SENSING OF REMOTE EM SOURCES (PHYSIOLOGICAL CORRELATES)

• Document Type: CREST
• Collection: STARGATE
• Document Number (FOIA) /ESDN (CREST): CIA-RDP96-00788R001300350001-0
• Release Decision: RIFPUB
• Original Classification: U
• Document Page Count: 44
• Document Creation Date: November 4, 2016
• Document Release Date: December 7, 1998
• Sequence Number: 1
• Publication Date: April 1, 1978
• Content Type: REPORT

Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
6).03
Final Report
SENSING OF REMOTE EM SOURCES
(PHYSIOLOGICAL CORRELATES)
By: RUSSELL TARG EDWIN MAY
SRI International, Menlo Park, California
HAROLD PUTHOFF
DAVID GALIN ROBERT ORNSTEIN
Langley Porter Neuropsychiatric Institute, San Francisco, California
Prepared for:
NAVAL ELECTRONICS SYSTEMS COMMAND
WASHINGTON, D.C. 20360
Attention: MR. JAMES FOOTE
ELEX 03X
CONTRACT N00039-76-C-0077
333 Ravenswood Avenue
Menlo Park, California 94025 U.S.A.
[phone]
Cable: STANRES, Menlo Park
TWX: 910-373-1246
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Final Report
Covering the Period November 1975 to October 1976
SENSING OF REMOTE EM SOURCES
(PHYSIOLOGICAL CORRELATES)
By: RUSSELL TARG EDWIN MAY
SRI International, Menlo Park, California
HAROLD PUTHOFF
DAVID GALIN ROBERT ORNSTEIN
Langley Porter Neuropsychiatric Institute, San Francisco, California
Prepared for:
NAVAL ELECTRONICS SYSTEMS COMMAND
WASHINGTON, D.C. 20360
Attention: MR. JAMES FOOTE
ELEX 03X
CONTRACT N00039-76-C-0077
SRI Project 4540
Approved by:
ROBERT S. LEONARD, Director
Radio Physics Laboratory
DAVID D. ELLIOTT, Executive Director
Systems Research and Analysis Division
333 Ravenswood Avenue ? Menlo Park, California 94025 ? U.S.A.
For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
UNCLASSIFIED
SECURITY CLASSIFICATION OF THIS PAGE (When Data Entered)
REPORT DOCUMENTATION PAGE
READ INSTRUCTIONS
BEFORE COMPLETING FORM
1. REPORT NUMBER
2. GOVT ACCESSION NO.
3. RECIPIENT'S CATALOG NUMBER
4. TITLE (and Subtitle)
SENSING OF REMOTE EM SOURCES (PHYSIOLOGICAL
CORRELATES)
5. TYPE OF REPORT & PERIOD COVERED
Final Report
Covering the period November
1975 to October 1976
6. PERFORMING ORG. REPORT NUMBER
SRI Project 4540
7. AUTHOR(s)
Russell Targ Edwin May Harold Puthoff
David Galin Robert Ornstein
8. CONTRACT OR GRANT NUMBER(s)
Contract N00039-76-C-0077
9. PERFORMING ORGANIZATION NAME AND ADDRESS
SRI International
333 Ravenswood Avenue
Menlo Park, California 94025
10. PROGRAM ELEMENT, PROJECT, TASK
AREA & WORK UNIT NUMBERS
12. REPORT DATE
April 1978
13. NO. OF PAGES
46
11. CONTROLLING OFFICE NAME AND ADDRESS
.
ELEX 03X
Naval Electronics Systems Command
Washington, D.C. 20360
15. SECURITY CLASS, (of this report)
UNCLASSIFIED
14. MONITORING AGENCY NAME & ADDRESS (If diff. from Controlling Office)
15a. DECLASSIFICATION/DOWNGRADING
SCHEDULE
16. DISTRIBUTION STATEMENT (of this report)
17. DISTRIBUTION STATEMENT (of the abstract entered In Block 20, if different from report)
18. SUPPLEMENTARY NOTES
19. KEY WORDS (Continue on reverse side if necessary and identify by block number)
EEG Electrophysiology
Electroencephlograph Psychoenergetics
Brainwaves Subliminal perception
20. ABSTRACT (Continue on reverse side If necessary and identify by block number)
We have investigated the ability of certain individuals to perceive remote
(faint) stimuli at a noncognitive level of awareness. To investigate this we
have looked for systematic changes in a subject's brainwave (EEG) production
occurring at the same time as light flashes are generated on a random schedule
in a remote.laboratory. We have found in this investigation that statistically
significant correlations do appear to exist between the times of light flashes
and the times of brainwave alterations. However, we consider these data to be
DD, FrAINM731473
UNCLASSIFIED
SECURITY CLASSIFICATION OF THIS PAGE (When Data Entered)
EDITLON OF 1 NOV.65 IS.OBSOLETE
Approved ror Release 2000/08/07: CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
CONTENTS
LIST OF ILLUSTRATIONS
LIST OF TABLES vi
PART ONE: SRI ANALYSIS 1
I INTRODUCTION AND SUMMARY 2
II PILOT STUDY AT SRI 3
III TECHNICAL APPROACH 6
IV DISCUSSION 11
PART TWO: LANGLEY PORTER ANALYSIS 14
I INTRODUCTION 16
II REANALYSIS OF PREVIOUS DATA 17
III FIRST EXPERIMENT AT LANGLEY PORTER: ATTEMPTED REPLICATION
OF SRI STUDY 20
IV SECOND EXPERIMENT AT LANGLEY PORTER: ATTEMPTED
REFINEMENT 25
V DISCUSSION 32
APPENDIX: OPTICAL ISOLATION EXPERIMENTS AT LANGLEY PORTER . . . 34
REFERENCES 40
iv
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
ILLUSTRATIONS
PART ONE
1 Typical Power Spectrum Averaged Over Twenty 8-s
Epoch for One Subject 5
2 Schematic of the Remote-Sensing EEG Experiment 8
APPENDIX
A-1 Floor Plan and Apparatus for Light-Leakage
Measurements
V
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
37
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
TABLES
PART ONE
1 Summary of Results of the Replication Experiments
Showing Power Means and Statistical Results for the
Various Experimental Conditions
PART TWO
12
1
Reanalysis of Dr. Rebert's SRI Tapes Played Through
the Langley Porter EEG Alpha Analyzer
18
2
First Experiment at Langley Porter:
Means of Summed
EEG Alpha in Relative Power Units
25
3
Second Experiment at Langley Porter:
All Trials,
Sender vs No Sender
28
4
Second Experiment at Langley Porter:
Guess Trials
Only, Sender vs No Sender
29
5
Second Experiment at Langley Porter:
Pass Trial Means
30
6
Second Experiment at Langley Porter:
Correct Guess
Trial Means
31
vi
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
PART ONE: SRI ANALYSIS
Russell Targ
Edwin May
Harold Puthoff
SRI International
Menlo Park, California
1
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
I INTRODUCTION AND SUMMARY
A. Introduction
The objective of the program reported here was to investigate
the characteristics of, and, if possible, determine the mechanisms
responsible for, the coupling of remote electromagnetic stimuli to human
detection modalities, utilizing physiological response (specifically,
EEG) as an indication of such registration.
In a number of laboratories evidence has been obtained indi-
cating the existence of an as-yet-unidentified channel wherein informa-
tion is coupled from remote electromagnetic stimuli to the human nervous
system, as indicated by physiological responses. This coupling can be
measured, even though overt subject responses such as verbalizations or
pressing of a key provide no evidence for such information transfer.
Physiological measures have included plethysmographic responsel and EEG
activity.213 Kamiya, Lindsley, Pribram, Silverman, Walter, and others
have suggested that a whole range of EEG responses such as evoked poten-
tials (EPs), spontaneous EEG, and the contingent negative variation (CNV)
might be sensitive indicators of the detection of remote stimuli not
mediated by usual sensory processes.4
A pilot study was therefore undertaken at SRI to determine
whether EEG activity could be used as a reliable indicator of informa-
tion transmission between an isolated subject and a remote stimulus.
Following earlier work of others, we assumed that perception could be
indicated by such a measure even in the absence of verbal or other overt
indicators.
To aid in selecting a stimulus, we noted that Silverman and
Buchsbaum attempted, without success, to detect EP changes in a subject
in response to a single stroboscopic flash stimulus per trial observed
by another subject.5 Kamiya suggested that because of the unknown
temporal characteristic of the information channel, it might be more
References are listed at the end of this report.
2
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
appropriate to use repetitive bursts of light to increase the probability
of detecting information transfer.9 Therefore, in our study we chose to
use repetitive light bursts as stimuli.7-9
B. Summary
In Section II we describe the pilot study conducted at SRI,
which led to the present investigation. It was in this study that we
obtained our first indication that remote light-flash stimuli could
cause systematic changes in the observed EEG production of an individual
shielded against direct perception of the light.
Section III presents the technical approach employed in the
most recent investigations. This includes a description of the battery-
operated strobe light equipment, shielding precautions, experimental
protocol, and finally data analysis.
Section IV presents a detailed discussion of the results from
all phases of the study.
Part Two of this report is the technical evaluation prepared
for us by our independent consultants at Langley Porter Neuropsychiatric
Institute. In their report they analyze the present investigation from
the point of view of practicing perceptual psychologists.
In the Appendix to this report we describe the optical measure-
ments made to determine the extent of possible light leakage from the
stimulus roam to the subject room.
II PILOT STUDY AT SRI
In the design of the study it was assumed that the application of
remote stimuli might result in responses similar to those obtained
under conditions of direct stimulation. For example, when normal sub-
jects are stimulated with a flashing light, their EEG typically shows a
decrease in the amplitude of the resting rhythm and a driving of the
brain waves at the frequency of the flashes.1? We hypothesized that if
we stimulated one subject in this manner (a putative sender), the EEG
of another subject in a remote room with no flash present (a receiver)
3
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
might show changes in alpha (8 to 13 Hz) activity, and possibly EEG
driving similar to that of the sender, either by means of coupling to
the sender's EEG, or by coupling directly to the stimulus.
We informed our subject that at certain times a light was to be
flashed in a sender's eyes in a distant room, and if the subject per-
ceived that event, consciously or unconsciously, it might be evident
from changes in his EEG output. The receiver was seated in a visually
opaque, acoustically and electrically shielded double-walled steel room
located approximately 7 m from the sender's room.
We initially worked with four female and two male volunteer sub-
jects. These were designated "receivers." The senders were either
other subjects or the experimenters. We decided beforehand to run one
or two sessions of 36 trials with each subject in this selection pro-
cedure, and to do a more extensive study with any subject whose results
were positive.
A Grass PS-2 photostimulator placed about 1 m in front of the sender
was used to present flash trains of 10 s duration. The receiver's EEG
activity from the occipital region (Oz), referenced to linked mastoids,
was amplified with a Grass 5P-1 preamplifier and associated driver ampli-
fier with a bandpass of 1 to 120 Hz. The EEG data were recorded on
magnetic tape with an Ampex SP300 recorder.
On each trial, a tone burst of fixed frequency was presented to both
sender and receiver and was followed in one second by either a 10-s train
of flashes or a null flash interval presented to the sender. Thirty-six
such trials were given in an experimental session, consisting of 12 null
trials--no flashes following the tone--12 trials of flashes at 6 flashes
per second (fps) and 12 trials of flashes at 16 fps, all randomly inter-
mixed, determined by entries from a table of random numbers. Each of
the trials consisted of an 11-s EEG epoch. The last 4 s of the epoch
were selected for analysis to minimize the desynchronizing action of the
warning cue. This 4-s segment was subjected to Fourier analysis on a
LINC 8 computer.
4
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Spectrum analyses gave no evidence of EEG driving in any receiver,
although in control runs the receivers did exhibit driving when physi-
cally stimulated with the flashes. But of the six subjects studied
initially, one subject showed a consistent alpha blocking effect. We
therefore undertook further study with this subject. We note that of
our six subjects, this one had the most monochromatic EEG spectrum.
Figure 1 shows a typical occipital EEG spectrum of this subject.
0 2 4 - 6
8
10
FREQUENCY - Hz
12
14
16
FIGURE 1 TYPICAL POWER SPECTRUM AVERAGED OVER TWENTY
8-SECOND EPOCHS FOR ONE SUBJECT
Data from seven sets of 36 trials each were collected from this
subject on three separate days. This comprised all the data collected
to date with this subject under the test conditions described above.
The alpha band was identified from average spectra; then scores of
average power and peak power were obtained from individual trials and
subjected to statistical analysis. The final analysis showed that power
values were less in the 16-fps case than in the 0-fps case (no light) in
all seven sets of peak power measurements and in six out of seven
average-power measurements.
Siegel's two-tailed t approximation to the nonparametric randomi-
zation test" was applied to the data from all sets, which included two
5
Approved For Release 2000/08/07: CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
sessions in which the sender was removed. Average power during trials
associated with the occurrence of 16 fps was significantly less than
when there were no flashes (t = 2.09, df = 118, p < 0.04). The second
measure, peak power, was also significantly less in the 16-fps conditions
than in the null condition (t = 2.16, df = 118, p < 0.03). The response
in the 6-fps condition was in the same direction as that associated with
16 fps, but the effect was not statistically significant.
As part of the experimental protocol for each trial, the subject
was asked to indicate conscious assessment as to which stimulus was
generated. The guess was registered by the subject via one-way tele-
graphic communication. An analysis of these guesses has shown them to
be at chance, indicating the absence of any supraliminal cueing, so
arousal as evidenced by significant alpha blocking occurred only at the
noncognitive level of awareness.
Several control procedures were undertaken to determine if these
results were produced by system artifacts or by subtle cueing of the
subject. Low-level recordings were made from saline of 12-kilohm
resistance in place of the subject, with and without the introduction
of 10-Hz, 50-?V signals from a battery-operated generator. The standard
experimental protocol was adhered to and spectral analysis of the results
were carried out. There was no evidence in the spectra associated with
the flash frequencies, and the 10-Hz signal was not perturbed.
In another control procedure a five-foot pair of leads was draped
across the subject's chair (subject absent). The leads were connected
to one input of a C.A.T. 400C "averager." Two-second sweeps, triggered
at onset of the tone, were taken once every 13 s for approximately two
hours, for about 550 samples. No difference in noise level between the
fore-period and the onset of flicker was observed.
III TECHNICAL APPROACH
A. Replication Studies at Langley Porter
The main effort of the program was directed toward replication,
by an independent laboratory, of the original SRI study of EEG response
6
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
to remote strobe-light stimuli. Arrangements for replication were made
with the Langley Porter Neuropsychiatric Institute, University of
California Medical Center San Francisco. Dr. David Galin and
Dr. Robert Ornstein were our principal consultants.
As a special precaution against the possibility of system
artifacts in the form of electromagnetic pickup from the strobe-light
discharge or associated electronic equipment (e.g., through the power
lines), SRI developed an entirely-battery-operated package for use as a
stimulus generator for the EEG experimentation. It consists of a
battery-driven incandescent lamp, whose CW output passes through a
mechanical chopper continuously driven by a battery-powered motor as
shown in Figure 2. A 10-Hz timing generator (computer triggered) con-
trols the generation of a 1-kHz warning tone 2 s before onset of the
experimental period, and also drives a locking circuit that determines
the presence or absence of the 10-s light stimulus, again all battery
operated. Thus everything on the left of the diagram of Figure 2 is
battery operated and therefore independent of the power-line system.
Further, the arc-discharge strobe lamp was replaced by an incandescent
lamp to eliminate the possibility of direct subliminal pickup of audio
or electrical signals from possible transients associated with the arc
discharge or associated electronics.
Finally, the possibility of artifact due to light-leakage
between the lamp (stimulus) and subject (receiver) was investigated by
the use of a photomultiplier in conjunction with a phase-sensitive
detection system. The results, presented in detail in the Appendix,
indicated no light leakage between the rooms in question to the limit of
our measurement capability (125 dB attenuation).
1. Description of the EEG Processor
A hardware single-channel power spectrum analyzer was
constructed from a commercial bandpass filter with corner frequencies
of 9.0 and 12.0 Hz, and a roll-off 48 dB down at 8.0 and 13.0 Hz. Analog
multipliers convert the filter output to a signal proportional to in-band
7
Approved For Release 2000/08/07: CIA-RDP96-00788R001300350001-0
0-1.0009?00?1.00N99/00-96dC1N-V10 : /0/90/000Z eseeieN .10d peA0.1ddV
LIGHT CHOPPER MOTOR
RUNS CONTINUOUSLY
-e iiii-_
0
0 0
o
1 kHz TONE
GENERATOR
2 sec BEFORE
EACH TOTAL
PERIOD (Light,
or No Light)
LIGHT
e10 sec LOCKING
CIRCUIT CONTROLS
LAMP
BUFFER
AMP
10 Hz
TIMING
GENERATOR
TIMING
CHANNEL 5
EEG CHANNELS
1, 2, 3, 4
<
2 sec
WARNING
TONE
FIGURE 2 REMOTE SENSING EEG EXPERIMENT
SHIELDED ROOM
SA-4540-11
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
power. To confirm that this system is equivalent to the standard FFT
analysis used in the pilot study, the analog data on the entire pilot
study were reanalyzed, and the result was found to be consistent with
the earlier analysis.
2. Experimental Protocol
Each experimental session consisted of 40 trials, 20 each
for the 0-fps (no light) and the 16-fps case of the remote light stimulus.
A trial is defined as a warning tone followed by a 10-s period consisting
of a 2-s wait, and two 4-s data-collection periods. The trial rate was
one trial every 30 ? 1 s. The trial sequence was randomized subject to
the following conditions: (1) in each group of 10 trials there were
equal numbers of each condition, and (2) no more than three in a row of
a single type were allowed. Seven 40-trial sequences were made according
to this prescription and recorded separately on audio tape. During the
session, trials were generated from one of these tapes chosen blind to
the experimenters so that the sequence was unknown to them. As in
standard EEG protocol, and in accordance with preestablished criteria,
certain trials were deleted after the session for three reasons only:
artifact, logic circuit failure, or abnormal EEG power. If a trial was
rejected, a trial of the opposite stimulus condition was rejected at
random from the particular set of 10 trials in question. If more than
10 trials of a given type were rejected from a session, the entire
session was deleted. (This occurred twice in each of the two experiments
of the study, described below.)
Six channels of EEG and one logic channel taken from the
sequence tape were recorded on a multiplexed FM analog tape recorder.
The logic on the tape differentiated the trials as to flashing and
nonflashing conditions.
In pretesting the equipment, we ran the experiment using
unselected subjects such as laboratory personnel, in order to test the
adequacy of the experiment and to determine whether there were any
correlated electronic or mechanical discharges from the apparatus. In
9
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
20 sessions of data acquisition, of 40 each (800 trials) there were no
significant differences between the null and 16-Hz conditions. The
absence of differences can also be taken as a demonstration of the non-
responsiveness of the unselected subjects.
B. Results
Using the above protocol, two experiments were conducted in
the course of the study, the first consisting of twelve sessions, the
second, fourteen. For half of the twelve sessions in the initial experi-
ment, the subject was asked to press a button when she felt the light
was flashing. For the six sessions (105 trials each for the 0- and
16-fps conditions) when she was not asked to overtly indicate her feelings
about the light, there was a slight decrease of in-band EEG power measured
over the left occipital region of the brain. For the six sessions (107
trials each for the 0- and 16-fps conditions) when she was asked to
respond overtly, there was a significant decrease of in-band EEG power
(p s 0.037, using an F ratio test derived from a two-way analysis of
variance). In considering the experiment as consisting of the combined
212 trials in each stimulus condition regardless of the overt response
contingency, we find a statistically significant decrease in in-band
EEG power (p < 0.011, using F ratio test as above).
During the second experiment of fourteen sessions, three months
later, a different contingency was added to determine if a "sender" was
necessary to produce the effect we had observed earlier. For a given
session, a random procedure (with equal trials) was used to determine if
a person (called the "sender" person) would be looking at the photo-
simulator. There was no one present with the photo-stimulator otherwise.
For the 7 "non-sender" sessions (121 trials each for the 0- and 16-fps
conditions) we find a statistically significant increase of in-band EEG
power measured over the mid-occipital region of the brain (p < 0.039
using an F ratio test as above). During the 7 "sender" sessions (123
trials in each stimulus condition) there was a slight increase of in-
band EEG power. When all 244 trials were analyzed, regardless of
"sender" condition, there was a statistically significant increase of
10
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
in-band EEG power (p < 0.008 using an F ratio test as above), and there
was no significant difference found between "sender" and "no-sender"
conditions.
For both experiments, we considered in-band EEG power for the
0-to-4-s and 4-to-8-s time periods independently, to determine if the
effects were time dependent. Although some of these isolated subinter-
vals were statistically significant, no systematic relationship emerged.
Thus the effect appears to be cumulative over the 8 s. The 0-to-8-s
results are summarized in Table 1.
IV DISCUSSION
Although our pilot experiment and the two replication studies all
showed significant changes in EEG production correlated with the presence
or absence of a remote light stimulus, the sign of the systematic change
in power in the third study was opposite to that of the first two. We
therefore undertook a detailed frequency analysis of the EEG data tapes
from the last two experiments, since the pilot experiment had already
been subjected to fast-Fourier-transform (FFT) analysis. We conjectured
that the observed power change in these experiments might be the result
of a very small frequency shift, which could become translated into a
large amplitude change due to discriminator action of the alpha-band
filter. In a chapter on alpha blocking, Kooi, in his Fundamentals of
Electroencephalography says, for example, "...attentiveness is associated
with a reduction in amplitude and an increase in average frequency of
spontaneous cerebral potentials.... The center frequency of the alpha
rhythm may be influenced by the type of ongoing mental activity. Shifts
in frequency may be highly consistent as two different tasks are per-
formed alternately." As compared with the first experiment in which the
average peak EEG power fell midband, the FFT analysis for the second
experiment showed that the average peak EEG power occurred most often
near 8 Hz, and thus fell slightly below the hardware summing window
(?3 dB at 8.7-12.4 Hz), enhancing a possible discriminator effect. The
FFT analysis further showed that there was an overall increase in fre-
quency of peak power but the shift was statistically nonsignificant.
11
Approved For Release 2000/08/07: CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Table 1
SUMMARY OF RESULTS OF THE REPLICATION EXPERIMENTS SHOWING
POWER MEANS AND STATISTICAL RESULTS FOR THE VARIOUS EXPERIMENTAL CONDITIONS
Experiment I
Experiment II
Guessing
Sessions
Non-Guessing
Sessions
Combined
Sender
Sessions
Non-Sender
Sessions
Combined
No light flash
957
704
832
854
766
810
Light flash
873
647
761
860
844
852
F ratio
4.39
2.20
6.47
0.017
4.33
7.03
dfl' . df2
1; 202
1; 198
1; 400
1; 232
1; 228
1; 460
P
0.037
0.14
0.011
0.90
0.039
0.0083
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
This slight shift of 0.11 Hz could possibly account for the observed
power increase due to the highly nonlinear discriminator effects. In
examining other portions of the spectrum for further effects, we found
that systematic amplitude changes are highly dependent on the location
in the frequency spectrum from which the power sum is taken. This is to
be expected, since almost all EEG phenomena are known to be strongly
frequency dependent.
In the pilot study the frequency region for analysis was centered
about the subject's dominant EEG output frequency, with bandpass deter-
mined by the full-width ten-percent power points. In the two replication
studies we used hardware filters in this same frequency. FFT analysis
showed clearly that if other filter bands had been chosen, significant
correlations would not have been found in some cases. Thus, although
our filter selection was made before the collection of any data, other
experimenters might have reasonably chosen other criteria for frequency
selection. Therefore, although we have found statistically significant
evidence for EEG correlates to remote light flash stimuli in all three
experiments, we consider these data to be only suggestive, with a defini-
tive result requiring further experimentation. In particular, further
analysis techniques would have to be developed before any putative EEG
effect could be considered as part of an information transmission/
processing channel.
We wish to acknowledge the valuable contributions made to this
project by SRI consultant Mr. David B. Hurt, who designed and constructed
the battery powered light stimulator and logic elements used in the
latter two phases of the program.
13
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
PART TWO: LANGLEY PORTER ANALYSIS
14
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Final Report on:
EEG STUDIES OF RESPONSES TO REMOTE STIMULATION
SRI International Subcontract No. 14157
SRI Consultants
David Galin, M.D.
Robert Ornstein, Ph.D.
Langley Porter Institute
University of California Medical Center
San Francisco, California 94143
15
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
I INTRODUCTION
The study of perception of information despite shielding of all
known sensory channels would be greatly aided if a paradigm could be
devised that used a physiological response as the major dependent
variable. Such a paradigm would have the advantage of not requiring
verbal report or even requiring the subject's awareness of the perception.
Effects of remote stimulation might be too weak to enter a subject's
awareness but nevertheless be detectable as a change in some physiologi-
cal index, such as the EEG. This is the case with conventional sub-
liminal sensory stimulation and scalp evoked potentials.
Filtering responses to remote perception through conscious aware-
ness must add variability to such phenomena. Removing this variability
might allow the study of remote perception in a wider sample of subjects,
because weaker effects may be widespread in unselected subjects, while
only unusual or "psychic" subjects may be sufficiently sensitive for
the remote stimulation to enter awareness. Therefore a demonstration
of a reliable EEG effect would have far-reaching implications.
A paradigm that might measure EEG response to remote stimulation
was devised by Drs. C. Rebert and A. Turner in cooperation with Dr. H.
Puthoff and R. Targ of Stanford Research Institute. After screening
six potential subjects, they identified one who seemed to produce reli-
able results and showed a highly monochromatic occipital EEG spectrum.
When normal subjects are stimulated with a flashing light, their
EEG normally shows a decrease in the power of the alpha rhythm and a
driving at the frequency of the flashes (Hill and Parr, 1963). The SRI
group hypothesized that if they stimulated one subject in this manner
(a sender), the EEG of another subject in a remote room with no flash
present (a receiver) might show a decrease in alpha activity, and
possibly EEG driving similar to that of the sender.
The SRI group showed a significant decrease of occipital alpha
power in this single subject compared with control trials, in which the
light did not flash (Targ and Puthoff, 1974).
16
Approved For Release 2000/08/07: CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
We attempted to replicate their finding and to refine the technique.
This was done, first by reanalyzing data tapes of their results; second,
by attempting to replicate their results in our laboratory using the
same subject; and finally by attempting to isolate the variables necessary
to produce the phenomenon at will. We particularly examined whether it
was necessary to have an observer (sender) present with the light flash
stimulus and whether it was helpful to have the subject guess whether
the light was flashing or not. In our second (final) experiment, we also
took extreme precautions that all procedures were double-blinded, and
that all previously measured electrode sites were included.
II REANALYSIS OF PREVIOUS DATA
We played the EEG tapes of Hella Hammid, gathered by Dr. Rebert of
SRI, through our hard-wired alpha analysis system. The results of this
reanalysis are provided in Table 1. They show power means at Oz in the
null condition (no flash, 0 Hz) during the second 4-s epoch at t = 2.08,
p < 0.05, df = 119. This reanalysis confirmed the published effect and
also ensured the compatibility of our systems. This system was used in
all of the experiments as well as in the reanalysis of the SRI data.
The specifications and logic of our analyzer system are given below.
The amplified EEG signals are taken from the amplifier outputs and
lead to the alpha EEG filters. Up to six channels can be accommodated.
The filters were built by Kinetic Technology, Inc., of Mountain View,
California, to high specifications: Corner frequencies 9.0 and 12.0 Hz,
48 dB down at 8.0 and 13.0 Hz with rejection over the rest of the stop-
band greater than 30 dB, and pass-band ripple less than 0.2 dB p-p. The
filters also had a 30-to-60-mV dc offset. Therefore, a high-pass filter
with f = 8 Hz was designed that blocked the dc offset and satisfactorily
attenuated the delta contamination, comparing adequately with alpha data
generated by a fast-Fourier-transform program on our computer. Alpha
levels at the filter output are usually less than 400 mV. Operational
amplifiers invert and amplify the filtered alpha (gain = 50) to provide
17
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Table 1
REANALYSIS OF DR. REBERT'S TAPES
PLAYED THROUGH THE LANGLEY PORTER EEG ALPHA ANALYZER--
FEBRUARY 2 THROUGH 6, 1976
Analysis of Variance with Repeated Measures
Treatment
Session
x0 Hz
N
x16 Hz
N
?
1:1
578
12
490
11
3:1
465
12
397
11
2:1
553
12
466
12
2:3
232
12
174
15
2:4
308
12
236
12
Mean:
427.2
340.8
Results of t-Test
Langley Porter:
t =
2.08,
df = 119
p < 0.05
SRI:
t =
2.09,
df = 118
optimum (near maximum) input to the squared circuits. The alpha signals
are squared by analog multipliers (Analog Devices #533K) to yield instan-
taneous power, an approximation to the FFT computation. The transfer
function is X2/10, with a maximum of ?10 V input yielding +10 V output.
After this stage, the signal processing is commanded by a microprogrammed
controller, hard-wired in TTL logic, except for read-only memories (ROMs)
that control formatting in the digital printer. A master clock is
synchronized with the power line (60 Hz).
When the experimenter is ready to begin data acquisition, he selects
the summation time (1 to 99 s) with a two-digit thumbwheel switch and
pushes the START button. In our experiment, two 4-s intervals were
chosen. The START function resets the summing integrators, commands the
18
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
printer to print a line of special characters signifying the beginning
of the record, delivers a 50-ms pulse to the polygraph marker channel,
and connects the outputs of the six squaring units to the integrators.
Switching is handled by reed relays for lower leakage. Solid-state
switching devices were used initially, but leakage currents (10-9 A)
were too high for the accuracy and stability required. The integrators
make use of low-loss polystyrene capacitors, and FET-input amplifiers
with ultra low (5 X 10-12 A) input offset current. This design makes it
possible to use long summing intervals or interrupts with a drift error
of no more than 17,.
After the summation time has elapsed, the outputs of the six summing
integrators are sequentially connected to the analog-to-digital converter
(ADC) by the analog multiplexer. When each conversion is finished (10 bits
BCD) the data is parallel loaded into shift registers. The shift regis-
ters are then clocked by the controller to send the data in digit serial
form to the printer. This being completed, the next integrator is con-
nected to the ADC and the process repeats. This continues until all six
integrators have been read and the summed power of each EEG lead is
printed. The digital printer is a MC4000 Monroe Datalog. A fiberoptic
cathode ray tube exposes light-sensitive paper quickly and (most important
for our research) silently. All standard alphanumeric characters are
printable.
Next, the controller commands a line feed from the printer. The
second line of data for this sample consists of the log ratios of pairs
of integrators representing homologous pairs of brain electrodes. The
log of the ratio rather than the ratio is desired because it is linear
around zero--e.g., a ratio of 2/1 = 1/2 = ?0.301. The analog multiplexer
is then commanded to connect the first two integrators to the two inputs
of the log ratio module. Its output is an analog voltage representing
the log ratio of the two channels to a 10-bit BCD number. The printing
process is the same with the addition of a polarity bit indicating which
hemisphere has a higher output for the task.
19
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
The process is repeated for computation of log ratios of the other
two pairs of integrators. After the last data are printed, the controller
resets the integrators, then reconnects the squaring units to their inte-
grators. One count is added to the trial display register, which tells
the experimenter at a glance how many epochs have been collected. The
digitizing, computation, and printout takes 2 s, primarily due to the
switching speed of the relays.
III FIRST EXPERIMENT AT LANGLEY PORTER: ATTEMPTED REPLICATION OF
SRI STUDY
A. Method
1. Subject
The first experiment was conducted with a single subject,
Hella Hammid, who had been selected from an initial group of four female
and two male subjects tested by the SRI group. She was selected because
she had had the most consistent positive results in their study.
2. Setting and Procedure
The first experiment was conducted in our EEG laboratory
at Langley Porter Institute. The setting is diagrammed in Figure 2 in
Part One of this report. The "sender" sat before the photic stimulator
in a divided room separated by a curtain from the logic, equipment, and
monitors. The number of people in the room with the stimulator and the
"sender" varied from one to three. Noise from the street and hallway
also varied.
The "receiver" sat upright in an isolated sound-attenuated
darkened metal chamber located in a laboratory adjacent to the roam in
which the sender was located. The "receiver" (subject) made no overt
responses for Set 1, but was required to press a button to indicate her
guess of trial type about 12 s after the warning tone for Set 2. The
subject was familiar with the nature of the experiment, and was not
formally instructed for each session.
20
Approved For Release 2000/08/07: CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
The sender and receiver were both presented with a one-
second warning tone that signalled the beginning of a trial. This was
immediately followed by the light flashing for 10 s at 16 Hz in the
active condition, and by no flashes in the null condition. The stimu-
lation period was divided into 4-s segments for comparability with the
Rebert study; the first 4 s following flash onset (or null onset), and
the Second 4 s starting with the sixth second from onset. (The fifth
second was blocked from the analyzer by the print cycle, and we dis-
carded the tenth second.) A "set" was defined as at least 100 acceptable
trials of each type. After the first set, minor modifications (such as
a button to indicate guesses) were added, and a second set was run. The
total number of acceptable trials was 212 of both types, or 424 total.
In pretesting our equipment and procedure we used un-
selected subjects such as laboratory personnel, in order to test the
adequacy of the protocol and to determine whether there was any cor-
related electronic or mechanical interference from the apparatus. In
these pilot sessions of data acquisition there were no significant dif-
ferences between the null and 16-Hz conditions.
3. EEG Recording Procedure
Electrodes were placed at C3, C4, 01, 02, and Oz, all
referenced to Cz. The occipital leads were not identical to the montage
used by the SRI group (Oz referenced to linked mastoids), but our montage
was used to provide more information about localization within the
occipital region, hemispheric specialization, and possible related
activation at other leads.
The EEG output (J6) of the Grass model 7 polygraph was
sent to our data analysis system, described above, and also to a Hewlett
Packard FM tape recorder, through a Vetter multiplex system. The FM tape
provided the capacity for any possible future reanalysis.
4. Editing
The hard-copy EEG output from the polygraph was edited
for artifact by experimenters blind to the stimulus conditions of any
21
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
trial. Any trial showing evidence of electrode slippage or of EMG or
EOG contamination, was deleted from the data set.
Trials were deleted after the session for three reasons
only: EEG artifact, logic circuit failure resulting in a breakdown in
the trial sequence, or out-of-range EEG power (under 50 or above 1299 on
printout). In each case, the linked (or previously paired trial of any
trial discarded was also discarded along with data from all leads for all
8 s. If more than 10 trials altogether were deleted for any session of
40 trials, the session was deleted. Only in the case where it would make
the difference in saving or discarding a session were the tapes of the
session played back and reanalyzed at different gain levels to recover
out-of-range epochs. This was done for three sessions.
5. Stimulus
The stimulus was a battery-driven, continuous-wattage
incandescent lamp, chopped by a continuously rotating apertured disc.
When the lamp was lit, the visible stimulus was a 16-Hz flash. In null
trials the lamp was not lit. The stimulus was stationed in a remote room
approximately 10 m from the subject.
6. Control of Stimuli
The trials in the experimental sessions were triggered by
pulses from one of a set of several tapes so that no human operator was
involved in the triggering of the trials in either the 16-Hz or the 0-Hz
condition (once the session had begun). These tapes were made at our
laboratory during the month preceding the experiment.
Randomized tables for the tapes were generated with a
Texas Instrument SR-51A electronic random-number generator. Random
sequencies of -Fs (16 Hz) and -s (0 Hz) in lengths of 40 were generated,
constrained by the requirements that: (1) the trials be pseudorandomized
within each block of 10 trials (i.e., groups 1-10, 11-20, 21-30, and
31-40 each contain five of each kind of trial); and (2) not more than
three trials in a row of either type be allowed.
22
Approved For Release 2000/08/07: CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
The following procedure was
alternately within each block of 10 trials
sequences of numbers generated by the key.
used: +s and -s were assigned
according to the random
For example, if the 21-30
block was being filled, and the random sequence of numbers was 14, 38,
45, 27: first a + would go to 1, then a - to 4, then + to 3, - to 8,
+ to 5, - to 2, + to 7, etc., until the block was filled, and then on
to the next block (repeated digits were ignored). Furthermore, each +
and the succeeding - were linked in the record for editing purposes (see
above).
Blocks of +s and -s were discarded if it was
they would include sequences of four or more consecutive +s
if a sequence of four or more -1-s or -s was created from the
clear that
or -s; also,
juxtaposition
of two blocks of 10, the latter block was reversed (+s changed to -s, and
vice versa). A single block of 10 trials was discarded because of cal-
culator failure in the middle of generating the block, and another was
discarded because of a possible recording error on the part of the opera-
tor; otherwise each trial that was generated was kept.
a 4-track Ampex
Control
stereo
signals for trials were recorded on a tape with
FM tape deck in direct mode, with pulses on one
channel for the 16-Hz condition and on another for the 0-Hz condition.
The pulses were
produced by two Grass stimulators. They were recorded
30 s apart, then checked afterward on playback. The inter-trial interval
was checked and found to be within 1 s of 30 s, consistently, with no
detectable systematic difference between conditions. The tapes were
played back on a Tandberg two-track stereo tape recorder into a logic
circuit that triggered the type of trial corresponding to the channel.
There were only rare failures in trial triggering due to errors in the
trial tapes at any point during the experiment.
The coded tapes were selected by number with no pre.
arrangement except that a different tape be used for each session in a
set until all tapes were used once. Only the operator of the logic
equipment had the knowledge of which tape was being used, and no person
knew before any trial what the trial type would be: that information
was coded in the tape. The coded tapes produced 5-V pulses which,
mediated by the digital logic, triggered the appropriate stimulus type
23
Approved For Release 2000/08/07: CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07: CIA-RDP96-00788R001300350001-0
for any trial. Inter-trial interval was fixed by the spacing of pulses
on the tape to be 30 ? 1 s. The command box of the photic stimulator,
when triggered, produced a 1-s warning tone to both sender and receiver
and, following another 1-s interval, flashed a light for 10 s when a
16-Hz trial was ordered, or did nothing if a null trial was ordered.
The digital logic sent pulses to turn the analyzer on and off at 1 s and
10 s, respectively, from the onset of the trial. For each trial the
digital logic generated timing pulses to be recorded on the Hewlett
Packard tape for use, if necessary, in off-line computer analysis of the
data. A 16-Hz trial was differentiated from a 0-Hz trial by the presence
of an initial 0.5-V positive pulse for the 16-Hz trial. This was the
only electronic signal differentiating the two conditions, that entered
the recording area while the experiment was in progress. Operators in
the recording area did not monitor this signal.
B. Results
The means of summed alpha power in arbitrary relative power
units are presented in Table 2. The differences found to be significant
at greater than 0.05 are starred. Significance levels were assigned by
F-tests from repeated measures of analyses of variance. The analyses
of variance were performed separately for each lead and epoch and
repeated across condition means.
The SRI experiment reported a decrement at Oz-linked mastoids
at 16 Hz compared with the null condition, in the second 4 s. At our
Oz-Cz lead we did not find a change in the second 4 s of either set, but
did find a significant decrement in the first 4 s of the first set.
None of the other Oz comparisons attain significance and the combined
Set 1 and Set 2 first 4 s is not significant. Therefore we did not
specifically confirm the SRI findings. However, 01 and 02 showed sig-
nificant decrement in the first 4 s of Set 1, and 01 showed enough
decrement to reach significance in the last 4 s in Set 2, and in all 8 s,
with both sets combined. Because our electrode montage was not identical
with the SRI montage, our first experiment showed some decrements at
other locations.
24
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Table 2
FIRST EXPERIMENT AT LANGLEY PORTER:
MEANS OF SUMMED EEG ALPHA IN RELATIVE-POWER UNITS,
EXPRESSED AS NULL CONDITION/16-Hz CONDITION
Set 1
Lead
C3
C4
01
02
Oz
0-4 s
276/261
276/245
353/305*
425/336**
419/352*
o.
4-8 s
284/267
268/249
358/347
415/413
420/413
0-8 s
560/528
544/494
711/655
841/749
839/765
Set 2
0-4 s
305/288
389/386
469/461
534/536
532/563
0
4-8 s
303/274
411/371
479/419*
511/477
484/516
0-8 s
608/562
800/757
948/880
1045/1013
1015/1079
0
Combined Set 1 and Set 2: 01 Only
0-4 s 411/383
4-8 s 419/383*
0-8 s 830/766*
*
* *
Significant at p < 0.05.
Significant at p < 0.01.
IV SECOND EXPERIMENT AT LANGLEY PORTER: ATTEMPTED REFINEMENT
Since the data from the SRI study and our first one were compatible
if not identical, we decided to continue to explore possible sources of
variance in this phenomenon. Specifically, we pursued the dependence
upon the "sender" and the possible importance of a trend for the trials
in which the subject correctly guessed that the light was on or off to
show the greatest EEG power differences. In addition, we examined the
differences between our montage of EEG sites (01, 02, C3, C4 ref. Cz)
and the SRI montage (Oz to linked mastoids); and used more careful pro-
cedures ensuring double-blinding of experimenters and subject.
25
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
A. Method
1. General
The method was the same as in the previous experiment,
although carried out in a new location (see Figure A-1 in the Appendix),
and with the addition of several refinements. Sessions with a "sender"
alternated with sessions without a "sender". The subject was also given
a "pass" or no-response option, so that she could indicate when she did
not feel sure of her response and would not hazard a guess about the
stimulus condition. The pass option was signaled by the subject pushing
a button that marked the polygraph record, so that we could discriminate
guess trials from non-guessed ones. There was an additional EEG channel
used, so that in addition to recording from 01, 02, 03, C4, and Oz
referenced to Cz, as in our first experiment, we also recorded the Oz
referenced to linked mastoids configuration used by the SRI group.
The subject ("receiver") was again Ms. Hammid, and the
"sender" was always R. Targ.
2. Double-Blind Procedures
In addition to generating the stimulus-condition sequence
types, we ensured that no one had information concerning the trial-type
sequence. The person who generated the tapes gave them to another member
of the laboratory staff for labeling: A, B, C, etc. A third staff mem-
ber chose a tape at random before the session. No one had knowledge of
the trial-type sequence until after the session was completed and the
stimulus marker sequence played back from the Hewlett-Packard recorder.
The operator who played the tape coded the trial-type neutrally (i.e.,
A and B rather than + and -). The operator who did this step knew the
code and was not allowed to handle data further. The personnel who
analyzed the data were blind to which of the two labeled conditions cor-
responded to the actual trial-type.
The trial-type was monitored with a light-sensitive
recorder--i.e., no one was in the room during trials without a "sender".
During "no-sender" sets, Mr. Targ spent the duration of the set with one
26
Approved For Release 2000/08/07: CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
of the experimenters. He did not know in advance whether a given set
would be "sender" or "no-sender."
B. Results
1. General
The means of summed alpha power are presented in Table 3
and Table 4. Table 3 lists the means from all trials, showing the
trials with a sender separately from those without a sender. Table 4
shows the means from only those trials on which the subject guessed the
stimulus condition. Significance levels were again assigned from F-tests
from repeated measured of analyses of variance performed separately for
each lead, repeated across condition.
This experiment shows an effect opposite that of the SRI
experiment and our first study: instead of alpha power decreasing
during the active condition, it increased. The effect reaches signifi-
cance only at Oz referenced to Cz, although the direction is the same at
other leads. Also of interest is the finding that for the SRI montage,
Oz referenced to linked mastoids, there was no significant difference
between stimulus conditions. The only F-value to reach the 0.05 level
is that of Oz referenced to Cz (F = 4.33), no sender. If this is com-
bined with the trials with sender (which is not significant by itself--
F = 2.647), the overall F for 0-8 s is 7.03, p < 0.01. However, this
is a different site than that used for the original SRI finding, and
our first experiment, and in the opposite direction.
2. Guess Effect
For trials in which the subject was required to guess
the stimulus condition (guess trials), our first experiment showed a
tendency to exhibit a greater alpha decrease than in trials in which
the subject did not guess. Trials in which the subject correctly
guessed the stimulus condition showed this tendency to a greater degree.
We investigated this tendency more rigorously in the second experiment.
27
Approved For Release 2000/08/07: CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Table 3
SECOND EXPERIMENT AT LANGLEY PORTER:
ALL TRIALS, SENDER VS NO SENDER MEANS AND F VALUES
OF SUMMED EEG ALPHA IN RELATIVE POWER UNITS,
MEANS EXPRESSED AS NULL CONDITION/16-Hz CONDITION
(Seven Sessions, 121 Trials in Each Condition)
Lead
>
Sender
01
02
C3
C4
Oz
Oz to
Linked
Mastoids
--All Trials
0.041
0.032
0.107
0.744
0.000
0.116
1.071
1.221
2.647**
0.182
0.024
0.017
0-4 s
4-8 s
0-8 s
0.073
0.039
0.089
1.687
2.556
3.853
0
4.1
0-4 s
479/486
484/517
275/264
320/306
408/439
439/449
as
co
4-8 s
428/432
441/480
253/255
280/281
384/410
415/410
0-8 s
907/919
925/997
527/519
595/586
792/849
854/860
No Sender--All Trials
QD
co
0-4 s
2.760
1.560
1.912
0.902
2.750
0.364
4-8 s
0.374
0.541
0.046
1.020
2.440
0.700
C.)
0-8 s
2.090
1.869
0.611
0.001
4.330*
0.770
0
rr.1
0-4 s
404/442
401/491
262/284
335/318
382/423
499/521
as
a)
4-8 s
399/411
441/461
271/267
297/314
385/422
466/498
0-8 s
802/853
902/952
533/551
633/632
766/844*
966/1018
F > 3.93
F > 6.84
(1,
(1,
120).
120).
Significant at p < 0.05,
**
Significant at p < 0.01,
28
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Table 4
SECOND EXPERIMENT AT LANGLEY PORTER:
GUESS TRIALS ONLY, SENDER VS NO SENDER
MEANS AND F VALUES OF SUMMED EEG ALPHA IN RELATIVE POWER UNITS,
MEANS EXPRESSED AS NULL CONDITION/16-Hz CONDITION
0
0.
r7-1
F Values
0
cU
F Values
cu
Lead
Oz to
Linked
01 02 C3 C4 Oz Mastoids
Sender--Guess Trials
0-4 s
0.003
0.470
0.642
1.081
0.833
0.017
4-8 s
0.312
0.676
0.005
0.002
0.097
0.758
0-8 s
0.083
1.058
0.330
0.402
0.614
0.422
0-4 s
489/491
495/514
268/253
318/300
415/438
450/447
4-8 s
445/432
461/482
258/257
284/285
400/408
436/408
0-8 s
934/922
956/997
525/510
602/585
815/846
886/855
211 Trials: 104 null, 107
No Sender--Guess Trials
16 Hz
0-4 s
1.900
0.752
1.182
1.638
1.894
0.044
4-8 s
0.344
0.490
0.013
0.563
1.755
0.901
0-8 s
1.603
0.840
0.663
0.168
3.069
0.533
0-4 s
413/447
475/498
261/278
340/314
387/423
512/520
4-8 s
408/420
451/469
270/272
301/314
392/426
481/519
0-8 s
820/868
928-963
531/550
641/628
779/849
993/1040
211 Trials: 106 null, 105 16 Hz
29
Approved For Release 2000/08/07: CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
The "pass option," described above, to eliminate forced guessing when
the subject was not certain, made it less likely that incorrect guesses
would be included.
Comparison of Table 4 to Table 3 shows that the effect of
guessing certainly does not increase the effect noted at Oz. Alpha
increases in the active condition compared to the null condition were
not significant. The logical inference of this is that the pass trials
should show a greater difference than "guess" or "all" trials. Table 5
shows the pass trials. They do exhibit greater differences in the same
overall direction of this experiment, an alpha increase in the active
Table 5
SECOND EXPERIMENT AT LANGLEY PORTER:
PASS TRIAL MEANS,
SHOWN AS NULL CONDITION/16 Hz CONDITION
Lead: Referenced to cz
Means
,4
c.)
a
w
Sender
01
02
C3
C4
Oz
Oz to
Linked
Mastoids
--Passes
423/523
331/475
754/998
0.81
0.70
0.76
316/356
227/256
543/612
0.89
0.89
0.89
328/344
268/246
596/590
0.95
1.09
1.01
353/454
301/428
654/882
0.78
0.70
0.74
403/440
308/406
711/846
0.92
0.76
0.84
0-4 s
4-8 s
0-8 s
0-4 s
4-8 s
0-8 s
420/468
360/427
780/895
0.90
0.84
0.87
Ratios
of
Means
Means
4
a
w
No Sender--Passes
277/307
269/206
546/513
0.90
1.31
1.06
321/323
299/315
620/638
0.99
0.95
0.97
345/399
303/397
648/796
0.86
0.76
0.81
364/516
333/374
697/890
0.71
0.89
0.78
0-4 s
4-8 s
0-8 s
0-4 s
4-8 s
0-8 s
366/382
357/343
723/725
0.96
1.04
1.00
393/417
393/378
786/795
0.94
1.04
0.99
Ratios
of
Means
30
Approved For Release 2000/08/07: CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
condition, opposite to what was seen in the first study. Further post-
hoc statistical tests were not performed on the pass trials because the
differences were clearly too mall to reach significance. A very con-
servative criterion is required for a post-hoc analysis of an isolated
observation for which there are previous data in the opposite direction.
The second factor in the trend toward a guess effect in
the first experiment was an even greater alpha difference on those trials
in which the subject guessed correctly. The correct guesses for the
second experiment are shown in Table 6. The correct guesses without a
sender show a small alpha increase in the active condition compared to
Table 6
SECOND EXPERIMENT AT LANGLEY PORTER:
CORRECT GUESS TRIAL MEANS,
SHOWN AS NULL CONDITION/16-Hz CONDITION
Lead: Referenced to Cz
Means
4
8
4
a.
w
Sender
01
02
C3
C4
Oz
Oz to
Linked
Mastoids
--Correct Guesses
291/233
277/236
568/469
1.25
1.17
1.21
348/276
303/292
651/568
1.26
1.04
1.15
463/407
417/407
880/814
1.14
1.02
1.08
481/445
387/459
868/904
1.08
0.84
0.96
0-4 s
4-8 s
0-8 s
0-4 s
4-8 s
0-8 s
518/463
465/429
983/892
1.12
1.09
1.10
561/497
473/492
1034/989
1.13
0.96
1.05
Ratios
of
Means
0
Means o
Ratios
0
of
Means w
No Sender--Correct Guesses
0-4 s 425/449 469/505 284/256 356/298 386/420 529/540
4-8 s 421/411 456/453 302/262 331/306 394/406 496/585
0-8 s 846/860 925/958 586/518 687/604 780/826 1025/1125
0-4 s 0.94 0.93 1.11 1.19 0.92 0.98
4-8 s 1.02 1.01 1.15 1.08 0.97 0.85
0-8 s 0.98 0.97 1.13 1.14 0.94 0.91
31
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
the null condition (not significant). Correct guesses with a sender
show a small alpha decrease compared to the null condition, in the
direction of the first experiment, but too small to be significant.
V DISCUSSION
In two studies, one at SRI and one in our laboratory, a decrease in
EEG alpha power was observed as a function of remote photic stimulation,
but at different scalp locations and in different time periods following
the stimulus onset. The third study, on our laboratory, showed an in-
crease in alpha power at another site and failed to replicate the
original finding.
Dr. Edwin May of SRI has done a full power spectral analysis of
these data and reports that his analysis shows results similar to ours.
We also observed that the subject's alpha frequency seemed to be slightly
lower in the second experiment than the first, and Dr. May's analysis
confirmed this. However we see no easy way to account for the negative
or inconsistent experimental results in terms of frequency changes. We
cannot account for this frequency change. The subject underwent major
pelvic surgery in the three months between our first and second experi-
ments, but there is no apparent reason to relate this to the EEG change.
Statistical analysis of these data necessitated many significance
tests on the same subject. One must always question the independence
of such tests, and also keep in mind that 5% of them will be significant
if the 0.05 level of confidence is adopted. In addition, repeated
measures on one subject and simultaneous measurements from different
leads always brings the independence of these measures into question.
If the findings were replicable, the statistical questions would
be less important. However, since there are many possibilities and we
failed to replicate either our earlier finding or the SRI results, it
is necessary to take a more conservative statistical position. The
three separate experiments all show effects in the occipital cortex, but
it is not the same effect, either in direction or site. We believe the
32
Approved For Release 2000/08/07: CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
three experiments should notItherefore, be lumped together as jointly
supporting each other: They could just as easily be taken as jointly
nullifying each other.
In conclusion, on each occasion, alpha changes occurred that differ
from chance, but we could not reproduce the same effect, in either
direction or site. There remains the possibility that a useful pheno-
menon exists here, but if so, we have clearly not isolated the variables
necessary to produce it at will.
33
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Appendix
SRI OPTICAL ISOLATION EXPERIMENTS AT LANGLEY PORTER
34
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Appendix
OPTICAL ISOLATION EXPERIMENTS AT LANGLEY PORTER
1. Introduction and Summary
As described in Sections I-B and III-A of this report, the purpose
of these measurements was to determine whether there exists any measur-
able light leakage between the stimulus roam in which a chopped light
source operates, and the subject room in which EEG recordings are being
made.
Using phase-sensitive detection, a photomultiplier with an S-20
photocathode, and a synchronously chopped light source, we obtained a
system sensitivity of 5 X 10-15 W/cm2 with unity signal-to-noise ratio,
limited by photo-tube dark current. This sensitivity, in W/cm2, is
3 x 10-13 of the flash lamp intensity used in our experiments (125 dB
down). Our findings using this equipment are as follows:
(1) In order to obtain a measurable light leakage signal it
was necessary to open two of the three doors separating
the stimulus room from the subject room (see Figure A-1)
and direct the output of the flash lamp on the door frame
of the subject room. Under these conditions a signal
could be detected with 20:1 signal-to-noise ratio by the
photomultiplier placed in the subject chair.
(2) When any two of the three doors between the subject room
and the flash lamp were closed, no signal could be
detected, up to the measuring capability of the apparatus,
with a 10-s integration time.
To examine the possibility of light transmission through
the air conditioning ducts, we placed the photo-tube
detector in the duct in the subject room and directed the
output of the flash lamp on the duct in the stimulus room.
Again, no detectable signal was found.
(3)
From this investigation we find that to the limit of our measurement
capability and under the conditions defined by the experimental protocols
and geometry, there was no light leakage between the rooms in question;
thus we conclude that the experimental findings of this program cannot be
attributed to light leakage.
35
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
2. Laboratory Arrangement
The laboratories evaluated in this experiment consist of two groups
of rooms on either side of a hallway. (See Figure A-1.) The doors of
the two laboratories face each other as they open onto a hall. The
stimulus room housing the lamp is a single room with one door. The sub-
ject room is entered from the hall, by first passing through a control
room housing EEG recording equipment. The subject room is considered a
darkroom, and has windows on the outer wall (opposite the control room)
that are taped and boarded up. A small but visible amount of light
enters the room around the tape. The room is separated from the control
room by a wooden door having plastic weather stripping on its edges, as
have the outer doors of both labs. Thus, from where the subject sat in
the dark room, the light from the stimulus room would have had to pass
through three closed and sealed doors to reach her.
3. Apparatus
The light-detecting system used in this measurement is shown in
Figure A-1. An electron multiplier photo-tube (RCA 7265) is powered by
an 1800-V dc power supply, and its anode current is measured across a
100-kilohm resistor. The voltage thus developed goes to the input channel
of a PAR (Princeton Applied Research) phase-sensitive detector, having a
tuned input filter, phase-sensitive amplifier, and variable time con-
stants. We derive the reference signal for the amplifier from a small
photodiode taped to the face of the chopped light source in the stimulus
room. The chopping frequency is 16 Hz.
The output of the light source was measured with a 16-junction
Eppley thermopile and a microvoltmeter. This measurement gave an
intensity of 15 mW/cm2 at 10 cm from the light source.
Using this now-calibrated light source, together with calibrated
neutral density filters, we measured the sensitivity of the phototube
-15
2 .
and phase-sensitive detector to be 5 x 10W/cm , with an integrating
time constant of 10 s.
36
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
SEALED
WINDOW
SUBJECT ROOM
PHOTO TUBE
HV
RCA 7265
SIG
SEALED
DOOR
IN
SEALED
DOOR
HV
1800 V
SIGNAL
REFERENCE
CHART
RECORDER
PHASE-
SENSITIVE
DETECTOR
41111.
CONTROL ROOM
HALLWAY
LAMP
AND
CHOPPER
STIMULUS ROOM
SEALED DOOR
/01 REFERENCE
PHOTODIODE
FIGURE A-1 FLOOR PLAN AND APPARATUS FOR LIGHT-LEAKAGE MEASUREMENTS
37
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0
4. Results
In these experiments the conditions were exactly as they were during
the actual EEG experiments described in detail in Section IV-A of Part 2
of this report. In particular, there was a small amount of outside day-
light leaking into the subject room through the electrical tape used to
seal the window cracks. This was quite a small amount of light, amounting
to approximately six times the dark current of the phototube--i.e.,
3 x 10-14 W/cm2? This amount of light is barely detectable to the dark-
adapted eye, but does give rise to incoherent shot noise in the phototube,
increasing the magnitude of the minimum detectable signal.
The subject room and the light stimulus room are connected to a
common air-conditioning system. The air-conditioner ducts make eleven
right-angle bends in traversing the 30 feet between the two rooms, and
this ducting was a prime suspect for possible light leakage between the
two rooms. To determine the presence of such leakage, the phototube was
placed inside the duct in the subject room, and the light stimulator had
its output directed at the duct in the stimulus room (even though this
was not the condition during actual experiments, when it was pointed
toward the side wall of the room). With the phototube in the duct, its
background was no longer determined by the ambient light in the room,
but rather by phototube dark current, giving a phase-sensitive amplifier
reading of 5 ?V, corresponding to 5 x 10-15 W/cm2? This signal was not
affected by the presence or absence of light from the flash lamp directed
at the other end of the air-conditioning duct.
In measurements with the flash lamp directed along a vector across
the intervening hall toward the door frame and closed door of the
experimental room, no position or manipulation of the flash lamp produced
any effect whatever on the output of the phase-sensitive detector. A
signal could be detected only when both of the two hall doors of the
laboratory rooms were left open to allow the light to leave and enter
the rooms (a condition counter to those set up during experimentation).
The signal in this case was twenty times the shot noise background due
to ambient light in the darkroom, with its door closed.
38
Approved For Release 2000/08/07 : CIA-RDP96-00788R001300350001-0