PROPOSED GRILL FLAME PROTOCOL: TASK II (S), PROPOSED SRI INTERNATIONAL PROTOCOL FOR RESEARCH ON REMOTE PERTURBATION TECHNIQUES (S)

• Document Type: CREST
• Collection: STARGATE
• Document Number (FOIA) /ESDN (CREST): CIA-RDP96-00788R001300110001-6
• Release Decision: RIFPUB
• Original Classification: S
• Document Page Count: 20
• Document Creation Date: November 4, 2016
• Document Release Date: June 18, 1998
• Sequence Number: 1
• Publication Date: November 15, 1979
• Content Type: REPORT

OR= j C; For Release 2000/08/07 7888001300110001-6
PROPOSED GRILL FLAME PROTOCOL: TASK II (S)
PROPOSED SRI INTERNATIONAL PROTOCOL FOR RESEARCH
ON REMOTE PERTURBATION TECHNIQUES (S)
CLASSIFIED BY: Msg, 7 Jul 78,
DA(DAMI-ISH), Wash., D.C.
REVIEW ON: 15 November 1999
This document consists
of 19 pages.
Copy No. /p
ed For Rel~l/~ I~~(A1 001-6
 Approved For Release 2000/08/07: I 788R001300110001-6
REMOTE PERTURBATION TECHNIQUES: PROJECT DESCRIPTION AND
EXPERIMENTAL PROTOCOL (U) . . . . . . . . . . . . . . . . . . . . . 1
A. Background (U) . . . . . . . . . . . . . . . . . . . . . 1
B. Scientific Merit (U) . . . . . . . . . . . . . . . . . . 2
C. Objective (U) . . . . . . . . . . . . . . . . . . . . . . 2
D. Plan (U) . . . . . . . . . . . . . . . . . . . . . . . . 2
E. Participants (U) . . . . . . . . . . . . . . . . . . . . 10
F. Facilities (U) . . . . . . . . . . . . . . . . . . . . . 12
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
 Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
(S) REMOTE PERTURBATION TECHNIQUES:
PROJECT DESCRIPTION AND EXPERIMENTAL PROTOCOL (U)
(S) There have been many reported accounts of phenomena variously
known as telekinesis, psychokinesis, teleportation, etc. Most of these
accounts are clearly derived from carefully staged tricks which are re-
vealed whenever they are studied under controlled and well recorded condi-
tions. There are a few, however, which describe serious research by
reputable investigators. Included among these are experiments in which
the subject attempts to perturb, by mental processes alone, the outcome
of an otherwise random event. This kind of remote perturbation experiment
is appealing in that it involves no subjective interpretation--the results
may be expressed entirely in probabilistic terms. Reference 2 contains a
summary description of these experiments. As an overall evaluation of this
data base, it is unlikely that the apparent RP effect is simply an artifact
of selected reporting by the laboratories involved; even if one were to
assume that there were ten unreported nonsignificant experiments for each
reported significant one, the entire expanded data base would still show
significant effects with odds against chance of better than 2000:1.
(S) There are, however, two characteristics of this data base which
pose problems. First, the effects are rarely stable with one individual's
RP effort, the quoted results being averages over a number of individuals.
Secondly, the physical environment of the random event sources and associated
electronics was not discussed in any detail for any of the experiments, so
it is possible that some of the effects may be the result of normal and
possibly subtle electronic interference.
r% r, 9r
9% an-w-
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
 Approved For Release 2000/08/07 : 788R001300110001-6
(S) Scientists have been slow to admit that such phenomena as RP may
be consistent with the formalism of established science. The interpreta-
tion of quantum physics is flexible enough to encompass a number of seeming
paradoxes and some scientists have suggested quantum mechanisms to explain
RP without violating the basic conservation theorems of physics. Mattuck1
describes one such mechanism and shows that effects on a macroscopic scale
are consistent with the theory. While these hypotheses do not prove the
existence of RP phenomena, when coupled with the findings discussed in
Reference 2 they do strongly indicate that further research is appropriate.
(S) This experiment is designed to replicate the type of experiments
described in Reference 2 under more rigorously controlled conditions in
order to evaluate the claims of having demonstrated the RP phenomenon.
(S) Expressed in simplest terms, the experiment may be described as
(1) (S) A truly random sequence of Os and is (binary
sequence) is generated by processing and sampling the
behavior of a noise source or random event generator
(REG).
(2) (S) The RP participant is instructed to bias the
composition of the sequence in favor of is or Os by
exercising mental processes.
(3) (S) Feedback is provided the participant, so that he
can instantaneously be aware of the degree to which he
is succeeding.
(4) (S) The associated computer equipment simultaneously
carries out a statistical analysis of the sequence of
digits until it can be deemed perturbed (biased as
instructed) or not whereupon the trial (or test) is ended.
~&E$RE9F
.r Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
 Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
(S) The testing phase of the experiment consists of a large number
of such tests by each participant.
(S) The basic elements of the experiment are:
a. (S) Sources of Randomness: Two sources for genuinely
random binary sequences are to be employed as well as one pseudo-random
binary sequence generator. The radioactive decay of prometheum 147 (147PR),
occurs by the emission of beta particles which are readily detected. The
other random event generator is an electronic noise diode which is subject
to avalanche discharge which is also easily detected electronically. The
theory of both these devices and the distribution of random events ($-
emission or avalanche discharge) is well understood and has been well
verified experimentally.a,4 The random events are Poisson distributed in
time. Each time a random event occurs, it will trigger a change in a
flip-flop device--from one to zero or from zero to one depending upon the
state upon detection of the random event. The flip-flop device thus
randomly switches back and forth between zero and one with the occurrence
of random events. This device is sampled periodically (at intervals much
longer than the average event rate) to produce the desired random binary
sequence.
(S) In addition, a pseudo-random binary sequence will also be utilized
as needed. This differs from a random sequence in that, although the numbers
appear to be random, they are actually calculated by the computer from a
starting number or "seed" and thus for any given seed are deterministic.
If the RP phenomenon should exist, a great deal can be determined about
the nature of the phenomenon by the difference to which truly random and
deterministic events are affected by it.
b. (S) Analysis Capability: The random (or pseudo-random)
binary sequence must be statistically analyzed in order to determine if it
is actually random or is biased. Computational capability is also used to
.r Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
 Approved For Release 2000/08/07 4-WP 88R001300110001-6
fto in NO I% so V__
(S)
govern the sampling process described above, to drive the displays and to
catalog and store certain results of the tests. The analysis is carried
out in real time in order to provide effective feedback to the RP participant.
c. (S) Feedback Display: In order that the RP participant can
benefit from almost instantaneous feedback, he will be provided with a
color video display driven by the output from the statistical analysis of
the binary sequence. A variety of preprogrammed displays will be made
available including simple clock motion or 'race-car's displacement as
indicators of perturbation as well as more technical ones such as the
progress of the sequential analysis (see Reference 2). An auditory feed-
back channel will also be provided wherein the frequency or intensity of
a tone (or complex waveform) varies to indicate perturbation. The partici-
pant is expected to choose the particular feedback mode which he feels is
most effective.
(S) It has been tacitly assumed that any RP effect would act upon
the random event generator. This assumption will be checked by recording
the incidence of random events, the resulting binary sequence, and the
output of the statistical analysis on magnetic disc storage for subsequent
stage-to-stage analysis. The analysis of this for consistency along with
the recorded output of the feedback channel (video or audio tape) will
isolate the source of any perturbation as being in the random source, the
computer equipment or the display equipment.
(S) We intend to address the two problems with previous experiments
by first focusing our attention upon a limited number (eight or less) of
participants who have shown previous expertise as remote viewers. By
using experienced remote viewers, it is anticipated that we should be able
to observe an effect within this limited number, rather than having to
average over large numbers of individuals as in the data base experiments.
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
 Approved For Release 2000/08/07 ? 0788RO01300110001-6
(S)
Also, by using more sensitive analysis techniques than have been used
previously, even a small effect can be stabilized.
(S) Secondly, to assure ourselves that the noise sources are suffi-
ciently free of even subtle (but normal) electronic nonrandomness, we
intend to use the most rigorous construction and design techniques possible
(battery power, optically coupled signals, etc.) to isolate the sources
from normal environmental influences. Furthermore, the noise sources
will be chosen for their internal simplicitya and thus may be amenable to
realistic mathematical modelling. Using the models, we are able to calcu-
late by Monte Carlo techniques a noise source's dependence on various
external and internal physical parameters.
(S) The experiment will be conducted in two phases. Phase I consists
of the assembly and checkout of the apparatus and exhaustive validation
testing. During the validation testing the entire system will be exercised
over a wide range of environmental and testing conditions to assume the
randomness of the binary sequence and the proper functioning of the computer
and displays. Appendix A provides further information on the validation
testing procedures.
(S) Phase II consists of the actual RP participant testing. There
are three primary independent variables that may be varied during the
course of the investigation. Two of them, the random source (beta decay,
noise diode, pseudo random) and the mode of feedback have been discussed.
The third, and prime independent variable is the presence or absence of an
RP participant. The final measure of significance of the experiment is
essentially whether a consistent difference can be discerned in the randomness
of the binary sequence when the apparatus is operating with and without an
RP participant. The dependent variable, provided by the statistical analy-
sis of the binary sequence, is a quantitative measure of the degree to
which the composition of the sequence differs from what would be expected
Iff. .0% It"
16 NEO
400 Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
 Approved For Release 2000/08/07: CIA-RDP96- 788R001300110001-6
(S)
by chance. (Details of the statistical analysis are provided in Reference
2). For individual trials, the criterion for success will be that the
odds against chance expectation be greater than 20:1. The first four
months of the investigation will be used as a pilot period during which
participants will familiarize themselves with the experiment and explore
various techniques in a learning mode. This period will also be used to
optimize the device parameters.
(S) During the final two months of the investigation, each partici-
pant will be asked to contribute 100 trials. The following is a typical
succession of events during an RP trial session:
? The participant and experimenter will discuss the program
status to date. If the participant feels that set or
circumstance are unfavorable to his participation, the
session will be postponed.
? The variables or other changes in the experiment will be
described to the participant to whatever extent desired.
? The participant will select whatever mode of feedback he
prefers for the trials in the forthcoming session.
? The participant initiates successive trials at his own pace
at the instant of his choosing until the (nominally) 30-
minute session is concluded or sooner if the participant
wishes.
? The participant is then debriefed. He is told of the
outcomes of the trials, the statistical significance or
lack thereof and given the answers to any other questions
concerning the session or his cumulative performance.
(U) At no time will any participant be exposed to radiation or be
instrumented for monitoring of any physiological function--even by remote
sensors. No drugs, hypnosis, special sensory or proprioceptive stimuli,
liminal, or subliminal, electrical, or electromagnetic, will be used.
(U) After a participant has contributed one hundred trials, we will
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
 Approved For Release 2000/08/07 : 9AERW
(U)
greater than 20:1. If this total number of runs is greater than ten (the
number required by exact binomial calculation to meet odds against chance
of greater than 20:1) then we will declare that participant to have a
significant result, To assess whether the entire investigation is sig-
nificant, we shall combine the results of all participants using standard
statistical procedures.5
(S) One primary consideration in the planning of the experiment is
assuring the credibility of the findings. If positive results are reported,
critics will challenge virtually every aspect of the test and data analysis.
The following measures are being taken to reduce the vulnerability of the
experiment to attack.
a. (S) Safeguards Against Deception: At no time will the RP
participant have access to the equipment room in which the computer and
random source hardware is kept. All raw data are recorded, back up copies
made and stored each day in a secure GSA approved safe after each session.
Generally, the last safeguard is to examine the results of the completely
separate experiment to be executed by the contracting agency.
b. (S) Validity of Data Sample: All data taken in Phase II
will be saved and made available for further analysis if the data analysis
techniques should be questioned. No data will be discarded. If any part
of the data taken subsequent to the four-month pilot period is not included
in the combined statistical assessment, it will be so identified. Any
circumstances during experimentation which are out of the ordinary will be
recorded.
C. (S) Validity of Analysis Methodology: After Phase I vali-
dation of the functioning of the apparatus, the entire system will be
exercised using artificially biased binary sequences (Monte Carlo testing)
to the extent that the probability of errors of the first and second kind
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
 Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
W% 11
(5)
are accurately known. These probabilities can also be calculated prob-
abilistically to provide an independent check. Basically we are testing
the hypothesis that the binary sequence is composed of 50% zeros and 50%
ones. An "error of the second kind" consists of rejecting this hypothesis
when it is true; an "error of the first kind" consists of accepting the
hypothesis when it is false. The previously cited criteria for each trial
of judging the sequence to be perturbed if the odds against its occurrence
by chance are greater than 20:1 correspond to a probability of error of
the first kind of 0.05.
d. (S) Influence of Uncontrolled Parameters: It can be
hypothesized that such uncontrolled factors as solar storms, cosmic radia-
tion, the phases of the moon, etc., may affect the experiment. In some
cases, the threshold values at which the level of the parameter might be
significant can be determined. In general, however, we rely on two safe-
guards to minimize any effect due to uncontrolled parameters. During
validation testing and before and after trial sessions, the experimental
apparatus will be exercised and any anomalous behavior recorded. Second,
the chief independent variable--presence or absence of RP participant--should
be uncorrelated with the uncontrolled parameters, hence a clear dependence
of any measured RP effect upon presence of a participant can be attributed
to the participant. In other words, it would be highly improbable that
any uncontrolled parameters would be effective only when the RP participant
is present.
e. (S) Interpretation of Results: This experiment is an attempt
to determine if remote perturbation as defined exists. The nature of the
experiment was suggested by earlier research which has claimed some degree
of success. The scope of this project does not permit much extension in
the list of independent variables. As a result, such potentially important
variables as shielding, device/participant separation or the simultaneous
use of multiple participants cannot be investigated.
..r Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
 Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
A" Eft 90 X
(S) It has been assumed that if any RP effect is detected it is due
to the designated participant. Obviously, there is a large number of
individuals who might be in a position to exert an RP influence on the
device, in particular, the experimenter. No attempt has been made in the
design of the experiment to isolate the agent of any RP influence.
(S) At the end of the current program, the result can be characterized
(1) Positive: The experiments meet or surpass the overall
criteria stated on page 5.
(2) Negative: The experiments do not meet the stated criteria
for the entire experiment or the overall performance of a
single subject.
(3) Indeterminate: Anything other than positive or negative
as defined above.
(S) If the result is positive, we suggest that further experiments
be undertaken in a follow on program to determine the limitations of the
phenomenon for military exploitation. Such experiments would include an
assessment of the effects of distance and shielding, possible interference
from other physical phenomena and the potential for refinement for greater
consistency or sharper definition. It is also likely that some experimen-
tation would be undertaken to attempt to determine the mechanism for RP
and the physiological source.
(S) If the result is negative as defined above, we would suggest
that an additional experimental phase be added on only if there were certain
well characterized portions of the data showing effects which are quite
unlikely to have occurred by chance alone. This additional effort would
be used to ascertain what conditions (sampling rates, feedback, etc.),
prompted the unlikely results.
- Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
 Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
? If only one of the experiments is positive, an additional
phase of testing might be undertaken with an exchange of
certain critical hardware as well as other modifications
which might be indicated by an analysis of the results.
? If one or two participants show a significant RP capability
as defined on page 5, further experiments might be con-
sidered to determine if the capability can be transferred,
or detected to be present in enough individuals for military
consideration.
? If any observed effect seems to be inconstant--on again,
off again--the data will be analyzed to seek correlations
which might explain the fluctuation. A limited number of
additional tests might be necessary to validate possible
causes.
E. Participants (U)
(S) Some relation between remote perturbation and remote viewing might
be expected since in ordinary physical interactions information about the
state of a system (as in RV) can be obtained only by way of some inter-
action with the system, which in some cases may be supplied by the observer.
We might therefore expect an observable dependence between RV and RP under
certain conditions. For this reason, we plan to seek only experienced
remote viewers to participate in this investigation, and we will work with
no more than eight individuals. All participants are to be in general
good health, as determined by a standard employment physical examination.
(S) SRI International has individuals who are presently, or have
past experience in, participating in remote viewing experiments. These
participants are consultants or members of the SRI staff and will work
if selected in a contractual arrangement. Only,those individuals who
indicate a positive desire to participate, after familiarization with the
remote perturbation experiments and procedures, will be accepted into the
RP program. The information and consent form to be used is included in
Reference 2. Some of these participants may have had some familiarity
~RoEf
aw Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
 Approved For Release 2000/08/07 : - - 788RO01300110001-6
(S)
with RP experiments of a similar nature. The difference in levels of
naivete among the RP participants will be offset by the oral and written
briefing material (see Reference 2) and by the four-month pilot period.
(S) The total duration of Phase II of the investigation is six
months. Experiment sessions will take place during the normal work week
and will not exceed 1/2 hour length each. There will be no more than two
sessions per day, one in the morning and one in the afternoon. The total
number of trials by any individual during the pilot period and the ultimate
contribution of 100 trials will not exceed 300. This should require
approximately 30 half-hour sessions from each participant.
(U) A medical doctor will be briefed on the program and will be
available during all periods of experimentation on a standby basis.
Reference 2 describes the available medical facilities, the procedures
employed, and insurance coverage applicable to the various participants.
? Session Debriefing: At the end of each experimental
session, the subject will be given all the data available
pertaining to that group of trials. Any statistical
significance or lack thereof will also be clearly explained
to him.
? Experiment Debriefing: At the conclusion of experimental
testing, a final unclassified report will be prepared,
summarizing all results from the experiment, together with
any conclusions or scientific findings that may have come
out of the study. The report will be given to each par-
ticipating subject. Finally, any remaining questions that
the subjects may have about the experiment will be answered.
Specific Debriefing Protocols: We do not anticipate that
participation in these experiments will have any effect on
the day-to-day life of the subjects. This study is a purely
intellectual activity, and we believe that the debriefing
provided by the final technical report will be a suitable
termination of the experiment for the subject.
OW re Am%
41F IM %0
no Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
 Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
(U) The SRI experiment is to take place in the Building 44 laboratory.
(S) The room to be used by the RP participants for this work is a
comfortable, carpeted, air-conditioned environment. It is lit by a combi-
nation of fluorescent fixtures in the ceiling and incandescent table lamps.
There is a couch, an easy chair, and two tables. The computer graphics
terminal stands on one of these tables. The participant will be seated
on a conventional reclining swivel chair.
(S) In an adjacent room, the random event generators, computational
equipment, programmable terminals and associated peripheral equipment will
be installed. This room may be secured independently from the rest of the
facility.
Approved For Release 2000/08/07 :t I - !~ 0788RO01300110001-6
 Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
1. R. D. Mattuck, "Thermal Noise Theory of Psychokinesis: Modified
Walker Model with Pulsed Information in Rote," Journal of Psycho-
energetic Systems, August 1977.
2. "Proposed Grill Flame Protocol: Task II (S)," "Proposed SRI
International Protocol for Research on Remote Perturbation Techniques
(S)," SRI International, Menlo Park, CA (8 May 1979).
3. R. H. Haitz, "Controlled Noise Generation with Avalanch Diodes,"
IEEE Transactions on Electron Devices, Vol. ED-13, No. 3, pp. 342-346
(March 1966).
4. P. C. Stevenson, NAS-NS-3109, "Processing of Counting Data," National
Academy of Sciences/Nuclear Regulatory Commission Scientific Series
on Radiochemical Techniques (May 1966).
5. R. Rosenthal, "Combining Results of Independent Studies," Psychological
Bulletin, 1978, Vol. 85, No. 1, pp. 185-193.
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
 Approved For Release 2000LIVJCAAQS>C67PM1300110001-6
UNCLASSIFIED
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
 Approved For Release 2000U/08/07: L A S~S6i ff31300110001-6
Sub-System Testing
(1) Measure filtered and raw power spectra, and filtered pulse
height distribution as a function of the following
physical parameters:
(b) Reverse current of 100 u
amp (manufacturers
recommendation)
? Temperature -40? to +40? C
6000 gauss (parallel and
perpendicular to the diode
junction
? Low intensity gamma 1.33 MeV
ray irradiation
(2) Measure filtered and raw power spectra, and filtered pulse
height distribution at 100 p amps reverse current at 20?C
with the diode and temperature sensor mounted in its final
package.
(1) With detector and source mounted in the final assembly,
confirm manufacturer's noise specifications, and measure
the 147Pm beta decay pulse height spectrum and compare with
the known spectrum.
(2) Confirm that amplifier and pulse shaping system ICs are
functioning per manufacturer's specification.
? Pseudo random generator
(1) Generate the entire number set (28-1) and compare with
expected number set.
(1) Using tested computer algorithms for random number generation,
Monte-Carlo test the sequential analysis formalism.
UNCLASSIFIED
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
 Approved For Release 2000/U/NCIf-M--f (1!b1300110001-6
(1) As there appears to be no evidence to the contrary, we will
assume that the manufacturer's specifications of mean-time-
between-failure to be an approximate measure of these sub-
systems' susceptibility to possible environmental factors
such as geophysical phenomena and cosmic rays.
(2) We shall consider what thresholds must be exceded before
extreme environmental factors such as gamma irradiation
become important.
(3) These sub-systems must, of course, meet the usual
acceptance criteria applied to such apparatus before they
are declared operational.
UNCLASSIFIED
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
 Approved For Release 2000/CtM CJL-FAF.950VPIEU300110001-6
Completed System Testing
(1) Each of the "true" random sources will be subjected
to at least the following statistical tests.
(a) Frequency test
(b) FFT tests
(c) Serial test
(d) Gap tests
(e) Yule test (5-digit sums)
(f) Autocorrelations test
(g) Conditional bit tests
(h) D2 test
(i) Runs-of-length-n test
(j) Sum-of-n test
(k) Maximum minimum-of-n test
(2) The pseudo random generator must produce the expected
sequence.
(3) All generators will be used individually in Monte-Carlo
confirmations of the sequential analysis formalism.
(4) Finally, the various procedures to be used in Phase II
will be simulated with extensive Monte Carlo runs.
UNCLASSIFIED
.0 Approved For Release 2000/08/07 : CIA-RDP96-00788R001300110001-6
 Approved For Release 2000/08/07 CIA-R P96-00788R001300110001-6
?-- NOT RELEASABLE TO FOREIGN NATIONALS --
Approved For Release 2000/08/07 : -CIA RDPY96=W788R001300110001-6