From Abracadabra to Zombies | View All
Zener ESP cards
There are five kinds of cards: one continuous curve that makes a circle, two straight lines that cross in the middle at the perpendicular, three wavy parallel vertical lines, four straight lines that form a square, and a five-pointed star. Because of their distinctness, there should be no ambiguity regarding any symbol. A deck of Zener cards consists of five of each symbol. The cards would be shuffled and a receiver would then try to guess the cards that a sender would try to telepathically communicate. Or a subject might try to guess which card from the deck would be turned up next. Rhine recognized that this test couldn't distinguish telepathy from clairvoyance. That is, there would be no way to tell whether a receiver was receiving messages from the mind of the sender or was receiving information independently of the sender's mind. Of course, Rhine had no way of knowing whether the information was coming from the guesser's own subconscious mind, from the Pleiadians, the Akashic record, or even from Zeus or the local baker's mind. But he was sure he had a way to rule out chance as an explanation for any apparent information transfer in his experiments.
Since there are twenty-five cards in the deck and five kinds of cards, there is a one in five or 20% chance that any given card is on top of the deck or being viewed by a sender. Of course, these odds change as soon as a card is removed from the deck. The odds for each selection can remain identical only if each card is put back in the deck and the cards are thoroughly reshuffled after each selection. From the point of view of the guesser, however, as long as no feedback is given while going through the deck with a sender, each card has a one in five chance of being the card on top at any given time. A correct "guess" is called a "hit". Anything significantly higher than 20% hits in the long run would indicate that something other than chance is at work. In the short run, higher percentages are expected, on occasion, by chance. Thus, if a guesser got nine out of twenty-five correct (36%) going through the deck once, that would not necessarily be indicative of anything important. If the guesser got 36% correct over 100 runs through a deck of 25 cards (i.e., 2,500 guesses), that would indicate that something else besides chance is going on. Maybe you're psychic, maybe there is sensory leakage, or maybe you're cheating. (We know, for example, that some of the decks of Zener cards that were printed for Rhine's lab were very thin and translucent, allowing receivers to see through the card to identify which icon the sender was looking at.)
Rhine would become ecstatic when anyone was found who could do
significantly better that 20% in guessing. Some were so phenomenal (e.g.,
Adam J. Linzmayer, George Zirkle, Sara Ownbey, and Hubert E. Pearce, Jr.)
that skeptics assume there must have been cheating. Rhine denied it. He
didn't think he could be deceived and thought his testing methods were
adequate. Nevertheless, charges of cheating plagued Rhine. Nobody thought
Rhine was cheating but many thought he had been duped by his subjects
several times. According to Milbourne Christopher “there are at least a
dozen ways a subject who wished to cheat under the conditions Rhine
described could deceive the investigator" (Christopher 1970: 24-25). Also,
once Rhine took precautions in response to criticisms of his methods, he
was unable to find any high-scoring subjects to match his early phenoms
(Christopher 1970: 28). Rhine even used a magician to observe Pearce; his
performance sunk back to chance levels. When not so observed, his scores
were significantly higher.
Rhine was undaunted by criticism. He claimed in his first book (Extra-Sensory Perception, 1934) that he’d done over 90,000 trials and could justifiably conclude that ESP is “an actual and demonstrable occurrence.” There were attempts to duplicate these trials at Princeton, Johns Hopkins, Colgate, Southern Methodist, and Brown without success. Critics could not find evidence in Rhine’s report that he was as systematic and careful as one would expect a scientist to be who was making such an extraordinary claim. There was no evidence, for example, that Rhine even realized how important it was to discuss how the cards were shuffled when doing the tests. He showed no awareness that the 1 in 5 odds that represent pure chance with the Zener deck could change if the cards were not perfect (which they weren’t), weren't randomly presented (which they weren't), and that certain strings of guesses would be ruled out with a universe of only 25 entities. For example, no one in his right mind would guess 6 or more circles in a row because the deck only contains 5, but in a truly random distribution of circles, 6 or more items of the same kind would be expected to come up occasionally. It's probably unlikely that there were many guesses of 4 in a row or even 3 in a row, yet in a truly random distribution of 5 possible symbols 3 and 4 in a row of any given symbol would occur with calculable frequency. In fact, given the small size of the deck, the actual odds of guessing any given item might be significantly different from the theoretical odds, which are based on the assumption of extremely large numbers of trials where each item always has exactly the same odds of coming up. Gamblers "and subjects in ESP experiments tend not to repeat a given choice immediately after it has been made" (Zusne and Jones 1989: 176) and in feedback trials this increases the odds of guessing correctly so that chance might be closer to 6/25 rather than 5/25 (Gatlin 1979). In real life, with only 25 cards to remember, card-counting and biased selection were certainly likely and frequent occurrences.
In fact, it is obvious that Rhine and his colleagues didn’t consider the relationship of theoretical probabilities with real probabilities. Others, however, had. In the 1930s, a magician by the name of John Mulholland asked Walter Pitkin of Columbia University how one determined the odds against matching pairs with five possible objects. Of course, Mulholland didn’t have a computer to do his dirty work for him, so he printed up 200,000 cards, half red and half blue, with 40,000 of each of the five ESP card symbols. The cards were mechanically shuffled and read by a machine. The result was two lists of 100,000 randomly selected symbols. One list would represent chance distribution of the symbols and the other would represent chance guessing of the symbols. How did they match up? Well, they didn’t. The actual matches and what would be predicted by accepted theoretical odds didn’t match up. The total number was 2% under mathematical expectancy. Runs of 5 matching pairs were 25% under and runs of 7 were 59% greater than mathematical expectancy (Christopher 1970: 27-28). The point is not whether these runs are typical in a real world of real randomness or whether they represent some peculiarity of the shuffling machine or some other quirk. The point is that Rhine assumed that statistical probability—which assumes true randomness and a very large number of instances—applies without further consideration to decks of 25 cards shuffled who knows how or how often.
Rhine and all other psi researchers have assumed that any significant departure from the laws of chance is evidence of something paranormal. While cheating should be of concern to paranormal investigators, there should be more concern with this assumption. Besides the Pitkin study, other studies have shown that even when no subjects are used there is significant departure from what would be expected theoretically by chance (Alcock 1981: 159). For example, Harvie “selected 50,000 digits from various sources of random numbers and used them to represent “target cards” in an ESP experiment. Instead of having subjects make guesses, a series of 50,000 random numbers were produced by a computer.” He found a hit rate that was significantly less than what would be predicted by chance. “If such significant variation can be produced by comparing random strings with random strings, then the assumption that any significant variation from chance is due to psi seems untenable (Alcock 1981: 158-159).
Another example of Rhine’s lack of sophistication with probabilities comes from the fact that when he found subjects who consistently scored below chance, he did not take this as what would be expected by the laws of chance. Rather, he took this to be evidence of a psychic phenomenon. He claimed that subjects who didn't like him would guess wrong to spite him (Park 2000: 42). Parapsychologists accepted his explanation for what they now call psi-missing.
Another indication that Rhine and his colleagues had
little understanding of how theoretical statistics should be applied in
the real world is revealed by their being puzzled by the fact that the
longer a successful subject was tested, the more his scores tended toward
a chance distribution. Rather than take this as natural
regression toward the mean (over time, all
subjects should move toward chance if nothing paranormal is happening),
Rhine and other parapsychologists explain regression away by saying that
it is due to the boring nature of the testing. They even have a name for
it: the decline effect.
Rhine did improve his testing techniques over the years, however. For example, he explained how it took some time before researchers realized that letting the subjects handle the cards or envelopes holding the cards opened the door to cheating. Also, in the early experiments, the experimenter and the subject were separated only by a screen. Later, they were placed in separate rooms or separate buildings to avoid the possibility of cheating or inadvertent communication by sensory cues. And Rhine was well aware that there was no way to distinguish in principle between telepathy and clairvoyance or precognition. Had he used a control, there would have been no way to prevent the control from being controlled by ESP or PK from some entity in this or some other universe.
In any case,
Rhine did not convince the scientific community that his research strongly
supported the reality of ESP, despite his claims that his subjects had been “carefully
witnessed” and that he had put into place “special conditions” that
“completely eliminates all chance for deception” (Christopher 1970). Zener
cards may have made it easy to calculate the odds against chance of any
psi performer, but it is a gross exaggeration to claim, as Dean
Radin has, that Rhine’s “statistical analysis is essentially valid” (Radin
See also decline effect, parapsychology, psi, psi assumption, and A Short History of Psi Research by Robert Todd Carroll.
books and articles
Brugger, Peter and Kirsten Taylor (2003). “ESP – Extrasensory Perception or Effect of Subjective Probability?” In Psi Wars, Getting to Grips with the Paranormal, Imprint Academic. (click here to see a copy of the article)
Gatlin, L.L. (1979), ‘A new measure of bias in finite sequences with application to ESP data’, Journal of the American Society for Psychical Research, 73, pp. 29-43.
How NOT to test your psychic ability (it is essential that a large number of trials be done to get statistically significant results)