Content uploaded by Stefan Schmidt
Author content
All content in this area was uploaded by Stefan Schmidt on Sep 17, 2018
Content may be subject to copyright.
Müller, Schmidt & Walach
The Parapsychological Association Convention 2006
85
REMOTE STARING DETECTED BY CONSCIOUS AND
PSYCHOPHYSIOLOGICAL VARIABLES
COMBINING AND IMPROVING TWO SUCCESSFUL
PARADIGMS
Susanne Müller
1
, Stefan Schmidt
1
& Harald Walach
2
1
Department of Evaluation Research in Complementary Medicine
Institute of Environmental Medicine and Hospital Epidemiology
University Hospital Freiburg, Germany
2
University Northampton, School of Social Sciences and Samueli Institute,
European Office
A
BSTRACT
Findings in parapsychology suggest an effect of distant intentionality. Two laboratory set-ups explored this
topic by measuring the effect of a distant intention on psychophysiological variables. The DMILS (direct
mental interaction in living systems) experiments investigate the effect of various intentions on the
electrodermal activity (EDA) of a remote subject. The “Remote Staring” experiments examine whether gazing
by an observer (starer) covaries with the electrodermal activity of the person being observed (staree).
In two meta-analyses (Schmidt, Schneider, Utts & Walach, 2004) it became obvious that the remote staring
studies had a lower overall quality than the DMILS studies. While there are some high quality DMILS studies
(score over 90%) the highest quality in Remote Staring studies is 71%. Thus there is a lack in studies with good
methodology to assess the remote staring paradigm.
We conducted a remote staring study that intended to overcome methodological shortcomings of earlier studies
Fifty participants were invited to take part as starees. After completing questionnaires on mindfulness, mood,
personality and paranormal belief they rested in a comfortable position in front of a video camera while their
EDA was continuously monitored. The experimenter also acted as the starer and either observed or did not
observe the participant through a closed circuit television system according to a random schedule. EDA during
stare and non-stare epochs was compared for significant differences.
In addition to this basic (replication) set-up two new hypotheses were tested. The participant had the possibility
to press a button whenever s/he feels stared at. This added a conscious response variable without engaging into
the disadvantages of the standard conscious guessing paradigm (guessing strategies, response bias etc).
Furthermore the distraction of the starer’s intention during non-stare epochs was varied. In one condition s/he
was mentally occupied by a cognitive task, in the other s/he was just told not to stare (standard condition). We
hypothesized that the distraction from the target in the standard condition was too weak to avoid an unwanted
intentional effect in the staree.
Overall we did not find any staring effect at all, not in the EDA data and not in the ‘conscious’ open response
situation. Thus the experiment failed in demonstrating any Psi effect.
I
NTRODUCTION
The era of examining the “everyday phenomenon” of feeling unseen gazes in a laboratory setting has a
long history and can be differentiated in two main paradigms: the simple test setting and the more
sophisticated one:
In the first kind of experiment in a randomized series of trials the starees are looked at (or not looked
at) directly by another person sitting behind them, watched through a one-way mirror or watched by
closed circuit television system and for each trial they had to make – in forced-choice way - guesses if
Remote Staring Detected by Conscious and Psychophysiological Variables
Proceedings of Presented Papers
86
they were looked at or not (Titchener, 1898; Coover, 1913; Poortman 1959; Peterson, 1959; Williams,
1983; Sheldrake, 1994, 1998, 1999, 2000).
All of these “the way of being stared at” or “unseen gaze” – studies were quite informally conducted
and hardly shielded against sensory cueing.
A radical change in methodology towards the more sophisticated testing began with the studies of
Braud et al in 1990. He and his colleagues pointed out that the effect sizes of the former studies were not
very impressive because all these studies “were designed to encourage deliberate conscious guessing in
order to identify staring periods” (1993a, p.376) and that stronger effects could be obtained if relatively
“unconscious” autonomic nervous system activity were used as the indicator of staring detection, rather
than conscious guessing “…because … autonomic reactions might be less distorted by higher cognitive
processes and therefore might provide a purer and more sensitive indicator” (1993b, p. 392).
From the late 1970´s until the early 1990´s Braud and his colleagues conducted altogether a series of
37 so called DMILS studies in the laboratories of the Mind Science Foundation in San Antonio (Schlitz &
Braud, 1997).
DMILS is the abbreviation of “direct mental interactions with living systems”. In DMILS research an
individual (agent) tries by means of mental intention and volition to interact with the behavioral or
psychophysiological response of another, sensory isolated living target system (receiver). The most
frequently response system that is studied in DMILS research is the electrodermal activity (EDA) of the
receiver which the agent attempts to calm or activate during an experimental session. A typical DMILS
session consists of 10 “activate” periods, 10 “calm” periods (and 20 “rest” periods in between) whose
ordering is randomized and counterbalanced. Under the null hypothesis there is no difference to be
expected between the receivers EDA arousal during the calm or activate periods.
The first remote-staring experiments were conducted by Braud and colleagues in the early nineties
(Braud, Shafer & Andrews, 1993a, 1993b). In these EDA “remote staring” DMILS studies the paradigm
of former DMILS-studies was combined with a quite simple design to test if people were able to detect
when they are watched (or stared at) by a sender in a distant room. An experimental session consisted of
two periods in random sequence: during a staring period the starer (agent or sender) was told to look
intently at the real-time image (on a monitor or screen) of the staree (receiver) which is conveyed to him
through a closed-circuit video camera system. His task was not to calm or activate the staree like in the
DMILS-studies; in the non-staring periods the starer just turned away from the monitor and kept his mind
busy with anything else than the experiment.
Altogether the four series of remote-staring experiments revealed significant results. The starees EDA
was significantly more activated during the staring periods compared to their EDA level during the non-
staring periods.
The experiments that followed the same paradigm (Schlitz & LaBerge, 1997; Wiseman & Schlitz,
1997; 1999; Wiseman & Smith, 1994; Wiseman, Smith, Freedman, Wasserman & Hurst, 1995) can be
regarded as conceptual replications of Braud’s experiments. Most of them also revealed significant results
by measuring the extent of activation during staring periods but the way the EDA was measured showed
some shortcomings and thus leaves some doubt about the validity of the results (Schmidt & Walach,
2000).
Most popular today are the studies conducted and promoted by Rupert Sheldrake, run in style of the
older - more simple - paradigm that made the conscious guessing of the staree the interesting outcome-
variable. Sheldrake conducted or supervised big series of experiments mostly in schools and could always
prove overall positive results with extremely high significances: while the responses at staring epochs
were constantly above chance they were around chance at non-staring epochs (Sheldrake, 1998, 1999,
2000, 2001, 2003, 2005). Anyway there are still a lot of controversies concerning the methodology as well
as the statistical analysis of these experiments (Schmidt, 2003;, 2005) .
Taken the impressive results of Sheldrake it looks like Braud and colleagues were wrong by
speculating that “…in bypassing the cognitive processes, one avoided a “noise” source that could
potentially obscure the relatively weak psi signal.” (Delanoy, 2001, p. 34).
So far a comparison between these two paradigms has just taken place in a few studies (i.e., Tart, 1963;
Targ & Puthoff, 1974; Dean, 1974; Delanoy & Sah, 1994; Lobach & Bierman, 2004). The results of these
Müller, Schmidt & Walach
The Parapsychological Association Convention 2006
87
studies support the idea that the unconscious, autonomic responses measured in EDA tasks may provide a
more sensitive instrument for detecting staring/psi effects whereas the conscious guess measure did not
differ meaningfully from chance expectations.
In our study we made a direct comparison between these two measures:
Additionally to the measure of the autonomic reaction of the staree (EDA) we also measured the
“conscious reaction”. This procedure has some advantages:
By conscious guessing alone it will be difficult for the staree to avoid guessing strategies, response
biases or intellectual analyzing and so forth. On the other side the phenomenon in everyday life is
described by a conscious response and the question is open how this is modulated.
By measuring nervous processes and the possible difference in staring and non-staring periods and by
comparing them with the conscious guessing it may be possible to find results about their interactions and
the possible correlation of this interaction with other variables (i.e. personality, belief in Psi, mood).
So far conscious reports in staring experiments were received through a forced-choice procedure where
“… the receiver must engage in overt cognitive processing to provide a response to the target” (Delanoy,
2001, p. 35). But this procedure is rather different from a daily life experience, where one is not asked
whether one has been stared at but notices so in a particular moment. While the forced-choice procedure
invites for certain guessing strategies and reflections on the subsequent answer in dependence from the
last one, this is not true for the daily life situation.
Thus we introduced in our experiment an open response situation that has a higher ecological validity
than a forced-choice procedure. The starees in our study didn’t have to make a decision about staring
“yes” or “no” at a particular moment but were asked to press a button whenever he or she had the feeling
that the starer tried to get connected with them and to release this button again when this feeling vanished.
To avoid any cognitive strategies they were not informed on the number of staring vs. non-staring epochs
their length or their starting point.
We furthermore introduced a new procedure on the side of the starer. Normally the agent/starer is
asked to maximize his/her attention during staring periods and to think about something else during non-
staring periods. Presuming that staring at somebody in daily life is just a process of focusing on a specific
intention (unconscious or conscious) which in turn can be detected by a staree (unconscious or conscious),
it is absolutely necessary that the agent maximizes his or her attention in order to make the experimental
situation as close to real life as possible. But the harder part of the agent’s task is to turn off his or her
attention and intention in non-staring periods! Everybody knows that trying to do so usually produces the
opposite. S/he will think exactly about what they should not think about, no matter how hard they try to
make themselves busy. Presuming that there is the possibility of agent and receiver to get connected
somehow by focused attention it can be assumed that the agent will be confused by this “steady attention”
and may feel to be stared at even during the non-staring condition.
In order to avoid this we did not only ask the starer to maximize his attention during staring periods,
but also made him or her minimize his/her attention in the non-staring periods by occupying him/her with
a mental task that should result in “forgetting” about the staree for a short time. The starer was asked to
perform a demanding cognitive test, exciting enough to capture the mind but also simple enough to turn
away from upon the start of the next staring period. This procedure aimed at increasing the variance in the
intentions of the starer between staring and non-staring epochs and thus we expected also larger effects in
the reactions (conscious and unconscious as well) of the staree.
To find out whether there is a difference in the magnitude of the reactions (conscious and unconscious)
of the staree, half of the sessions for each starer were conducted in the “normal” way, (this means he/she
was told just to turn away from the screen in the non-staring periods and try hard to think about anything
else but the staree). In the other half the starer was occupied with the cognitive task in the non-staring
periods.
To find out more about the role of the starers´ task to minimize his attention can be of great advantage
for further parapsychological experiments. In spite of Braud and his colleagues´ assertion “…that the
quality of the starers´ attention is important in determining the nature of the experimental outcome” (2001,
405p.), no study so far has tried to find out about this “pure attention component”.
Remote Staring Detected by Conscious and Psychophysiological Variables
Proceedings of Presented Papers
88
Altogether our objectives were as follows:
- To replicate the two main staring paradigms (conscious report, EDA) in one experimental setting
under best controlled conditions with state of-the-art EDA measurement (Schmidt & Walach, 2000).
- To assess a new method of recording conscious responses that overcomes the disadvantages of the
standard forced choice method.
- To find out about the moderating variables of a possible staring detection effect. Related to this is
the need to better understand the different roles of the various participants in a staring experiment
(starer/agent and staree/receiver).
- To try to increase the staring detection effect by eliciting a larger variance in the attention/no
attention task of the agent.
M
ETHODS
SETTING
Location
All experimental sessions were conducted in a two-floor-building belonging to the Institute for
Environmental Medicine at the University Hospital in Freiburg. Starer and staree were located in two
entirely separated rooms in different wings of the building. The staree´s room was located on the ground
floor and the starers´room on the first floor approximately 20m in distance. The physical separation of
these rooms together with the standard use of a closed-circuit video system made any conventional
communications impossible.
Figure 1: Plan of the location where the experiment took place
During the experimental session the staree sat quiet and relaxed in a comfortable chair. Jewelries and
mobil phones had to be removed before the start of the session. In order to eliminate any noises from
outside which could influence the measurement of the EDA the starees wore special headphones, which
extremely attenuate sounds from the environment (Sennheiser HD 280 Pro). In addition the participants
Gggro
Ground
floor
First floor
stare
r
staree
Müller, Schmidt & Walach
The Parapsychological Association Convention 2006
89
listened to a special music (Dr. Harold Moses – “The Drone”) which did not contain any sudden changes
or beats in order to make them relaxed and comfortable without interfering with the EDA measurement.
Participants (starees /starers)
Fifty participants - recruited through leaflets – took part in the experimental sessions in exchange for
10 Euros. Twenty-six participants (52%) were female, twenty-four participants (48%) were male and the
mean age of all participants was 32.5 years (SD = 10.3, range 20-57). There were no special criteria for
the participants; everybody who was interested to participate was invited. Thereby the participants were a
self-selected sample.
The whole study was conducted by two experimenters - one female and one male - who also acted as
the starers (agents). They were hired for this study and extensively trained in a pilot study.
The complete organization and conductance of this study was carried out by the first author SM.
Materials
Video equipment
A video camera (AxisNetcam206) was positioned half in front and half sideways of the staree and
transmitted real-time images to a monitor in the starer’s room. This one-way closed circuit video system
allowed the starer (the experimenter) to see the participant but not the other way round.
EDA-measurement
The staree’s electrodermal activity was assessed by measuring skin conductance (SC-) with a constant
voltage method (0.5 V) according to the guidelines by the Society for Psychophysiological Research
(Fowles, Christie, et al. 1981). Skin conductance level (SC) was recorded with two 8mm Ag/AgCL
electrodes placed on the thenar and hypothenar eminences of the non-dominant hand. An isotonic
electrode gel was applied and a time lag of at least 20 minutes between electrode application and start of
measurement was maintained to allow for skin adaptation processes.
SC-data was recorded by a skin conductance device built especially for this purpose by the University
Hospital’s electronic workshop. This device splitted the data into two channels, the tonic signal (SCL) and
the phasic signal (SCR). The latter one was derived by treating the SCL with a time constant of 10s. Data
was digitized (12 bit) and sampled at 20 Hz for each channel.
Two variables were derived from this data (i) Number of non-specific skin conductance responses
(NS.SCR for more details see Schmidt & Walach, 2000), (ii) skin conductance level (SCL) calculated as a
mean over all samples during an epoch. While the first one is the main dependent variable for this study
the latter one was recorded for exploratory use.
The signal generated by the button presses of the participant was recorded together with the EDA raw
data in a separate channel. Respiration was recorded by a piezo based respiration belt placed on the upper
abdomen and also stored in a separate channel.
Conscious guessing
In order to record whenever a staree felt stared at the participants were provided with a tiny switch on
their dominant hand (EDA recording took place on the non-dominant hand) and were asked to push this
switch into the “stared at position” whenever they had the feeling that they were being observed. The
switch remained in this position by itself and participants were thus also asked to bring the switch back to
the “not-stared-at position”, whenever they felt there was nobody staring at them from a distance.
Questionnaires
Participants were asked to complete the following questionnaires:
1. “Belief-in-Psi”
Three questions similar to the ones that Wiseman and Schlitz used in their study (Wiseman & Schlitz,
1997). The three questions concerning the subjects’ attitudes toward psi. They will indicate their responses
on a seven-point scale ranging from –3 to +3.
Remote Staring Detected by Conscious and Psychophysiological Variables
Proceedings of Presented Papers
90
A general “belief-in-psi” score is obtained by summing the respondents´ responses over all three
questions. The questions are as follows:
Are you convinced about the existence of psi?
Certain -3 -2 -1 0 +1 +2 +3 Not at all
What best describes your own psi ability?
I have psi ability -3 -2 -1 0 +1 +2 +3 I have no psi ability
Do you believe you might be able to demonstrate any psi ability in this experiment?
Yes -3 -2 -1 0 +1 +2 +3 No
Each staree had to answer these questions before the experimental session.
2. “Mindfulness” (Freiburg Mindfulness Inventory; FMI)
Participants were asked to fill in the 14 item short form of the Freiburg Mindfulness Inventory (FMI)
(Buchheld, Grossman & Walach, 2001; Walach, Buchheld, Buttenmüller, Kleinknecht & Schmidt, in
press). This includes, apart from awareness for the environment, also the awareness of one’s own mental
processes, emotions and signals from the body. The short form of this questionnaire can be filled out by
people who are not familiar with the mindfulness concept itself.
The rational for measuring mindfulness in participants was to find out how the ability to be
continuously aware of the present moment is related to the detection of a remote stare observation. The
rationale for measuring mindfulness in the starer was to find out whether the ability to be aware of the
presence is related to a better performance. Mindfulness, intention and attention share aspects that are
likely to be important for these kinds of intention experiments.
Each staree had to fill out this questionnaire before the experimental session.
3. „Well-being“ (Befindlichkeitsskala; Bf-s)
General well-being was measured by the Bf-s (Zerssen, 1976), an adjective list which measures general
well-being in 28 pairs of adjectives with opposite semantic content arranged in a semantic differential.
The scale is a widely used, psychometrically sound scale for measuring short term changes in well-being.
The staree had to fill out this scale shortly before and directly after each session.
The starer had to fill out this scale shortly before each experimental session.
Procedure
The procedure was mainly equal to the “classical” staring experiments by Braud and colleagues (Braud
et al., 1993a, 1993b) with the following changes or extensions:
Each experimental session was conducted individually by one of the two experimenters (starers).
After the questionnaires had been filled out, the electrodes had been attached, the respiration belt
fastened and the headphones put into position the participant was left alone for five minutes in order to
record his/her basal skin resistance (baseline). The experimenter reentered the staree’s room, saved the
baseline data and then went to his room upstairs. Exactly 6 minutes after s/he had left the participant s/he
started the computer program that ran the whole session. This program was written in VBA; it sampled
and stored the incoming data, selected the random sequence, and controlled the monitor for the starer by
switching between the picture of the staree and a blank screen (standard condition) or a computer game
(distraction condition).
The only thing the experimenter had to do manually was to set up the program for either the standard
or the distraction condition. The sequence of these two conditions throughout all the sessions was in a
randomized order with the same number for both. Each of the experimenters had a package of sealed
envelopes locked in a cabinet drawer. Before starting the program s/he chose the envelope labeled with the
smallest number and like this found out about the ongoing condition.
During a staring epoch the experimenter saw the real-time image of the staree on the monitor. During
the condition where he/she was kept busy with the computer game through the non-staring period, the
game just appeared on the monitor and in the other condition s/he was presented a “non-staring”
instruction on the screen.
Müller, Schmidt & Walach
The Parapsychological Association Convention 2006
91
The sequence of these epochs was in a randomized order with the same number of staring and non-
staring trials. The sequence was balanced in a way to avoid linear trends (such as a steady decrease in skin
conductance or trends caused by shifts in the amplifier etc) potentially resulting in artefacts. An
algorithmic random process was employed to draw a sequence out of a pool of sequences fulfilling the
above criteria. One epoch lasted 60 seconds. There were 10 staring and ten non-staring epochs. Staring
and non-staring epoch were interspersed by short rest intervals of variable length (5-15 seconds, randomly
assigned). Thus the overall session length summed up to 23-30 minutes.
It is important to notice that the experimenter (the starer) was blind to the sequences of the epochs.
Also the starees were blind against the sequencing of these two types of epochs but also against the
number and the timing. They were just told that the camera would be switched on through the whole
session but that the starer would only look at the monitor at certain times.
After the session was finished the experimenter waited for five more minutes before s/he entered the
staree’s room. Thereby, a second baseline at the end could be recorded.
In a pilot study with ten participants the laboratory set-up, the measurement devices and all procedures
were tested in order to perform the main study according to a strict protocol. Before the beginning of the
main study in January, this protocol was deposited with Eberhard Bauer (IGPP, Freiburg).
HYPOTHESES
Our hypotheses were as follows:
Main Hypotheses
1. Staring Hypothesis (proof oriented)
Remote Staring: unconscious response (EDA)
There will be a significant difference in the starees´ electrodermal activity between staring and non-staring
periods. The direction of this difference will not be specified (two-tailed test).
Phasic Activity:
There will be a significant difference in the value „NS.SCR.frequencies > 0.01 µS“ between staring
and non-staring periods. The direction of this difference will not be specified (two-tailed test)
Tonic Activity:
The same analysis will be performed for SCL for exploratory use.
Remote Staring: Conscious response
Starees will press the staring button more often and for longer periods during staring epochs than during
non-staring epochs.
2. Distraction Hypothesis
Remote Staring: unconscious response (EDA)
The differences in the starees´ electrodermal activity (between staring and non-staring epochs
throughout all the experimental sessions) in the sessions where the starer will be occupied by a cognitive
task will be significantly higher than the differences in the starees´ electrodermal activity in the sessions
where the starer is asked just to “think about something else” while not staring.
Remote Staring Detected by Conscious and Psychophysiological Variables
Proceedings of Presented Papers
92
Remote Staring: Conscious response
Starees will press the staring button more often and for longer periods during non-staring epochs in the
condition when the starer is asked just to “think about something else” while not staring, than in the
condition when s/he will be kept busy by playing a computer game.
DATA ANALYSIS AND STATISTICS
SCR-
CHANNEL
The data was first transformed into standardized measured values and then treated with a 0,5 Hz low-
pass filter. Afterwards each of the twenty phases was analyzed for NS.SCR.frequencies as well as for
NS.SCR. sum of amplitudes with a special software (EDA-Para). The value threshold limits regarding to
analyzing the data were 0,01 µS.
Each of these values were added up separately for the two conditions (stare and non-stare) throughout
all epochs of one experimental session. Like this each session resulted into two pairs of values (consisting
of the sums of staring and non-staring epochs).
SCL-
CHANNEL
After being transformed to standardized measured values (µS) each of the 60 sec epochs were
averaged. These mean values were averaged over all epochs of the same type (10 * staring and 10 * non-
staring). Thus each session resulted in one pair of values (consisting of the mean value of the staring and
the mean value for the non-staring epoch).
R
ESPIRATORY ACTIVITY
Respiratory activity was recorded but not analyzed. The data is available for later analyses.
A
NALYSING
„
CONSCIOUS GUESSING
“
Changes in the switch position were also analyzed per epoch and then summed up over all epochs for
each of the two conditions. There were two variables extracted from this data: (i) number of times the
switch was moved from the “not stared at” to the “stared at” position. (ii) percentage of samples within an
epoch where the switch was in the “stared at”-position (with 100% indication all the time and 0%
indicating never).
While the first variable was analyzed confirmatorily, the second variable was for exploratory use.
C
RITERIA FOR EXCLUSION OF DATA
In our protocol we prespecified a set of exclusion criteria for the EDA data in order to exclude SCR
non-responders from the analysis. A dataset was excluded from analysis if:
• Mean-SCL-value for more than 4 60 sec. epochs was less than 0.5 µS
• The complete dataset showed less than 10 * NS.SCR > 0.01 µS
Datasets were also excluded from the analysis if
• One electrode was disconnected during recording
• The staree wanted to discontinue the experimental session
• Anything else happened that could invalidate data recording
Questionnaire data, calling sequences and guesses of the participants were entered into SPSS for
Windows 13.0. Further analyses were conducted in SPSS and in Excel.
For the Main-Hypothesis - Hypothesis 1 (Staring Hypothesis) we proceeded as follows:
Müller, Schmidt & Walach
The Parapsychological Association Convention 2006
93
For each participant one pair of values was available. These pairs of data were tested for normal
distribution (Kolmogorov-Smirnoff-test), and compared by a t-test for dependent variables..
For the Main-Hypothesis - Hypothesis 2 (Distraction Hypothesis) we calculated as follows:
Throughout all experimental sessions two pairs of values were available. These values were used to
calculate a per-session success score. Next session success scores of distraction sessions and standard
sessions were compared for significant differences. As the criterion “normal distribution” was fulfilled (p
> 0.05) a t-test for independent variables (one-tailed, α = 0.05) was computed. This was calculated for the
NS.SCR data. The same analysis was computed for the other scores (SCL, button presses) but only in an
exploratory fashion.
R
ESULTS
Of the 50 participants only 35 fulfilled the above inclusion criteria. We had a prespecified sample size of
40 in our protocol. But as the study proceeded more slowly than expected and more participants than
anticipated fulfilled the exclusion criteria we had to stop with 35 valid data sets only, as the experimenters
were no longer available. This decision was taken without any inspection or analysis of the data and was
motivated purely by pragmatic reasons.
1. Staring Hypothesis:
21,71 35 12,77 2,16
21,46 35 13,54 2,29
NS.SCR´s frequencies
in stare trials
NS.SCR´s frequencies
in non-stare trials
pairs
mean n standard deviation standard error
Table 1: NS.SCR frequencies during stare and non-stare trials
The average of NS.SCR-frequencies did not differ in stare (m=21.71, sd=12.77) vs. non-stare-trials
(m=21.46, sd=13.54); t=.21 n.s., p=.84 (two-tailed), effect size d = 0.04.
1
5,00 35 4,19 ,71
4,91 35 4,49 ,76
sum: NS.SCR in stare trials
sum: NS.SCR in non-stare
trials
pairs
mean n
standard
deviation standard error
Table 2:
NS.SCR. sum of amplitudes
during stare and non-stare trials (exploratory analysis)
The average sum of NS.SCR-amplitudes did not differ in stare (m=4.99, sd=4.19) vs. non-stare-trials
(m=4.91, sd=4.49); t=.28 n.s., p=.78 (two-tailed).
For the skin conductance level (SCL) the mean SCL did not differ in stare (m=3.82µS, sd=2.10) vs. non-
stare-trials (m=3.80µS, sd=2.13). The variables were normally distributed and thus we applied a t- test to
compare the means. The difference proved to be not significant (t = 1.25 n.s., p = .22).
1
Effect size calculated with d = t/√df
Remote Staring Detected by Conscious and Psychophysiological Variables
Proceedings of Presented Papers
94
Remote Staring: Conscious response
Participants were provided with a switch to signal whenever they felt stared at. Out of our complete
sample of 50 participants 7 never used this switch, and we only analyzed the remaining 43. Our main
analysis counted how often the switch was pushed into the “stared at” position during staring and non-
staring epochs. On average, participants pushed the switch 9.9 times during either “stared at” or “not
stared at” epochs (range 1-40, SD = 8.2). They used the switch more often during “stared at” epochs
(mean = 5.2, sd= 4.2) than during “not stared at” epochs (mean = 4.7, sd = 4.3). The variables were not
normally distributed and thus we applied a Wilcoxon Signed Rank test to compare the means. The
difference proved to be not significant (z = 1.47, p = .14).
As an additional exploratory analysis we calculated the percentage of time during which the switch was
kept in the “stared at” position by the participant. Overall participants had the switch 23.9% of the time in
this position (range 0.4 % - 60.3 %, sd = 15.7). During “stared at” epochs the switch was in 25.1 % (sd =
17.0) of the time in the stared at position, during “not stared at” epochs 22.8 % (sd = 16.9) of the time.
The variables were normally distributed. The difference was not significant (paired sample t-test, t =
1.17, df = 42, p = .25)
2.
D
ISTRACTION
H
YPOTHESIS
:
Remote Staring: unconscious response (EDA)
There were 16 sessions where the distraction condition took place (matrix task) and 19 with the
standard procedure (blank screen). Regarding the SCR variable (NS.SCR.frequencies;
NS.SCR.sum of amplitudes) the results can be seen in table 3:
16 2,13 7,60 1,90
19 -1,32 7,06 1,62
16 ,49 1,95 ,49
19 -,25 1,78 ,41
matrix
matrix
no matrix
matrix
no matrix
NS.SCR: frequencies: stare
trials - non stare trials
NS.SCR: amplitudes: stare
trials - non stare trials
n means standard deviation standard error
Table 3: Difference of NS.SCR frequencies and
NS.SCR. sum of amplitudes between
stare and non-stare trials (matrix vs.
no-matrix condition)
The average difference of NS.SCR-frequencies between stare and non-stare trials was larger in the
matrix (m=2.13, sd=7.6) than in the no-matrix condition (m= -1.32, sd=7.06). Also the average difference
of NS.SCR. sum of amplitudes between stare and non-stare trials was larger in the matrix (m=.49,
sd=1.95) than in the no-matrix condition (m= -.25, sd=1.78).
As the variables were normally distributed we applied a t-test to compare the means. The difference
proved to be not significant (independent samples t-test, t = 1.39, df = 33, p = .16).
Müller, Schmidt & Walach
The Parapsychological Association Convention 2006
95
For SCL the results looked as follows:
16 ,09 1,07 ,27
19 ,36 1,17 ,27
matrix
matrix
no matrix
SCL: stare trials
- non stare trials
n means
standard
deviation standard error
Table 4: Difference of SCL between stare and non-stare trials (matrix vs. no-matrix condition)
The average difference of the SCL between stare and non-stare trials was smaller in the matrix
(m=.01µS, sd =.11) than in the no-matrix condition (m= -.04µS, sd=.12).
As the variable was normally distributed we applied a t-test to compare the means. The difference
proved to be not significant (independent samples t-test, t = .72, df = 33, p = .48).
Remote Staring: conscious response
20 4,55 2,74 ,61
23 4,35 4,50 ,94
20 4,20 3,68 ,82
23 4,30 4,47 ,93
20 29,12 17,79 3,98
23 21,56 15,76 3,29
20 23,09 13,80 3,09
23 22,52 19,56 4,08
matrix
matrix
no matrix
matrix
no matrix
matrix
no matrix
matrix
no matrix
"button on" in stare trials
"button on" in non stare trials
(%)
"button on" in stare trials
"button on" in non stare trials
(%)
n means standard deviation standard error
Table 5: Distribution of button presses throughout matrix vs. no-matrix condition
,04 -,96
23 23
2,03 9,93
,35 6,03
20 20
1,95 14,85
mean
n
standard deviation
mean
n
standard deviation
no matrix
matrix
"button on" in stare trials -
"button on" in non stare trials
"button on" in stare trials (%) -
"button on" in non stare trials (%)
Table 6: Differences between the two conditions (matrix vs. no matrix)
Our main analysis counted how often the switch was pushed into the “stared at” position during staring
and non-staring epochs in the matrix (=distraction) condition compared to how often the switch was
pushed into the “stared at” position during staring and non-staring epochs in the no matrix condition.
(Also here we only analyzed the 43 out of 50 participants who used the switch at all).
As an additional exploratory analysis we also calculated the percentage of time during which the
switch was kept in the “stared at” and “non-stared at” position during the matrix condition compared to
how often the switch was pushed into the “stared at” position during staring and non-staring epochs in the
no matrix condition.
Remote Staring Detected by Conscious and Psychophysiological Variables
Proceedings of Presented Papers
96
As both variables were normally distributed we calculated an independent t-test to compare the two
conditions. The differences proved to be not significant.
Frequency of “button on” in stare trials - frequency of “button on” in non-stare trials (matrix vs. no-
matrix condition): Independent samples t-test, t = .50, df = 41, p = .62).
Percentage of time of “button on” in stare trials - Percentage of time of “button on” in non-stare trials
(matrix vs. no-matrix condition): Independent samples t-test, t = 1.84, df = 41, p = .07).
D
ISCUSSION
We conducted a remote staring experiment where we combined two strands of research which ran in
parallel for several years. So far researcher have investigated this phenomenon by either employing a
physiological (‘unconscious’) measurement (i.e. EDA) as dependent variable or by asking the subjects
directly whether they were stared at or not (conscious forced choice paradigm). Furthermore we replaced
the forced choice situation by an open response procedure, where the participants could indicate at any
time if they have feelings of being stared at (or not) without being prompted. Moreover we tested a
hypothesis on the distraction process of the starer while not staring at the participant. To do this, we either
distracted the starer completely from the staree by a demanding cognitive task, or we left him or her
(according to the standard paradigm) with just a blank screen. The difference between the two conditions
was not significant (p=.07) but for the condition where the starers were distracted (N=20) we found a
medium effect size for a staring effect according to our hypothesis of d =.43 (p =.085) for the difference
between staring and non staring epochs. This may be a chance result, but is large enough to follow-up the
distraction hypothesis in another experiment with a larger sample.
In all other analyses we did not find any staring effect, not in the EDA data and not in the ‘conscious’
open response situation. The experiment failed in demonstrating any Psi effect.
First of all this releases us from the burden to demonstrate that no conventional information transfer
was possible in our newly created remote staring set up. But why couldn’t we demonstrate similar staring
effects like other researchers active in this field? There are three well known lines of reasoning we would
like to follow in short: (i) there is no such thing as a ‘remote staring’ effect and other researchers have
mistaken artifacts for such an effect, (ii) there is such an effect which we missed for reasons to be
discussed and finally (iii) the psi phenomenon under consideration is not showing up in a stable and
replicable mode but moderated by variables in a larger context or by the systemic set-up of the
experimental paradigm in general.
The first (i) position focuses mainly on the methods and set-ups of earlier work and we would not like
to engage in such a discussion here. This has already taken place elsewhere (see Schmidt, Schneider, Utts
& Walach, 2004). There we have explained that in our view some effects can be attributed to
methodological shortcomings but that there remains a substantial effect which lacks a classical
explanation.
Regarding the position (ii) we can put our study in line with the two other recently published studies
(Lobach & Bierman, 2004; Schlitz, Wiseman, Radin & Watt, 2005), which also failed to replicate staring
effects with EDA as well as conscious guessing as dependent variables. Several reasons are possible for
such a failure and we will restrict ourselves to just two. In our meta-analysis (Schmidt, Schneider, Utts &
Walach, 2004) we calculated an effect-size for the EDA paradigm with d = 0.13 which corresponds
approx. to an eighth part of a standard deviation. For an effect so small in size, all remote staring
experiments conducted so far, including the one presented here, are underpowered by far. One would need
studies with several hundreds of participants in order to achieve a reasonable power. But maybe the effect
could just not be demonstrated because our sample was too small. Another possibility is that we didn’t
have the adequate experimenters or participants employed who might be necessary to initialize a psi-
conducive system. Especially remote staring experiments proved to be sensitive to experimenter effects,
(e.g. the Wiseman-Schlitz studies but see also Juniper & Edlmann, 1998). Regarding participants so far
only unselected samples have been tested and it was suggested to perform pretests or screening trials to
select participants who can perform this task successfully.
Müller, Schmidt & Walach
The Parapsychological Association Convention 2006
97
The third option (iii), psi phenomena just showing up in an unstable manor, is of course a valid option,
but we cannot discuss the various theoretical models referring to this position here. Some of this
discussion has been presented elsewhere (Walach & Schmidt 2005).
As there was no overall Psi effect several of our other hypotheses could not be tested. This is especially
true for the question, whether occupying the starer by a distracting task during the non-stare epochs can
increase the differences between staring and non-staring epochs. Our data pointed into this direction but
this can be of course mere chance findings. We nevertheless think that this interesting process oriented
hypothesis should be followed up in future experiments. Unfortunately it is not testable in a system which
does not show a basic psi effect, as this happened here.
A
CKNOWLEDGEMENT
This study and the work of Susanne Müller was funded by the Bial foundation. Stefan Schmidt is
funded by the Samueli Institute for Information Biology, as was Harald Walach at the time the experiment
was conducted.
We are grateful to Isabell Neu and Marco Petrucci for their contribution as experimenters in this study
and to Helge Liebner for his work as software engineer.
Our special acknowledgement however goes to the participants who acted as starees. Without them the
study wouldn´t have been possible.
R
EFERENCE
L
IST
Braud, W., Shafer, D. & Andrews, S. (1993a). Further studies of autonomic detection of remote staring: Replication,
New Control Procedures, and Personality Correlates. Journal of Parapsychology, 57, 391 – 409
Braud, W.G., Shafer, D. & Andrews, S. (1993b). Reactions to an Unseen Gaze (Remote Attention): A Review, with
New Data on Autonomic Staring Detection. Journal of Parapsychology, 57, 372 – 390
Buchheld, N., Grossman, P. & Walach, H. (2001). Measuring mindfulness in insight meditation (vipassana) and
meditation-based psychotherapy: The development of the Freiburg Mindfulness Inventory (FMI). Journal
for Meditation and Meditation Research, 1, 11-34.
Colwell, J., Schröder, S. & Sladen, D. (2000). The ability to detect unseen staring: A literature review and empirical
tests. British Journal of Psychology, 91, 71-85
Coover, J.E. (1913). The feeling of being stared at. American Journal of Psychology, 24, 570 – 575
Dean, D. (1974) reported in Dean, D., Mihalasky, J., Ostrander, S. & Schroeder, L. (1974) Executive ESP, Prentice-
Hall, Inc., Englewood Clifts, N.J., p. 49
Delanoy, D.L. & Sah, S. (1994). Cognitive and physiological psi responses to remote positive and neutral emotional
states. In: D.J. Bierman (Ed.) Proceedings of Presented Papers of the 37th Annual Parapsychological
Association Convention, Durham, NC: The Parapsychological Association.: pp 128-137
Delanoy, D.L. (2001). Anomalous Psychophysiologial Responses to Remote Cognition: The DMILS Studies.
European Journal of Parapsychology, 2001, 16, pp 29 -40
Fowles, D. C., Christie, M. J., Edelberg, R., Grings, W. W., Lykken, D. T. & Venables, P. H. (1981). Publication
Recommendations for Electrodermal Measurements. Psychophysiology, 18, 232-239.
Juniper, Z. & Edlmann, M. (1998). The Relationship Between Perceptual Vigilance/Defensiveness and
Psychophysiological Responses to Remote Staring in the Ganzfeld Condition. University of Edinburgh.
Lobach, E. & Bierman, D. J. (2004). The invisible gaze: three attempts to replicate Sheldrake's staring effects. The
Parapsychological Association, 47th Annual Convention, August, 5-8, 2004 (pp. 77-90). Durham, NC: The
Parapsychologic Association.
Remote Staring Detected by Conscious and Psychophysiological Variables
Proceedings of Presented Papers
98
Poortman, F. F. (1959). The feeling of being stared at. Journal of the Society for Psychical Research, 40, 4-12
Schlitz, M. & LaBerge S. (1997). Covert observation increases skin conductance in subjects unaware of when they
are being observed: a replication. Journal of Parapsychology, 61, 197 – 207
Schlitz, M. J. & Braud, W. G. (1997). Distant Intentionality and Healing: Assessing the Evidence. Alternative
Therapies in Health and Medicine, 3, 62-73.
Schlitz, M. J., Wiseman, R., Radin, D. I. & Watt, C. A. (2005). Of two minds: skeptic-proponent collaboration
within parapsychology. The Parapsychological Association, Inc. 48th Annual Convention, August 11th -
15th, 2005 (pp. 171-177). Durham, NC: The Parapsychological Association.
Schmidt, S. & Walach, H. (2000). Electrodermal Activity (EDA) – State of the Art Measurement and Techniques for
Parapsychological Purposes. Journal of Parapsychology, 64, 139-163.
Schmidt, S. (2002). Außergewöhnliche Kommunikation? Eine kritische Evaluation des parapsychologischen
Standardexperimentes zur direkten mentalen Interaktion. Oldenburg: BIS-Verlag
Schmidt, S. (2003). Rupert Sheldrake: "The sense of being stared at and other aspects of the extended mind".
Zeitschrift für Anomalistik, 3, 263-266.
Schmidt, S. (2005). Comments on Sheldrake's 'The Sense of Being Stared At'. Journal of Consciousness Studies, 12,
105-108.
Schmidt, S., Schneider, R., Utts, J. & Walach, H. (2004). Distant intentionality and the feeling of being stared at:
Two meta-analyses. British Journal of Psychology, 95, 235-247
Sheldrake, R. (1998). The sense of being stared at: Experiments in schools. Journal of the society for Psychical
Research 62, 311-323
Sheldrake, R. (1999). The “sense of being stared at” confirmed by simple experiments. Biology Forum 92, pp. 53-76
Sheldrake, R. (2000). The “sense of being stared at” does not depend on known sensory clues. Biology Forum 93,
pp. 209-224
Sheldrake, R. (2001). Experiments on the sense of being stared at: the elimination of possible artefacts. Journal of the
society for Psychical Research 65, pp. 122-137
Sheldrake, R. (2003). The sense of being stared at and other aspects of the extended mind, Hutchinson – London,
United Kingdom
Sheldrake, R. (2005). The Sense of Being Stared At Part 1: Is it Real or Illusory? Journal of Consciousness Studies,
12, 10-31.
Targ, R. & Puthoff, H. (1974). Information transmission under conditions of sensory shielding. Nature, 252: pp. 605
– 607
Tart, C.T. (1963) Physiological correlates of psi cognition. International Journal of Parapsychology, 5: 375 –607
Walach, H., Buchheld, N., Buttenmüller, V., Kleinknecht, N. & Schmidt, S. (in press). Measuring mindfulness—the
Freiburg Mindfulness Inventory (FMI). Personality and Individual Differences.
Walach, H. & Schmidt, S. (2005). Repairing Plato's Life Boat with Ockham's Razor. The Important Function of
Research in Anomalies for Mainstream Science. Journal of Consciousness Studies, 12, 52-70.
Williams, L. (1983). Minimal cue perception of the regard of others: The Feeling of being stared at. Journal of
Parapsychology, 34, 59 – 60
Wiseman, R. & Schlitz, M. J. (1997). Experimenter Effects and the Remote Detection of Staring. Journal of
Parapsychology, 61, 197-207.
Wiseman, R. & Schlitz, M. J. (1999). Experimenter Effects and the Remote Detection of Staring: An Attempted
Replication. The Parapsychological Association 42nd Annual Convention. Proceedings of Presented
Papers (pp. 471-479). Durham, NC: The Parapsychological Association.
Müller, Schmidt & Walach
The Parapsychological Association Convention 2006
99
Wiseman, R. & Smith, M. D. (1994). A Further Look at the Detection of Unseen Gaze. In: D. J. Bierman (Ed.), The
Parapsychological Association 37th Annual Convention. Proceedings of Presented Papers (pp. 465-478).
The Parapsychological Association.
Wiseman, R., Smith, M. D., Freedman, D., Wasserman, T. & Hurst, C. (1995). Examining the Remote Staring
Effect: Two Further Experiments. 38th Annual Parapsychological Convention. Proceedings of presented
papers (pp. 480-490). Durham, NC: The Parapsychological Association.
Zerssen, D.v. (1976). Die Befindlichkeitsskala: Manual. Weinheim: Beltz Test
