The Sense of Being Stared At

Second, it enables people who are unusually talented to be identified, and thus opens the way to more detailed experiments.

Third, it enables people who do not do particularly well to practice and find out if they improve with experience. It may be possible to train oneself to perform well under these conditions. And this too would open the way for further research.

In these experiments people work in pairs, one sitting with his or her back to the other. In a series of trials, in a random sequence, the looker either looks at the back of the subject for 20 seconds, or looks away and thinks of something else for 20 seconds. The random sequence is determined by tossing a coin before each trial: heads means look; tails means don't look. The looker indicates when a trial is beginning by a tap, click, or beep, and the subject then guesses whether he or she is being looked at or not. Uniform mechanical clicks or electronic bleeps are better than taps because they rule out the possibility of subtle cues being transmitted through the strength of the taps. The looker records the result, and then tells the subject whether the answer was correct or not. The looker then tosses a coin to determine what to do in the next trial. And so on. The procedure is quite fast, and an average speed of two trials per minute is easy to achieve. The results are recorded on a simple score sheet....with two columns. Each trial is entered on a separate line, and correct answers are indicated by ticks, wrong answers by crosses....

...The results of these tests can be analysed statistically by means of the paired-sample t-test....

I have found it best to keep test periods fairly short, up to about 20 minutes, during which time forty or more trials can be done. For statistical analysis, at least ten separate test periods are desirable, either with the same pair of people or with different pairs of people....

For those who do not perform particularly well in initial tests, it is good to practice, doing 15- to 20-minute test sessions whenever convenient. This makes it possible for a learning process akin to biofeedback to occur, whereby various subtle sensations or methods of visualization are tried out in the attempt to find an effective way of telling when one is being looked at. If there is a tendency to improve with experience, it should be revealed by a rising proportion of correct guesses in successive sessions.

If and when sensitive subjects have been identified, many further questions can then be asked. Here are some straightforward examples:

1. How much difference does the looker make? Are some people much more effective as lookers than others?

2. Does the sense of being stared at still happen when the person is looked at through a window? Does it still show up when looked at from a distance, for example through binoculars? By experiments of this type, it should be possible to rule out the possibility that in tests in the same room the subjects are being influenced by subtle cues, such as the looker moving his or her head. If the effect still shows up at a distance, or through soundproof windows, then this would greatly strengthen the evidence for a direct influence of looks.

3. Does the ability show up when the subject's reflection is looked at in a mirror?

4. Does this ability show up when the subject is looked at using closed-circuit TV, with the looker and the subject in separate rooms, or even separate buildings?....

5. If it works on closed-circuit TV, then what about actual transmissions? In this case, the effect of distance can be tested over hundreds, or even thousands, of miles, using satellite links. If preliminary experiments show it works on TV, then live experiments could be done involving millions of viewers. Here is one possible design for a TV show. Four sensitive subjects are kept in separate rooms in front of TV cameras that are running continuously. Then, in a series of trials, viewers see one subject at a time in a randomized sequence. At the end of each trial, all four subjects press a button indicating yes or no. Viewers see a scorecard on which the number of right and wrong scores for each subject is registered. The sequence of trials need take no more than about 10 minutes. A computerized statistical analysis could be available almost immediately, and the rest of the program could consist of a discussion of the results and their implications. If sensitive subjects are available, there would probably be no problem getting this kind of experiment performed, as I have found talking to TV producers in Europe and America. Such experiments would make good television and arouse much popular interest.

6. How closely related is the sense of being stared at to telepathy? Does looking at someone have a greater effect than just thinking about them without looking? The way to find out is by experiment. For example, the experiment could be modified to include a third condition, in which the lookers think of the receivers but do not look at them. In other words, there would be three kinds of trial in random sequence: looking; not looking and not thinking; not looking but thinking. My own guess is that the effects of looking will be greater than just thinking."