Early experiments in precognition
THIS ENTIRE SECTION HAS NOW BEEN RE-WRITTEN AND IS READY TO BE PUBLISHED
Most of the important early experiments in precognition was conducted by by J.B. Rhine at Duke University during the 1930s. These tests involved Zener cards, and the participants would be asked to predict what order the cards would be in after they were reshuffled by the experimenters. In at least some of these studies the results were promising. One subject, a Hubert Pearce participated in several large studies in which he attempted to predict the order of literally hundreds of packs of cards.
Each of these decks consisted of twenty five cards depicting five geometrical symbols, each featured on five different cards. This means that pure luck would have led to a Pearce correctly predicting the identity of five cards in each of the decks.
However, he often scored significantly h igher. In one study, Pearce correctly
predicted an average of 6.3 cards per deck, and in another study he averaged 7.1 cards per deck.
While this is not the spectacular kind of result that you see in TV shows and movies, it is considered to be statistically significant. Whether or not these two studies were rare exceptions or not I have not been able to find out. If they were exceptions than this means little. If he did this on a regular basis, it would be more significant. And there were some statistical and methodological flaws with Rhine’s studies. Critics pointed out that the Zener cards may not have been properly reshuffled, leading to an inflated hit rate. Later studies, however, used random number tables instead of shuffling by hand, and this supposedly resulted in the shuffling being truly random.
Interestingly, the results of Rhine’s precognition studies declined over time, which led researchers to distance themselves from card tests and try different types of experiments. In the 1960s, a physicist named Helmut Schmidt convinced his employers at the Boeing Scientific Research
Laboratories in Seattle to support a study into the possibility that precognition may be an extant phenomenon. Schmidt’s study used a device that had four buttons below four coloured lamps. The test subject was asked to guess which of the four lamps would light up in a few seconds’ time, and to press the button below that lamp. Pressing the button triggered a random number generator which in turn selected and turned on one of the four lights. The study used three volunteers, who made a total of 63,066 guesses. On chance alone the volunteers should have had a hit rate of 25% of the time when predicting which lamp would light up. Instead, they were right a little more often than 26%, which is considered to be statistically significant given the very large number of guesses. There were, however, some external criticisms of the study, including University of Oregon psychologist Ray Hyman’s suggestion that the selection of the lights was not actually random. Schmidt countered this with some five million demonstrations showing that the machine’s choice of which lamp to light up was indeed random. Such critics were in the minority and many other parapsychologists created similar types of RNG/precognition experiments.
There was also a meta-study carried out by the Psychophysical Research Laboratories in 1989, which looked at over three hundred precognition studies carried out by over sixty researchers between the years 1935 and 1987. The studies found that the results were statistically significant, and therefore supported the existence of precognition. I have not been able to find figures for this particular meta-study so I do not know how significant the results were.
