The claim that you can feel a gaze on the back of your neck with no known sensory channel to carry it. It is the most-tested folk belief in parapsychology, and the fight over it has run for more than 125 years.
Documented
The oldest debunk in the field
Edward Titchener founded the literature in Science in 1898 after his students insisted they could feel unseen gazes; his laboratory trials came back negative. J. E. Coover at Stanford published the first full protocol in the American Journal of Psychology in 1913, 24:570, pairs taking turns as starer and subject, and pronounced the belief groundless. Surveys ever since report the same thing his students said: large majorities of people claim the experience.
Assessment: Both papers are documented and both found nothing. What they established is the baseline: the feeling is near-universal, and the early lab evidence was null. The literature did not stay null.
Stories
The asymmetric hit
Rupert Sheldrake made the stare his flagship experiment: tens of thousands of trials by 2005, schoolchildren in Britain, Germany and America, and more than 18,700 pairs at Amsterdam's NEMO science centre, where visitors tested themselves under a sign asking if they have eyes in the back of the head. The pooled hit rate sat around 55 percent against the 50 of chance, with a repeating shape: clearly positive scores on looking trials, dead-chance scores on not-looking trials.
Assessment: The pattern is genuinely odd and consistently reported. It matches what a real sense would produce, and it also matches a simple response bias toward saying watched. Both readings remain live in the journals, which is what keeps the topic unkillable.
Interesting claims
The skeptics replicated it, then disputed the dice
Colwell, Schroder and Sladen (British Journal of Psychology, 2000) ran Sheldrake's own sequences and got strikingly significant results, then attributed them to hidden structure in his counterbalanced randomizations. Marks and Colwell took that case to the Skeptical Inquirer; Sheldrake answered that thousands of coin-tossed, feedback-free trials showed the same effect. Lobach and Bierman's 2004 replications came back null. In CCTV versions the result tracked the starer: Wiseman got chance, Schlitz got significance, same lab, same subjects.
Assessment: The fairest verdict in the literature is a shrug with footnotes: a small effect, sensitive to method and experimenter, confirmed by Schmidt, Schneider, Utts and Walach's 2004 meta-analysis in the same British Journal of Psychology, and absent in better-controlled attempts. Neither side has run the decisive experiment.
Sources and further reading
Titchener, The feeling of being stared at, Science 8 (1898), 895-897, the founding debunk
Coover, The feeling of being stared at, American Journal of Psychology 24 (1913), 570-575
Sheldrake, The Sense of Being Stared At: Experiments in Schools, JSPR 62 (1998), 311-323, and the NEMO centre dataset
Colwell, Schroder and Sladen, The ability to detect unseen staring, British Journal of Psychology 91 (2000), 71-85
Marks and Colwell, The psychic staring effect: an artifact of pseudo randomization, Skeptical Inquirer (2000), with Sheldrake's reply
Schmidt, Schneider, Utts and Walach, two meta-analyses, British Journal of Psychology 95 (2004), 235-247
Lobach and Bierman, The invisible gaze: three attempts to replicate, Parapsychological Association (2004)
Sheldrake, The Sense of Being Stared At (2003), the primary text, and the Journal of Consciousness Studies special issue 12:6 (2005)
Worth asking
When the debunkers replicate the effect and then explain it away, which side actually won: the phenomenon or the explanation?