessly assigned, could make the
difference between a significant sum of ranks or a non-significant one.
For example, in series D of Protocol #5, the rank sum was 14, a figure
with a non-significant p value of . 061. Had one transcript in that
series, which was ranked as 5, been assigned a rank of 4 (still not
an outstanding hit), the series would have been defined as significant
at p =. 036. The remaining five ranks in this series were 1, 3, 2, 2,
and 1. Again, the central point is that when the outcome of an entire
series is this sensitive to a single rank, the assignment of those
ranks should not be as vulnerable as they are to the subjective opinion
of a single individual.
A second basic problem with the present judging methods is their
relative insensitivity to description quality. A judge might assign
a rank of 1 to a transcript simply because, in his opinion, there is
some vague resemblance to a minor detail of that particular target
and less resemblance to any of the other targets, or, he might assign
a rank of 1 because the correspondences between the description and
target are so markedly accurate as to exclude the possibility of it
referring to any but the correct target. In either case, the rank is
the same and bears strongly on the statistical outcome of the series.
We attempted to demonstrate this variability by including a
"measure of confidence" indicator in the judging form drawn up for
these new evaluations. After making their selections, judges were
asked to rank the degree of confidence with which they chose their
first-place match, on an ascending scale of 1 to 5. Since we were
unable to compare these confidence indicators across all of the ranks
assigned any given transcript, there was insufficient data to reach
any definitive conclusion other than that not all first-place matches
were made with the same degree of confidence, even though they carried
the same weight. We did observe, however, that those transcripts
which obtained the lowest mean rank sums, particularly the three tran- :
scripts consistently ranked as 1, did appear to have received higher
confidence indicators than most of the other transcripts.
Another difficulty with these procedures is their insensitivity
to striking fidelity of individual trials of a total series. In Pro-
tocol #3, trial #5, the Danube River, was one of those three trans-
cripts which received a rank of 1 from all seven judges who evaluated
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
.

12
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
it. The fact that one of the most outstanding hits in these series
was part of a series which, overall, had lower significance levels,
and in some cases non-significant results, is an indication of this
problem. This insensitivity also applies in the context of evalu-
ating the critical detail within a single transcript. Since each
transcript is judged against a given pool, an unfortunate by-product
of the random target selection process is lack of control over po-
tential similarities of more than one target in a series. For ex-
ample, in Protocol #2, two of the targets, the Lindheimer Observa-
tory and the Grant Park Bandshell, contained rounded structures in
open fields with tall buildings in the background and Lake Michigan
within easy view.
Such similarities complicate the judges' task
and reduce the probability of first place ranks, even when the de-
scriptions are quite accurate, as they were in these two trials.
Unsolicited additional information, although accurate, may further
complicate the problem, as in the case of the bicycle ride mentioned
earlier, or in the case of a transcript which described, in part, the
agent walking through a parking lot and a grove of trees. The agent
did follow such a path en route to the target, but these details were
not specified in the photographs of the target. Still another ex-
ample can be found in the transcript of one of the percipients in
Protocol #5. The target was a young man sitting in an abandoned car
wreck, playing a trumpet. The percipient described the sound of a
horn blowing, but he also described elements which may have fit sev-
eral other targets as well.
(Note: this was one of the sequential
series which seem so sensitive to bleed-through or overlap of trials. )
Once again, the judge's subjective bias, his decision to make his
selection on the basis of the overall impression of the transcript
vs. specific details within the transcript, determines the final out-
come, and this decision may be easily influenced by the percipient's
unique descriptive style, as well as his (the percipient's) choice
of priority and order of transmission of the remotely perceived in-
formation. Some percipients tend to describe minute details, while
others are more general in describing their impressions. For this
reason, we have added to our judging forms, in addition to the con-
fidence indicator, two additional questions for the judge to answer.
1) Did you reach your decision more on the basis of the transcript's
explicit detail, overall impression, or both?, and 2) In reaching
your decision, were you more influenced by the transcript's symbolic
similarities, realistic features, or both?
We have already mentioned the problem of overlap or bleed-through
in series conducted over an extended period with no feedback to per-
cipients until all trials have been completed. This effect has also
been noted by Targ and Puthoff (personal communication) in series
they have conducted under similar conditions. Reference to a similar
phenomenon can be found in the work of Whately Carington (1940) when
he conducted experiments on paranormal cognition of drawings in series
of ten targets without feedback. It is as if, at some level of awarc-
ness, percipients regard the entire series as a single extended trial
when they have no information regarding the outcome of each individual
trial. After all, the task assigned the percipients requires them to
disregard temporal distinction, a difficult enough task without also
requiring that they be able to assiduously pinpoint their location
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

13
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
outside "real" time. The present evaluation procedures do not take
this effect into consideration, and are thus inadequate for assessing
series of this kind.
A final criticism of the current evaluative methods involves the
possibility of psi influences in the judging process itself. There
is no way of ruling out the possibility that the judges, perhaps due
to the influence of the experimenter, are making their matches via
some extrasensory process. In Protocol #5, the two sets of tran-
scripts which were significant belonged to a percipient in a pre-
cognitive mode who had never met the agent, and a percipient in a
retrocognitive mode who was a close friend of the agent's. The only
commonalities they shared were 1) both were experienced, successful
PRP percipients (Diane Freemand and Hella Hammid), and 2) these were
the two sets of transcripts which were judged in Chicago. It might
be noted that, while we had no familiarity with Hella Hammid's typi-
cal descriptive style, the authors both agreed that, in their personal
opinions, Diane's transcripts were not up to her usval descriptive
standards. Under the conditions of the PRP protocol, an argument
could be made to regard the entire procedure, up to the judging pro-
cess, as nothing more than an elaborate and convoluted technique for
selecting targets for an ESP matching test in which the judges are
the subjects. In a later section we will be discussing the role of
the experimenter in influencing the percipient's attitude and per-
formance, as well as the importance of a positive attitude on the
part of the percipient, for obtaining positive results. These same
factors may also be a determinant in the judges' performance. In the
light of this hypothesis, it may be possible that the non-significant
results obtained for some of the transcript sets of Protocol #5 may
be the result of having the judging undertaken under the supervision
of experimenters other than those conducting the experiment.
We are
currently having some of our earlier transcripts re-evaluated at dif-
ferent laboratories to test this.
It seems clear that some alternative strategy for evaluating PRP
experiments must be developed; one that is more sensitive to the in-
tricacies of the phenomenon and, at the same time, more objective in
its assessment of the transmitted information. This paper does not
presume to detail such a strategy, however, perhaps some relevant
factors can be delineated which may percipitate and aid the design of
a suitable process in the near future.
One possibility is the development of a more sensitive ranking
scale and a uniform process for training judges, thus reducing the
subjective bias in evaluation tasks of this kind. Transcripts might
be broken down to the elemental descriptive components of their con-
tent, and each informational bit ranked on a more sensitive scale
against a pool of potential targets. The problem with a procedure of
this kind is that, while it might provide a method for evaluating each
transcript on its own merit, rather than as a single element of an
overall series, it might also negate the influence of Gestalt impres-
sions which involve much more than simple superposition of details and
which provide a "feel" of a particular location to the judge, without
specifying the individual elements in detail. The interpretation of
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
-

14
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
symbolic similarities would still be dependent upon the subjective
perspective of the judge. One might account for the subjective bias
by having each transcript ranked by a number of different judges, as
we attempted to do, but this is cumbersome, time-consuming, and places
a strain on one's ability to recruit a sufficient number of competent
judges, especially when one is trying to evaluate a number of trials
in this fashion.
Some of the earliest free-response experiments attempted to
capitalize on subjective perspective by having the percipient judge
his own transcript against a pool of targets, since the percipient is
more capable than anyone else of interpreting his own impressions and
experiences during the trial. Such a procedure does render the evalu-
ation more sensitive and reduces the vulnerability of the process to
subjective interpretation by eliminating one of the links in the com-
munication chain, however, at the same time it renders the results
more susceptible to criticisms of experimenter/percipient collusion
by eliminating the substantiation of objective verification of the
results.
An alternative might be to eliminate the subjective component al-
together, by designing a procedure by which experimental results might
be evaluated by a computerized process. By altering the protocol some-
what from the unstructured free-response mode presently employed, the
percipient could be presented with a finite list of yes/no questions
regarding his impressions (i.e., is it outdoors?, is there water pres-
ent?, etc. ), the results of which could be quantified easily and eval-
uated by binary logic. However, this intrusion on the PRP process, by
attempting to force the percipient into a predominantly logical mode
of experience, could prove fatal to the phenomenon, if psi is indeed
evidence of "paranormal" perception of cognition. (Nevertheless, this
could prove an interesting line of research to probe the perceptual or
cognitive parameters of the phenomenon. ) Alternatively, a description
obtained in the traditional fashion could be broken down to binary bits
of information. Once again, the problem with this method is that there
would be difficulty in interpreting the many Gestalt impressions with
which PRP transcripts seem to abound.
It might be worth exploring the possibility of developing an al-
gorithm which would code and compare the elements of the target and
the elements of a free-response description, taking such factors as
narrative style into consideration, via electronic pattern recognition.
The solution does not have to lie in an either/or decision be-
tween human judge sensitivity and electronic objectivity. For example,
a hybrid system could be explored wherein each process could evaluate
that aspect of the protocol where its expertise lies. Two separate
scales could be developed and compared: an electronically generated
evaluation of descriptive detail, and a human judge generated evalu-
ation of the Gestalt of the narrative content - i.e. , mood, feeling,
and overall resemblances couched in comparative terms.
The classification of information, whether cognitive or scientific,
is a matter of drawing distinctions. These distinctions are usually
drawn in accordance with agreed-upon systems of rules or definitions
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
.
-

Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
which have proven useful from past experience. One of these rule
systems, that of statistical probability, was designed to organize
and classify large amounts of data, or events, in order to infer a
pattern of behavior which is typical of the group, so that fairly
reliable predictions can be made regarding future events which
appear to belong to the same classification. In psi research, we
are looking at deviations from these patterns; that is, we are ex-
amining a body of events which do not follow or conform to the
normal probability distribution describing this class of events.
We attempt to collect a sufficient amount of this non-conforming
data to begin to classify these events as a new category with
properties of its own, and to form generalizations and make predic-
tions regarding them.
The problem we are facing may lie in the fact that we have
been attempting to force these events which do not conform to our
already existing categories into new categories defined by the same
distincitions which define the behavior of "normal" events. It is
just possible that those rules do not apply to these "paranormal"
events, which is why they are distinguished as "paranormal" in the
first place. We may need to distance ourselves from our precon-
ceived expectations of how psi operates before we can begin to es-
tablish appropriate new categories to describe how it actually func-
tions. The evaluation an analysis of PRP and other parapsychological
experiments at present are predicated on rules which were developed
for the evaluation an analysis of "normal" events, and hence, may
not provide us with a useful basis for representing the essential
nature of psi. We might examine the precedent set by the example of
the development of Quantum Statistics, as a result of the inability
of the rule system of Classical Statistics to provide the appropriate
tools for the task of evaluating certain quantum events.
(Fowler
and Guggenheim, 1952.)
Human Factor Recognition
Parellel to the need for a sufficiently sensitive evaluation
and quantification procedure to measure the information transfer oc-
curring during PRP, is the necessity to develop an adequate framework
or paradigm within which to examine the nature of the information trans-
fer process. We have observed that, regardless how strict the adherence
to the experimental protocol, the degree of success in PRP is still un-
predictable. Apparently, there is still some variable not taken into
consideration in the experimental design, and, since PRP describes a
particular mode of human behavior, it is likely that this unknown fac-
tro is in some way related to the specific characteristics of the peo-
ple participating in the experiment.
A great deal of research has been reported in the literature
which has attempted to establish correlations between psychological
and personality characteristics of subjects and successful psi func-
tioning. (Carpenter, 1977.) However, the results of these experiments,
like so many others in this field, have often failed to be replicated,
and in some instances have even demonstrated contradictory results.
The fact that even subjects of the "ideal" personality type do not
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

Approved For Release 2000/08/07 : 0JA-RDP96-00787R000500390001-6
demonstrate consistently successful psi functioning, suggests that
there is still a missing ingredient in the formula. This raises
the issue of the role of the experimenter in PRP and other psi re-
search. Along with the problem of quantification and evaluation
of data, the questions of "experimenter effects" and performance
feedback are two of the major issues currently being confronted by
psychic researchers as potential sources for the resolution of the
replicability enigma.
-
Reported replications attempted in remote perception experi-
mentation, both formal and anecdotal, support the assumption that
the attitude of the experimenter may be as important a factor as
that of the percipient in producing the desired experimental out-
come. Researchers who hold the belief that remote perception is a
valid possibility which they would like to see confirmed tend to
obtain positive results, while those motivated to disprove the
phenomenon generally report chance, or even below chance results.
The role of the experimenter as a variable in any parapsychological
experiment has been acknowledged by researchers in this field for s.
long time (White, 1977), but very little empirical research has been
undertaken to ascertain to what extent the experimenter's contagious
enthusiasm affects the experimental outcome, or, for that matter, to
what extent the experimenter's, or possibly even the laboratory's,
past successes affect his contagious enthusiasm.
Earlier in this paper it was suggested that the experimenter
and/or agent was in a position to exert influence on the percipient's
(or judge's) performance, by establishing the experimental environment
via his words, actions, attitude, and expectations, thereby molding
the percipient's (or judge's) understanding of the task, contributing
to his confidence in accomplishing it, and providing the emotional
climate of the experimental trial. This suggestion is in accord with
the so-called "Rosenthal effect", in which it has been demonstrated
that an experimenter's wishes and expectations may bias the outcome
of the experimental data.
(Rosenthal, 1966.)
It may well be that the psi effects demonstrated in PRP research
are not simply evidence of an individual percipient's, or experiment-
er's, "paranormal" ability, but are by-products of the interaction be-
tween the experimenter and the percipient. , (None of the participants
in any of our trials considered themselves to have any unusual "psychic"
talents.) If this is the case, psi cannot be predicted or evaluated
simply on the basis of the personality parameters of either of the par-
ticipants, or even on the basis of an additive process, such as Per-
sonality A + Personality B = psi. The process, or interaction, is
complicated by factor E: the environment within which A and B inter-
act. Within the context of the experimental protocol, A and B together
may be capable of behavior of which neither is capable independently.
The personality factors which have been identified as being represen-
tative of good subjects, may be no more than the characteristics which
define those individuals best capable of entering into, contributing
to, and functioning within the type of bonded interaction which is
conducive to psi effects.
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
Perhaps one of the most relevant aspects of the experimental
procedure we followed in these experiments, which does not appear
on Table 1, is the fact that before each of the experiments began,
the percipients were briefed informally about the nature of the ex-
periment and the experimental protocol, as the agent took time to
talk casually with each percipient in an informal, friendly atmo-
sphere, attempting to establish a comfortable rapport and a' shared
enthusiasm for the project.
To date, psychic research has been proceeding along a linear
path, attempting to identify each of the individual components con-
tributing to the psi process. Yet, the subjective reports of in-
dividuals participating in PRP experiments suggest that there is
an element of the experience which defies cognitive categorization.
Two percipients who have each participated successfully in at least
seven separate trials, were asked to describe their experiences in
terms of their feelings. One expressed this in terms of feeling
"tuned into" the experimenter/agent, "like we were both on the same
wave length." She said she didn't think about the nature of the
task or the logical impossibility of accomplishing it, but relaxed,
tried to clear her mind of any extraneous thoughts, and concentrated
on trying to visualize the agent and sensing the rapport between them.
The second percipient explained that she had to "be in the right mood",
which involved a "willing suspension of disbelief and a general mental
posture of receptivity." In addition, she described a "sense of over-
all physiological alertness." Even after ten successful trials, she
does not consider herself to have any unusual psychic abilities. When
asked to what she attributed her remarkable success rate, she replied,
"Some kind of high energy level combined with intense concentration."
One of the authors, B. D., who has been a percipient in eight formal
trials and several informal ones, has described her experiences in
similar terms. In addition, as agent, she has sensed the same kind
of resonance with the percipient (s), a feeling of heightened alertness
and excitement, and an awareness of an emotional, as well as an in-
tellectual, involvement in the experiment which is similar to par-
ticipating in an exciting game. It is as if one "makes believe" that
one can accomplish the impossible to the extent that it becomes real.
Perhaps the current linear direction of parapsychological research
should be complemented by a holistic perspective of the phenomena.
This would involve, in part, an investigation of the nature of the ex-
perimenter/percipient, and the experimenter/judge, relationships. We
have observed, informally, that the more comfortable, intimate, and
warm the relationship between the agent and the percipient, the more
confident we have felt of success in the endeavor. The results of the
trials we have conducted appear to support this assumption, although a
formal analysis along these lines has yet to be attempted.
Concluding Observations
The course of future research in PRP will be dependent upon sev-
eral factors, such as whether experimentation is motivated by a desire
for more specific phenomenological demonstration or the development
of the utilitarian potential of the information transfer process. Both
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

.
18
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
lines of exploration are worthy of pursuit, although they would,
most likely, take different directions. The former would stress
the interactive psychological aspects of the phenomenon, employ-
ing a variety of percipient personality types and a broad range
of freely-selected target material. The latter would concentrate
on improving the fidelity index of transmitted information and limit
itself to working with a few highly trained percipients to develop
a method of optimal relay of clearly defined details of specific
types of targets. Appropriate evaluative techniques would be needed
in both instances.
In the former case, the emphasis would be placed
on the development of a method sufficiently sensitive to individual
narrative styles, abstract or symbolic representations, critical de-
tail, and Gestalt impressions. In the latter case, binary repre-
sentation of specific detail might be preferable for the determina-
tion of precise measurement of "signal-to-noise ratio, " data acqui-
sition rate, etc.
Up to this point, PRP research has provided a quantity of data
which bears strong evidence of some kind of non-sensory information
transfer taking place under the conditions of these experiments.
The results of this experimental program have supported the hypo-
thesis that this information transfer is not limited by temporal or
spatial barriers, in spite of the difficulty that such findings pose
in the light of generally accepted physical laws. Yet, we believe
that the consistently positive results we have obtained can be at-
tributed, in part, to the fact that the precognitive aspect of the
design reinforces the logical impossibility of the task, forcing
the percipient and agent into a "paranormal" mode of communication.
In addition, we have tentatively concluded that the specific temporal
distinction established by the protocol cannot always be enforced,
particularly in series of consecutive trials where no feedback is
available at the end of each trial.
Some of the additional findings of this program can be summa-
rized as follows :
1)
It appears as if the agent's attentional direction may be
more important factor in the process than the contents of the tar-
get envelope.
2) The method of target selection does not appear to be a
critical component of the process, as long as the target is chosen
in a random fashion and cannot be deduced by the percipient through
normal logical processes.
3)
Moving targets are detected as easily as stationary ones.
4)
The assumption that a relaxed, quiet environment is a pre-
requisite for successful PRP is not borne out, since several trials,
including the two CBS trials of Protocol #4, were conducted success-
fully under contrary conditions of high tension and excitement and
in the presence of television personnel and paraphenalia; however,
this tension and excitement were of a positive nature, which may
have had the effect of increasing motivation rather than producing
anxiety. This does not mean that a relaxed, quiet environment and
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

19
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
state of mind is not conducive to psi functioning, it merely sug-
gests that the process can function successfully under alternative
conditions as well.
5)
The PRP process is not limited to two-person interactions,
and appears to function equally as well with more than one percipient.
6) Ungifted percipients appear to be able to demonstrate this
ability without extensive training, and the roles of agent and per-
cipient appear to be interchangeable.
The shortcomings of the experimental design, discussed at length
in this paper, while presenting difficulties in the description and
quantification of the results, provide no evidence to deny the ex-
istence of the hypothesized communication channel. These problems
are not insurmountable, but rather present a challenge to scientists
dedicated to discriminating orderly patterns in apparently random
events .
The existence of "paranormal" phenomena is a fact, supported
by anecdotal reports since the beginnings of recorded history and
ever more scholarly investigations over the past century. If we can ..
not find a way to fit this fact into our existing models of reality,
then it is just possible that the models themselves are in need of
revision.
-
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

20
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
References
Bisaha, J.P. and Dunne, B.J.
Precognitive remote viewing in the
Chicago area: A replication of the Stanford experiment. Re-
search in Parapsychology, 1976: Proceedings of the Parapsy-
chological Association Convention, Utrecht, The Netherlands,
1976, Metuchen, N.J .: Scarecrow Press, 1977a.
Bisaha, J.P. and Dunne, B.J. Multiple subject and long distance
precognitive remote viewing of geographical locations.
Proc
7th International Conference on Cybernetics and Society, IEEE,
Washington, D.C., September, 1977b (invited paper) .
Bisaha, J.P. and Dunne, B.J.
Long distance and multiple subject
precognitive remote viewing of geographical locations.
Chapter
in Mind at Large, Charles Tart, ed. (in press) .
Bisaha, J. P., Dunne, B. J., Blauvelt, D. Precognitive remote view ..
ing for CBS-TV News Magazine: Two experimental trial: .
Re --
search in Parapsychology, 1978: Proceedings of the Parapsy-
chological Association Convention, St. Louis, Mo., 1978.
Metuchen, N.J. :
Scarecrow Press, 1979 (in press) .
Carpenter, J.C.
Intrasubject and subject-agent effects in ESP ex-
periments.
Handbook of Partpsychology, B. B. Wolman, ed.
New
York:
Van Nostrand Reinhold Co., 1977.
Carington, W. W.
Experiments on the paranormal cognition of drawings.
Journal of Parapsychology, 1940, 4, 1-134.
Diaconis, P.
Statistical problems in ESP research.
Science, 14
July 1978, 201, pp. 131-136.
Dunne, B.J. and Bisaha, J.P.
£
Multiple subject precognitive remote
viewing.
Research in Parapsychology, 1977: Proceedings of
the Parapsychological Association Convention, Washington, D. C.,
1977. Metuchen, N. J. :
Scarecrow Press, 1978.
Dunne, B.J. and Bisaha, J.P. Long distance precognitive remote
viewing.
Research in Parapsychology, 1978: Proceedings of
the Parapsychological Assn. Convention, St. Louis, Mo. , 1978.
Metuchen, N.J .: Scarecrow Press, 1979a (in press) .
Dunne, B.J. and Bisaha, J.P. Precognitive remote viewing in the
Chicago area.
Journal of Parapsychology, March 1979b (in press) .
Fowler, R. and Guggenheim, E.A.
Statistical Thermodynamics.
New
York:
Cambridge University Press, 1952.
Kennedy, J. E.
Methodological problems in free-response ESP experi-
ments.
The Journal of the American Society for Psychical Re-
search,
January 1979, 73, pp. 1-15.
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

21
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
Marks, D. and Kammann, R.
Nature, 274, 1978, pp. 680-681.
Morris, R.L. An exact method for evaluating preferentially matched
free-response material. Journal of the American Society for
Psychical Research.
October 1972, 66, p. 401.
Puthoff, H. E. and Targ, R.
Precognitive remote viewing.
Research
in Parapsychology, 1975: Proceedings of the Parapsychological
Association Convention, Santa Barbara, Calif. , 1975.
Metuchen,
N. J. :
Scarecrow Press, 1976a.
Rosenthal, R.
Experimenter Effects in Behavior Research. New York:
Appleton-Century-Cross, 1966.
Solfvin, G., Kelly, E., and Burdick, D.
Some new methods of analy-
sis for preferential-ranking data.
Journal of the American
Society for Psychical Research, April 1978, 72, p. 93.
White, R.A.
The influence of the Experimenter motivation, attitudes
and methods of handling subjects in psi test results.
Hand-
book of Parapsychology, B. B. Wolman, ed. New York:
Van Nos-
trand Reinhold Co., 1977.
.
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

:
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
Precognitive Remote Perception
22
Table 1
Sample Experimental Protocol
(Precognitive Remote Perception)
-
.
.
-
.
10:00
Outbound experimenter leaves with 10 envelopes containing target
locations and begins 20-minute drive.
Experimenter remaining with percipient elicits description of
location where outbound experimenter will be between 10:35 and 10:50.
10:15
Percipient response completed, at which time laboratory part of
experiment is over.
10:20
Outbound experimenter generates random number between 1 and 10,
counts down to associated envelope, opens it and proceeds to
.
target location indicated,
10:35
Outbound experimenter arrives at target and remains there for 15
minutes, until 10:50.
10:50
Outbound experimenter photographs the target and makes notes of
her impressions of it, then returns to laboraotry.
Experimental
trial completed.
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
Precognitive Remote Perception
23
Table 2
Cumulative Results of PRP Experimental Program
-
tocol
# of
.Percipients
# of
Trials
# of
Transcripts
# of
Judges
Times
Judged
X Sum
of Ranks
p-value*
1
2
8
8
27
6
17.2
.002
2-A
5
7
7
17
5
15.8
.041
B
6
7
7
17
5
15.4
.008
Total #2
7.
7
.14
34
5
15.6
>.001
*3
1
5
5
23
7
10.1
.079
$4
2
2
2
20
10
(27) (10)
75-A
1
6
6
6
1
13
.036
B
1
6
5
5
1
6
>.001
C
1
6
6
6
1
22
<. 145
D
1
6
6
6
1
14
.061
E
1
6
6
6
1
17
<. 145
F
1
6
6
6
1
19
<. 145
G
1
6
6
6
1
24
<. 145
tal #5
7
6
41
41
1
16.4
#6
2
6
6
6
1
16
.145
#7
1
6
6
6
1
15 (a)
.097
9 (b)
.006
.
8 (c)
.007
tals
19
40
82
157
.
* Solfvin, Kelly & Burdick, 1978
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
(a) Includes control trial.
(b) Fycludes col
,1 trial.

Precognitive Remote Perception
24
Table 3
Ranke Assigned by Judges in Protocol fl
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
Ranks Assigned
Brget
Judge 1
Judge 2
Judge 3
Ind J. 1
Ind. J. 2
Ind. J. 3
IPS
Bindmill Cemetery
1
2
3
1
1
8
2.7
Adler Planetarium
1
2
1
1
4
1
1.7
Playboy Building
3
1
3
3
1
3
2.3
Karina Towers
1
1
1
2
3
3
1.8
incoln Park Conservatory
1
2
2
1.
1
2
1.5
Elks Memorial & Headquarters
2
2
3
5
2
2
2.7
Ivanhoe Restaurant
2
1
1
6
8
3
3.5
Angel Guardian Florist
1
I
1
1
1
1
Sum of ranks
12
12
15
20
21
23
17.2
p value (one-tailed)
(Morris, 1972)
10-4
10-4
.0005
.01
.025
.04
.002
p value (one-tailed)
(Solfvin, Kelly &
Burdick, 1978)
>.001
>.001
>.001
.008
.012
.027
.002
.
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

Precognitive Remote Perception
25
Table 4
Ranks ( ssigned by Judges in Protocol #2
Ranks Assigned
Group A Transcripts
Group B Transcripts
JIA
J2A
J3A
Ind. IA
Ind.2A
X
J1B
J2B
J3B
Ind. 1B
Ind.2B
Plaza del Lago
(P4, P5)
1
5
6
2
3
3.4
2
4
2
4
1
2.68
Wrigley Field
(P10,Pg)
1
1
2
1
2
1.4
2
3
3
4
1
2.80
Techny Mission
(P4, P5)
3
I
1
7
6
3.6
3
1
3
2
2
2.480
Lindheimer Observatory
(P6, P7)
2
3
2
1
1
1.8
3
1
3
2
2
2.20
Madonna della Strada
(P5, P4)
3
1
2
1
4
2.2
2
2
2
4
3
2.00
NOR Station
(P6, Pg)
2
1
1
2
1
1.4
1
3
2
1
2
1.
Grant Park Bandshell
(P7, P6)
3
1
4
1
1
2
1
1
1
1
1
1
Sum of ranks
15
13
18
15
18
15.8 14
15
19
17
12
15.48
p value (one-tailed)
.01 .005
.04
.01
.04
.01 .005 .01
.10
.025
.001
.01
(Morris, 1972)
p value (one-tailed)
(Solfvin, Kelly, &
Burdick, 1978)
.008 .002
.036
.008
.036
.041 .005 .008 .055
. 023
.001
.008
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
Percipients
Target
(A
B)
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
1

Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
·
Table 5
Ranks Assigned by Judges in Protocol #3
Target
Rar.ks Assigned
Judge 1
Judge 2
Judge 3
Ind. J. ].
Ind. J. 2
Ind. J. 3
Ind. J. 4
Moscow Exhibition
1
2
2
1
3
1
1
0.6
Taxi to Ukraine Hotel
2
1
3
2
3
1
2
St. Michael's Church
3
3
5
3
2
2
1
.1
Tertrezeko Art Gallery
2
4
4
2
2
1
2
5.4
Danube River
1
.
1
1
1
1
1
1
Sum of Ranks
9
11
15
9
11
6
7
18.1
p value (one-tailed)
(Morris, 1972
.05
.20
<. 20
.05
.20
.005
.01
0
p value (one-tailed)
(Solfvin, Kelly &
Burdick, 1978)
.041
.139
<. 139
.041
.139
.002
.007
879
.
Approved For Relagse 2000/08/07 : SIA-RDP96200787R000500390001-6
.
Precognitive Remote Perception
26
,

Precognitive Remote Perception
27
Table 6
Ranks Assigned by Judges in Protocol #4
Target
Ranks Assigned
Rank Sum
x
-
Columbus, Indiana Public Library
1
2
2
2
3
3
3
3
4
4
27
Rockefeller Chapel
1
1
1
1
1
1
1
1
1.
1
10
1
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
Approved For Release 2000/08/07 : CIA-RDP90-00787R000500390001-6

Precognitive Remote Perception
28
Table V
Ranks Assigned by Judges in Protocol #5
Target
Ranks Assigned
Set A
Set B
Set C
Set D
Set E
Set F
Set G
=
x
Feathered Pipe Ranch Lodge (Mont. )
1
1
2
1
1
3
4
1.86
Pontiac wreck and trumpet player (Mont. )
3
1
5
3
1
4
3
2. 8
Herd of cows and culvert (Idaho)
2
1
4
2
1
6
6
3.2
Gambling casino (Nev.)
2
1
2
5
5
3
6
3.43
Coffee shop (Calif.)
3
2
4
2
4
2
4
3.00
Driving down Pacific coast (Calif.)
4
5
1
4
1
1
2.87
Sum of ranks
13
6
22
14
17
19
24
16.62
p value (one-tailed)
(Morris, 1972)
.04
. 0005
X.20.
₹. 05
<. 20.
<. 20
<. 20
p value (one-tailed)
(Solfvin, Kelly &
Burdick, 1978)
.036
>.001
<. 145
.061
<. 145
<. 145
<. 145
-
Approved For Release- 2000/08/078 CIARDI96-097878000500390001-6
.. .
.
.
..
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6
.
.

1
'Approved For Release 2000/08/07 : CIA-RDP96-00787P00050089000Remote Perception
29
Table 8
1
Ranks Assigned by Judges in Protocol #6
Target
Rank Assigned
Salt Water Lagoon.
4
Cafe Pollito
1
Lobby of Princess Hotel
1
River Boat at Disney World Village
3
2
Seaworld
1
Riding on expressway
5
Sum of Ranks
16
p value (one-tailed)
(Morris, 1972)
.20
p value (one-tailed)
(Solfvin, Kelly &
Burdick, 1978)
.145
·
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6

Approved For Release 2000/08/07 : CIA-RDP96-00787500050039000Gote Perception
· 30
Table 9
Ranks Assigned by Judges in Protocol #7
Target
Rank Assigned
River Boat at Disney World Village (retro)
Seaworld
1
Hotel room (no target)
6
Carib Hotel deck
2
Riding on expressway
2
Gatorland
3
Sum of ranks
15
(a) 9
(b) 8
p value (one-tailed)
.10
(Morris, 1972)
p value (one-tailed)
(Solfvin, Kelly &
Burdick, 1978)
.097
.006
.007
(a) Excluding control trial
-
(b) Excluding control and retrocognitive trials
Approved For Release 2000/08/07 : CIA-RDP96-00787R000500390001-6