Remote Staring Detection - Experiment
ABSTRACT: An experiment designed to examine potential relationships between Remote Staring Detection, Belief in the Paranormal and personal Self Consciousness. Warning - This was my BSc (Hons) dissertation and is obligatorily very lengthy and probably rather dull!! ;) Remote Staring Detection
my Undergraduate experiment BSc (Hons) Psychology

Most people can relate to having had at some point, the feeling of being stared at by an unseen observer, turned around and found that they were indeed correct. It appears that this strange, yet rather commonplace experience -estimated as having been experienced by 68-94% of the public (Braud, Shafer and Andrews, 1993) - has been ubiquitous through time and has been the subject of scientific scrutiny for over a century. A variety of laboratory experiments aimed at testing the REALITY of this very phenomenon have resulted in mixed results and endless debate over methodology. The feeling of being stared at some would say, is simply a statistical anomaly, and a mere myth to others; but after a hundred years of research, this “feeling” is still difficult to prove as a real psi-related phenomenon and is even harder to debunk. This potentially, unrecognised human ability has been termed remote staring detection (the term psi is used in parapsychology to refer to any extra-sensory perception (ESP) and psychokinetic phenomena).

In 1898, Titchener published an article declaring that after running a series of experiments in Cornwall University, “the feeling of being stared at” was nothing more than a chance observation. He purported that his experiments yielded negative results, suggestive of no true remote staring detection effect. This may well be true, however, we shall never know as Titchener failed to report any details on those experimental results. Titchener reasoned that the “feeling of being stared at” was due to a combination of superstition and nervousness in social settings, resulting in the nervous person attracting the gaze of another, looking around and then noticing this observer.

At Stanford University, Coover (1913) also attempted to study the “feeling of being stared at" using himself as the starer (the person who directs their attention to, and stares at a recipient) and 10 subjects who tried to determine which of 100 trials he was staring at them. Coover was seated behind the subject in the same room and each subject sat in front of him with their eyes closed and covered with one hand. The roll of a die was used to determine the order of stare/ non-stare trials, each lasting 15-20 seconds. Echoing Titchener's research, Coover found that subject scored much at chance level, (50.2% accuracy comparing to 50% accuracy at chance level).

In contrast to the findings of both Titchener and Coover, Poortman (1959) of Leyden University found that in his experiments there was a 59.55% accuracy rate of correct guessing. The design of this experiment was different to the previous studies as he served as the “staree” (the person who is the recipient of a stare) and a colleague who sat behind him either stared or looked away. The starer was seated in one of two open, but adjoining rooms. Poortman sat in the other with his back to the starer.

Although, there appeared to be mixed results and speculation as to what these experiments suggested, there is no reason to presume that a true remote staring detection had occurred in any case, as none of the experiments were well controlled. This allowed for subtle sensory cues, (such as the sound of the starer sub-consciously leaning forward or breathing more heavily on a stare trial) to give the staree systematic, subtle sensory clues. For true remote staring detection to occur there must be no information regarding the nature of the trial given to the participant acting as stare.

To eliminate this obvious problem in the research, Peterson (1978) employed the use of a one-way mirror between two closed cubicles, so that the starer could see the staree but not the other way around. The staree was also required to listen to white noise through headphones in order to prevent auditory cueing. The periods of time where staring and non-staring occurred was randomly determined and over the 36, experimental sessions (each of which lasted 6 minutes) the staree pushed a button whenever he or she felt they were being stared at. Each session contained three 30-second staring periods and the times at which the button was pushed were recorded and compared to the timing of the staring. Peterson's results offered a significant remote staring detection effect, with an accuracy rate of 54.86%. Since the trials are either stare or non-stare, correct guesses should average out at 50% accuracy, i.e. people should, by chance, guess correctly on half of the trials and erroneously on the other half.

A further tightening of experimental controls was achieved by Williams (1983) where a closed circuit television system was used with monitor and camera 60 ft apart and in separate closed rooms. Again, randomly determined stare/non-stare trials were run – 52 staring trials and 52 non-staring; each lasting 12 seconds each. A more subtle yet significant result was obtained – 51.31% accuracy of scoring.

Most recently, Braud, Shafer and Andrews (1993) conducted an experiment in which closed circuit television equipment was used in a similar way as Williams (1983) did in her experiment. There were two groups of 16 subjects, all of whom were blind to the order of trials. The subjects' responses to stare/non-stare trials in this experiment were however, monitored via their skin-resistance so as to measure an unconscious reaction, rather than making conscious guesses. Furthermore, “training” was given to one group of subjects and not the other. The training largely consisted of intellectual and experiential exercises with other people involving staring in to the other person's eyes and “feeling connected”. Braud, Shafer and Andrews' experiment revealed impressive data suggesting that not only was there a remote staring detection effect but that those who had received the “connectedness training” were markedly less aroused (45.45% accuracy) and affected during staring trials than those who did not receive training (59.38% accuracy in scoring).

If we are to presume that humans do possess an ability (however subtle) to detect the stare of an observer by means beyond the known senses, then it makes sense to understand its purpose. If we are able to detect the presence of someone else's intentional staring, then perhaps the advantages for self protection and defence against an attacker are of an evolutionary nature; thus allowing self-preservation from physical injury, rape and even death. Many a war anecdote, has been heard mentioning the “powerful gaze of the enemy” that allowed a soldier to avoid being wounded. The potential for remote staring detection to influence and manipulate social interactions would seemingly rely upon an awareness of the self, both internally so as to detect an others' stare; and an awareness of others' ability to “feel” the attention. In order for someone to detect a stare by means of ESP, they would have to be adept at noticing subtle internal cues and changes. Furthermore, a belief in remote staring detection is required in order for those cues and changes to be interpreted as a response to a watchers stare. Similarly, to make the best use of remote staring detection, one would have to understand what conditions are required for their stare to be noticed or unnoticed, depending of course, on intention.

Self-consciousness is a trait that can be measured and used to predict behaviour in private, public and in social situations hence the self-consciousness trait is defined as having three constructs; private self-consciousness; public self-consciousness and social anxiety. Private self consciousness is seen in the tendency of a person to direct attention on to their inner feelings and thoughts; public self-consciousness is described as the awareness of he self as a social object and thirdly, social anxiety demonstrates the intensity of discomfort of a person while in the company of others (Fenigstein, Scheier and Buss, 1975).

The assessment of individual differences in self-consciousness was enhanced in 1975 by the construction of a questionnaire by Fenigstein, Scheier and Buss which consisted of 23 items split across the three factors discussed: 10 questions testing private self-consciousness, 7 testing public self-consciousness and 6 testing social anxiety. Although the questionnaire demonstrated itself to be a reliable predictor of social behaviour, a few criticisms came from other researchers who later, made modifications to the original scale according to their findings. Burnkrant and Page (1984) suggested that certain items be dropped over issues of ambiguity and Mittal and Balasubramanian (1987) suggested that both public and private self-consciousness consist of two sub-constructs. Public, comprising of Style consciousness and Appearance consciousness ; and private self consciousness entailing an Internal-state awareness and Self reflection . For details of the criticisms and break down, see Buss, (1980); Burnkrant and Page, (1984) and Mittal and Balasubramanian, (1987).

The process of self –examination allows a person to gain insight into their own thoughts and actions whilst enabling them to recognise what motives and defences they are prone to. Self-awareness, as Argyle (1969) suggests, is of great importance in social interaction and assessments methods like the Fenigstein, Scheier and Buss self-consciousness scale are successfully employed to predict such cognitions and behaviours i.e. aggression (Scheier, 1976) and personal use of cosmetics (Miller and Cox, 1982). Similarly, focusing attention on to the self via external means such as mirrors and cameras has been found to influence such factors as self-esteem and again, aggressive behaviour (Ickes, Wicklund and Ferris, 1973). People who tend to seek self-insight- not surprisingly- have a tendency to "“turn inward” and introspect, focusing their attention on their inner private self and have a greater awareness of their personal reactions to such attention directing stimuli such as camera and mirrors. It is in this reasoning that the study of remote staring detection potentially relies on the subjective awareness of internal thoughts and feelings of those participating in the experiments. The mere presence of a camera or mirror may well serve to arouse self-conscious feelings in the participants regardless of whether a staring or non staring session is in progress. Interestingly, Fenigstein, et. al. suggests that highly self-conscious people are more likely to be successful in meditation techniques where their attention is required to be internally focused.

The extent of personal belief in the paranormal has been of great importance within the realms of parapsychological research; that is the varying beliefs of the general public and scientists –both parapsychologists and sceptics. According to Schmeidler, 1985, the bases of the carrying beliefs are generally thought to rely upon four explanatory categories:
• Beliefs serve to fulfil some psychological need.
• Beliefs are formed under the influence of a doctrine one accepts or of a person one respects i.e. the teachings of a particular religious group.
• Beliefs are based upon the outcome of intrepid analysis of relevant research.
• Beliefs originate from personal experiences which have remained unexplained by means of the current scientific paradigm.
Blackmore (1984) noted that 36% of the general population believed in extrasensory perception (ESP) and of those, 44% believed themselves to be experiencers of the phenomenon. More recent studies on the bases of paranormal belief have suggested that personal experience of phenomenon is highly correlated with belief and that this belief, in turn influences behaviour, such as frequency of attendance at “clairvoyant” readings (Roe, 1996; Irwin, 1993). Alternatively, disbelief in the paranormal is highly correlated with a lack of experience (McClenon, 1982). The relationship between belief in the paranormal and such experiences has endured much speculation; Lawrence (1993) suggests that these bases of belief and varying degrees of belief influence the control over “psi” related phenomena in experimental environments

Of course, the role of expectancy has much to do with the interpretation of anomalous experience; e.g. some one who is a believer in ESP is likely to attend events relating to that subject (i.e. “psychic readings”) and interpret ambiguous results to fit that expectancy, thus reinforcing their belief. Similarly, a sceptic who attends such an event will be likely to interpret the same ambiguous information as no evidence on which to change their belief. Non-believers who experience similar phenomenon to believers are more likely to rationalise the event and seek a more conventional explanation than those already believing in the existence of the paranormal. Not surprisingly, believers tend to have had more experiences than non-believers; however, the direction of this relationship is questionable.

The terms “sheep” and “goat” have, over the years, come to represent believer and non-believers in parapsychological phenomena. This terminology was first used by Scmeidler and McConnel (1943) to term subjects who believed in the possibility of ESP within the confines of their experiment as “sheep” and those who did not, “goats”. Since then, the conceptual boundaries have become very blurry and the “sheep/goat” terms are used to refer to belief/non belief in anything paranormal including the existence of extra-terrestrials, astrology and fairies. For the purposes of clarity, most researchers are confining the terms to parapsychological phenomena, alone.

In 1993 Thalbourne and Delin, constructed a scale to measure the extent of individual belief in the paranormal. There scale was a development of the Australian Sheep-Goat Scale (Thalbourne and Haraldson, 1980) which was used to measure individuals' extent of belief and experience of parapsychological phenomena. The scale was later modified from the individual responses on a visual analogue scale to a Lickert scale after criticisms of ambiguity in interpretation, Roe (1998). This experiment was designed to test for the presence of remote staring detection ability in the participants and also to examine the possibilities of relationships between their personal self-consciousness and belief in the paranormal and how that might affect their scoring on the detection experiment.

This study is designed on the basis of previous research which suggests a reality of remote-staring detection and the influence of individual differences on the accuracy of such an ability. The experiment is designed to test for the ability of people to detect, specifically the unseen stare of the experimenter, without sensory cueing and experimenter influence; and to examine any links between remote staring detection performance, personal self-consciousness and belief in the paranormal.

It is hypothesised that if there is a remote staring effect, that his will correlate with the extent of belief in the paranormal and the degree of personal-self consciousness, however, no specific predictions about the direction of any correlation is made.


This experiment was a within-subjects design and consisted of 3 factors: accuracy of guesses, belief in the paranormal and personal self-consciousness. The experiment essentially had two conditions- stare and non-stare resulting in accuracy of guesses for both stare and non-stare conditions, being a dependant variable.

38 participants (32 female and 6 male) between the ages of 18 – 68 who responded to various advertisements of the experiment participated. An e-mail regarding the study and the need for participants was circulated to the level 4 honours psychology class and an advert seeking participants was presented to the Scottish Society for Psychical Research- see appendix for details. Most participants volunteered from either of these groups, but ten were acquired via word-of-mouth from university students.

Two separate rooms which are built parallel to one another, in the Department of Psychology, at the University of Glasgow were used in this experiment; one for the experimenter (Starer) and one in which the participant (Staree) resided throughout. A blackout curtain was used against each room's window to minimise the interference of outside distracters. Similarly, a portable CD player (ref: Aiwa; EX35) was used to play a CD of white noise to minimise distracter sounds . The volume was kept at a constant moderate level throughout every trial, and was the same for every participant. A closed-circuit-television system was used comprising of a camcorder (ref: JVC, GR-AX860 VHS-C); a video cassette recorder (ref: NV-SD230B VHS); and a 14” television monitor (ref: Panasonic TC-1453R). Five E180 video cassettes were used to record 32 of the experimental sessions; the first six failed to be recorded due to a minor technical fault. The camcorder was mounted on a tripod (ref: Velbon D500) which rested on a desk adjacent o the stare. One red and one green diode was use to signal the beginning and the end of trials, and was controlled by two respective buttons in the experimenter's room.
A two switch button-box (built within the psychology department) was used connected to an electronic timer/counter (ref: Campden Instruments LTD, timer counter 566). This equipment recorded the number of times the participant pushed the button where they were -indicating when they had a “feeling of being watched”, in each trial. The redundant button was covered with masking tape. A large piece of opaque cardboard was used by the experimenter (acting as “starer”)to cover the television monitor screen during the non-stare trials. A clock was strategically positioned by the equipment to time the trials.
An LED indication unit incorporating high intensity, 10mm LEDs was used to indicate the beginning and end of trials. A green to indicate beginning and red to indicate end – the LEDs were switched on and off by the experimenter via connecting switches.
Experimenter/ Starer Room.
Television monitor
LED control unit
Cardboard screen

Participant/Staree Room.

Camera & tripod
CD player
Button box
Red/green LED unit.

Room 216. Experimenter/Starer room
Television monitor
LED controls
(Leads & cables visible)
Cardboard screen

Room 217. The Participant/Staree room
CD player
Button box
Red/Green LED unit
An information and consent form were provided for each participant along with a feedback sheet, on which participants could add there name and contact details if desired. See appendix for copies of these. A score sheet for indication of decision on each trial, were provided for each participant along with a pen. The experimenter had a score sheet to record trial data including trial type (stare/non stare), number of button presses per trial, additional information/ notes (see appendix).

A Private Self-Consciousness Questionnaire (Fenigstein, Scheier and Buss, 1975) and a Paranormal Belief Scale (Thalbourne and Delin, 1993) were provided for each participant. The private self-consciousness questionnaire was designed to measure individual differences in self-consciousness and comprised of 23 items split in to 3 constructs; private self-consciousness, public self-consciousness and social anxiety.

The paranormal belief scale (Thalbourne and Delin, 1993) comprised of 18 items designed to measure the extent and basis of individual belief in the paranormal. See appendix for both questionnaires

Using a computer programme (see appendix), written at the Koestler Parapsychology Unit the University of Edinburgh , fifty sets of 32 randomised and counterbalanced stare/non-stare trials were produced so that each participant completed a set of 32 trials without risk of the experimenter unintentionally indicating a sequence bias. Every set comprised of 16 stare and 16 non-stare trials randomised as two groups of either of the four possible pairings; stare/ non-stare; non-stare/ stare; stare/stare; non-stare/non-stare (see appendix).

The participants completed the experiment one at a time and independently of the other participants. After being shown the experimental equipment that was in use, the participants were seated in the second, “staree” room in front of a wooden-topped desk with the camera pointing directly at their torso and head areas. On the desk, there was a button box to the left of the participant and the red and green diodes just in front of them.
There was an information and consent form, one copy of each questionnaire, a response sheet and a pen. The participant was asked to make themselves comfortable and left to acclimatise for a few minutes as they read and signed the consent form at their leisure. Each participant was then asked to complete the two questionnaires in their own time and call the experimenter when finished. At this point, the participant was informed of the background research of remote staring detection and the aim of the experiment and how they were to participate was explained. The participant was then asked to select at random, one of the undisclosed randomised and counterbalanced trial sets. The experimenter took it from them without either party opening it out and seeing the order.
The participants were then given an opportunity to ask questions about the procedure and once settled, they were prepared to begin testing. At this point the playing of a compact disc of white noise was initiated.

The beginning of each 30 second trial was signalled by the illumination of the green diode in front of the participant; the diode was activated by the experimenter (starer) and remained on for approximately two seconds. At this “start” signal, the experimenter either stared at the image of the participant on the screen for the duration of the trial or, if it was a non-stare trial, the cardboard screen was positioned before the green light was illuminated and the experimenter looked away from the equipment entirely and tried to think of things other than the participant him/herself. Once 30 seconds had passed, the red diode was illuminated by the experimenter for approximately two seconds (as with the green), indicating the end of the trial and that the participant could relax. The next trial would begin at the next green light and so on. The experimenter attempted to leave only a 3-5 second gap between the red “end trial” signal and the green “start trial” on moving on to the next trial.
The order of sequence of stare, non-stare trials was determined exclusively by the trial set selected by the participant and the 30 second trial times were determined by the experimenter making use of a clock to the right of the equipment during a non-stare trial or via the on-screen clock during the stare trials.

Before each trial the experimenter used the numbered score sheet and noted the type of trial about to feature. After the red (end) signal was given, the number of button pushes for that trial was recorded alongside that trial. Any distractions ( i.e. noise from other experimental rooms in use) or unusual activity was also noted as foot notes on the sheet such as (unfortunate) static and mild electric shocks on the experimenter's part.

After the completion of the 32 trials the, the participant was thanked and given confectionery and a “Kirsty Voucher” (see appendix) styled thank you note from the experimenter, and again any queries about the experiment and related research were discussed.

The descriptive statistic show that the mean number of hits n the experiment did not differ greatly from chance (15.6316) and is not significant. The difference between the mean scores of hits in stare trials and hits in non-stare trials, is however rather large (stare: 6.39; non-stare: 9.23). The mean score for the original private self-consciousness scale was 81.1, and for the total revised scale, 67.8. The means for the five sub-factors of the scale were all similar to one another especially style consciousness, social anxiety and internal-state awareness; 14.6; 14.68; 14.92, respectively. Appearance consciousness and self reflection averaged at a slightly lower score- 11.23 and 12.36, respectively. The mean score for Thalbourne's paranormal belief scale was 53.86. The button presses were low for both stare and on-stare conditions avarageing at 1 button press per subject, per condition.

Descriptive Statistics

N Minimum Maximum Mean Std. Deviation
PSQtotal 38 59.00 102.00 81.1053 10.4125
total PSQr: total - 7,9,12,23 38 47.00 84.00 67.8158 8.6770
public sc - style consciousness 38 8.00 20.00 14.6053 2.7365
public sc –appearance consciousness 38 5.00 15.00 11.2368 2.5408
social anxiety 38 7.00 20.00 14.6842 3.0589
private sc- internal state self awareness 38 11.00 20.00 14.9211 2.4427
private sc- self reflection 38 4.00 18.00 12.3684 3.2334
TPBtotal 38 18.00 84.00 53.8684 13.7176
no. stare trial hits 38 .00 11.00 6.3947 2.5099
no.hits in non-stare trials 38 5.00 16.00 9.2368 2.7057
Total hits in experiment 38 11.00 23.00 15.6316 3.1829
total number of button presses in stare trials 38 .00 13.00 1.0263 2.7360
total number of button presses in ns trials 38 .00 18.00 1.0263 3.4601

PSQ total – average total score for all 23 original items on the personal self-consciousness scale.
Total PSQr – total of revised questionnaire. Total of items excluding 7, 9, 12 and 23.
public sc - style consciousness – total score for items 2, 6, 14 and 19.
public sc -appearance consciousness – total score for items 11, 17 and 21.
social anxiety – total score for items 4,8,10 and16.
private sc - internal state self awareness – total score for items 3, 13, 20 and 22.
private sc - self reflection – total score for items 1, 5, 15 and 18.
TPBtotal – total score for all 18 items in Thalbourne's Paranormal belief scale

The original Fenigstein, Scheier and Buss self-consciousness questionnaire comprised of 23 items which were were grouped under three factors: Private self-consciousness, Public self-consciousness and Social anxiety.

Private self consciousness was made up of 10 of the items:1, 3, 5, 7, 9, 13, 15, 18, 20, 22.
Public self-consciousness was made up of 7 of the items: 2, 6, 11, 14, 17, 19, 21.
Social anxiety was made up of the items: 4, 8, 10, 12, 16, 23.

Each item was rated on a scale of 0-4; with 0 meaning “extremely uncharacteristic” and 4 meaning “extremely characteristic”.

The data from the questionnaires was recoded from 0-4. to 1-5 for ease of analysis and the responses for each item were added to give total scores for each sub-factor and overall totals. Furthermore, due to conceptual ambiguities and queries of relevance, the Fenigstein, Scheier and Buss self-consciousness questionnaire, was later modified: the three primary factors were broken down into 5 sub-factors and items 7, 9, 12 and 23 were dropped (Buss, 1980; Burnkrant and Page, 1984; Mittal and Balasubramanian, 1987). The five sub-factors and their construction in the revised scale are:

Private self-consciousness : Internal state awareness [items 3, 13, 20 and 22]
Self reflection [items 1, 5, 15 and 18]

Public self-consciousness : Style consciousness [items 2, 6, 14 and 19]
Appearance consciousness [ items 11, 17 and 21]
Social anxiety : - [items 4,8,10 and16]

Due to this varying application of the scale, reliability tests were performed on the total score for the original 23 items, the 5 individual sub-factors with exclusion of items 7, 9, 12 and 23, and the total for the 5 sub-factors.

Alpha co-efficient for self-consciousness (23 item): 0.7909
Alpha co-efficient for self consciousness (excluding items 7, 9, 12, 16): 0.7596
Alpha co-efficient for Private self-consciousness (Internal state awareness): 0.4069 *
Alpha co-efficient for Private self-consciousness (Self-reflection): 0.679
Alpha co-efficient for Public self-consciousness (Style consciousness): 0.758
Alpha co-efficient for Public self-consciousness (Appearance consciousness): 0.6676
Alpha co-efficient for Social anxiety: 0.618
Alpha co-efficient for Thalbourne's paranormal belief scale – 0.9278

The alpha coefficients suggest that the reliability of the modified questionnaire did not differ greatly from the original 23-item scale. The questionnaire was found to be reliable in both forms and that the modifications to the scale made little difference within this experiment- probably due to the relatively small sample of 38 participants, however the further removal of items 8 and 18 may improve the alpha reliability co-efficients slightly. Although the questionnaire appears reliable, the alpha co-efficient for Private self-consciousness (internal state awareness) * was rather low in this sample.

8) I have trouble working when someone is watching me.
18) I sometimes have the feeling that I am off somewhere watching myself.

Item 8 is perhaps has more impact on the participants in this experiment as it is specifically relevant to their situation and may have influenced the participants to down-play their feelings somewhat. Item 18 perhaps is rather ambiguous, and a few participants did remark on their confusion over it's meaning.

The analysis suggests- although not a significant finding- that overall, people with high personal self-consciousness tended to score a lower number of hits across both stare and non-stare trials (Pearson's correlation of -0.139; sig. 0.4, 2-tailed) than those with a higher self-consciousness. The sub-factor of public self-consciousness; style consciousness did, however significantly correlate with number of hits in non-stare trials (-0.407; sig. 0.011 2-tailed), suggestive of a link between accuracy of detecting the absence of a stare and the level of style consciousness of people. The relationship is a negative correlation implying that the higher the style consciousness, the lower the accuracy of scoring on non-stare trials and vice versa. As with the self-consciousness trait, a higher or stronger belief in the paranormal correlates with a lower scoring of hits overall (correlation -0.139; sig. 0.42, 2-tailed). See appendix for table of correlations .

There was no significant relationship between belief in the paranormal and personal self-consciousness (r=0.09).

Each participant experienced 16 stare trials and 16 non-stare trial, giving 32 trials over all. Taking a “hit” to mean a correct guess and a “miss” to mean an incorrect guess, at chance level, participants were expected to average at 16 hits out of 32 and the overall mean for hits in this experiment was 15.63 yielding a 48.8% accuracy rate. Participants, overall in this experiment appear to have scored just below chance and with a fairly small sample this is quite likely to have risen by chance and not all that surprising.

There was no significant remote staring detection effect for this sample of participants, overall. However, interestingly the means for total stare trial hits and total non-stare trial hits differs significantly. To score at chance level, participants would have to average at 8 correct guesses and 8 incorrect guesses for both staring and non-staring trials. The mean score for hits on stare trial was 6.39 ; 39.97% accuracy rather than 50% and the mean score for hits on non-stare trials (that is, where participants correctly guessed they were not being stared at) 9.24 ; 57.73% accuracy.


A paired samples t-test revealed that the difference between the means for stare and non stare trials was significant, (t= 4.236; df.=37; p= 0.000) however, it is possible that this difference could have arisen by chance.
The responses to stare and non-stare trials as measured via button pushing, yielded very little data as rather few participants chose to make use of the button. The mean number of button pushes for both stare and non-stare trials was 1.0263. Resultantly, no further analyses were performed on the button pushing data.
The experiment conducted, attempted to explore the possibility of remote staring detection and any relationships it had with self-consciousness and belief in the paranormal. The experiment was designed to avoid the possibility of sensory cueing; by the use of CCTV across separate rooms, masking noise and the strict double-blind randomisation of trials engineered to avoid the possibility of guessing strategies causing a bias and experimenter influence.

In the past, remote staring detection studies produced mixed results; some suggestive of an extra-sensory ability and others not, yielding data that suggests no such ability. The lack of supporting data in these studies, perhaps implies the non-existence of remote staring detection. Most studies have concentrated on the percentage of correct guesses in stare trials under the supposition that, if remote staring detection is real, it would be unrealistic to expect people to detect the absence of staring. However, in this study the participants were more reliably detecting (or guessing correctly) on non-stare trials- they appear to be more “aware” of the absence of a starer. If this finding can be reliably replicated then the ability to detect an absence of staring must have a significant role in the understanding of ESP and the assumption that the detection of an absence is unlikely- is wrong.

If there is, as many studies suggest a remote staring effect; the effectiveness of the ability is thought to be affected by various factors: Experimenter effects, individual differences (both starer and staree), situation and environment. Braud, Shafer and Andrews (1993) suggest that remote staring detection relies on an unconscious awareness of an other's attention and is more reliably measured via autonomous responses i.e. skin resistance. With this in mind, perhaps the deliberate conscious guessing of participants in this experiment, allowed for a plethora of cognitive interference. Perhaps the true measure of remote staring detection via changes in skin resistance could well be enhanced by keeping the participants blind to the purpose of the study – that is that they are not aware of the possibility of being watched, and so not aroused by auto-suggestion.

Although there appeared to be no significant relationship between individual differences of self-consciousness, belief in the paranormal and accuracy of guesses, the self-consciousness and belief in the paranormal of the starer (experimenter) was not assessed. Previous research has suggested that the role of the starer is highly important. If some people are more effective at detecting the stare of an unseen observer, it follows that some people may be more susceptable to having their staring detected than others. Assuming that the feeling of being stared is one which causes psychological discomfort, i.e we feel uncomfortable when some one is watching us, then what are the factors that determine an “uncomfortable stare” ? Perhaps some people can have a “prickly” effect on us – make us feel uneasy by staring at us from an unobservable place, and others a calming effect by staring at us. If this is a possibility then it makes sense to say that in the absence of a “calming stare” starees would become more agitate.

The “training” which Braud, Shafer and Andrews (1993) gave to half of their subjects involved them becoming comfortable with the gaze of some one else- allowing them to feel “connected” to that person; these subjects were not as aroused by staring trials as those who did not receive training. Therefore, on the basis of their findings and of the results of this experiment, it would be interesting to examine the occurrence of calm and “connectedness” feelings in relation to detecting the watch different starers. Similarly, the role of feedback to the starees was considered important in the Peterson study; reporting a higher incidence of accurate stare detection among subjects who were given receiving immediate feedback of their guesses to help them identify the internal signals and cues that relate to stare trials. Perhaps the feeling of being watched is different depending on who the starer is and successful detection of their staring, requires a kind of “training” too- a familiarisation with their particular effect. Although, the participants in this experiment could reliably discriminate between stare and non-stare trials, we could speculate that perhaps the starer in this experiment did not produce the antagonistic feelings expected by the participants; hence not show the expected direction of discrimination.

On the other hand, The fact that the outcome of this experiment could have been expected by chance, suggests that it might not be representative of an unusual cognitive ability at all. We can expect that it would be levelled out to chance level through replication of the experiment and possibly in further meta-analyses of remote staring detection studies in general.

Lawrence (1993) suggested a link between the extent of belief in the paranormal and performance on psi- related tests and a similar trend was noted here where high believers in the paranormal scored a lower accuracy on the experiment than low believers did. This trend was not, however significant within the confines of this experiment. The self-consciousness trait of an individual was found to be of a similar trend, suggesting a need for further research into a link between personality and belief in the paranormal. Both subtle trends were of the same direction but contrary to common expectation, which suggests that high self-consciousness and belief in the paranormal would yeild a higher accuracy in remote staring detection than low self-consciousness and belief.

If there does exist, an underused extra-sensory ability to detect the stare of another, then the implications are phenomenal. From the everyday usage of avoiding those who intend harm, perhaps resulting in the decline of petty crime; to the implications remote staring detection has on our evolution and relationships with other humans and other species- predatory and prey.


Argyle, M. (1969) Social interaction . New York: Atherton Press.

Blackmore, S. (1984). A postal survey of OBEs and other experiences. Journal of the Society for Psychical Research , 52, 225-244.

Braud, W.G., Shafer, D. and Andrews, S. (1993a). Further studies of autonomic remote staring, new control procedures, and personality correlates. Journal of Parapsychology , 57, 391-409.

Braud, W.G., Shafer, D. and Andrews, S. (1993b). Reactions to an unseen gaze (remote attention): A review with new data on autonomic staring detection. Journal of Parapsychology , 57, 373-390.

Burnkrnat, R. E. and Page, T.J. (1984). A Modification of the Fenigstein, Scheier, and Buss Self-Consciousness Scales. Journal of Personality Assessment , 48, 6.

Buss, A. H. (1980). Self-consciousness and Social anxiety . San Francisco: W.H. Freeman.

Colwell, J., Scroder, S. and Sladen, D. (2000). The ability to detect unseen staring: A literature review and empirical tests. British Journal of Psychology , 91, 71-85.

Fenigstein, A., Scheier, M., & Buss, A. (1975). Public and Private Self-Consciousness: Assessment and Theory. Journal of Consulting and Clinical Psychology , 43, 4, 522-527.

Ickes, W. J., Wicklund, R. A., & Ferris, C. B. (1973). Objective Self-awareness and self-esteem. Journal of Experimental Social Psychology , 1973, 9, 202-219.

Irwin, H. J. (1993) Belief in the paranormal: A review of the empirical literature. Journal of the American Society for Psychical Research , 1-39.

Lawrence, T.R. (1993). Gathering in the sheep and goats: A meta-analysis of forced-choice sheep-goat studies, 1947-1993. Proceedings of Presented Papers: The Parapsychological Association 36 th Annual Convention, pp. 75-86.

McClenon, J. (1982). A survey of elite scientists: Their attitude toward ESP and parapsychology. Journal of Psychology , 46,127-152.

Mittal, B and Balasubramanian, S.K. (1987). Testing the Dimensionality of the Self-Consciousness Scales. Journal of Personality Assessment , 51(1), 53-68.

Peterson, D. M. (1978). Through the looking glass: An investigation of the faculty of extra-sensory detection of being stared at. Unpublished thesis, University of Edinburgh, Scotland.

Poortman, F.F (1959). The feeling of being stared at. Journal of the Society for Psychical Research , 40, 4-12.

Roe, C. A. (1996). Persuasion in the context of a psychic reading . Unpublished doctoral dissertaion. University of Edinburgh, Scotland.

Roe, C.A. (1998). Belief in the paranormal and attendance at psychic readings. Journal of the American Society for Psychical Research , 92 , 25-51.

Scheier, M. F. and Fenigstein, A., and Buss, A. (1974). Self-awareness and physical aggression. Journal of Experimental Social Psychology, 10, 264-273.

Scheier, M.F. (1976). Self-awareness, self-consciousness and angry agression. Journal of Personality, 44, 627-644.

Schmeidler, G. R. (1943) Predicting good and bad scores in a clairvoyance experiment: a preliminary report. Journal of the American Society for Psychical Research , 37, 103-110.

Schmeidler, G. R. (1945). Separating the sheep from the goats. Journal of the American Society for Psychical Research , 39, 47-49.

Schmeidler, G. R. (1985). Belief and disbeleif in psi. Parapsychology Review. 16 (1), 1-4.

Thalbourne, M.A. and Haraldsson, E. (1980) Personality charachteristics of sheep and goats. Personality and Individual Differences 1, 180-185.

Thalbourne, M.A. (1998). On the Australian Sheep-Goat Scale. Journal of the American Society for Psychical Research , 92 , 308-309.

Thalbourne, M.A. & Delin, P.S. (1993). A new instrument for measuring the sheep-goat variable: Its psychometric properties and factor structure. Journal of the Society for Psychical Research , 59 , 172-186.

Williams, L. (1983). Minimal cue perception of the regard of others:The feeling of being stared at . Paper presented at the 10 th Annual Conference of the Southeatern Regional Parapsychological Association, West Georgia College, Carrollton, GA.
The white noise CD has a spectral frequency of 1, and a bit rate of 1411kbps. The audio sample size was 16bit, two channels (stereo), with an audio sample rate of 44kHz. The white noise was edited using Cool Edit 2000 in a continuous block with no repeats to prevent any drop-outs. The CD was compiled (01/08/02) by Ian Baker at the Koestler Parapsychology Unit; Dept. of Psychology: The
University of Edinburgh.

I would like to thank John Morrison for building this LED indication unit for the experiment.

I would like to thank Dr. Paul Stevens and Ian Baker from the Koestler Parapsychology Unit for writing this programme.

© 2004

Copyright © 2005 Sinergy Group Media