ESP Marvels Documented by Rhine

In his book New Frontiers of the Mind: The Story of the Duke Experiments, Duke University professor Joseph Rhine discussed his search for people who would perform highly on tests of ESP. Rhine discusses tests that are mainly tests of clairvoyance, not telepathy. (The term ESP refers to both telepathy and clairvoyance.)  In a laboratory test of telepathy, one person will typically attempt to transmit a thought (or will look at some image), and another person will attempt to receive that thought or guess what image the other person saw. In a laboratory test of clairvoyance, one person will attempt to perceive something that cannot be seen with his senses at that time. 

Rhine and Hubert Pearce

The tests Rhine describes were done with Zener cards depicted below. Each card in the deck would have five possible symbols, the symbols depicted below. Each symbol would appear in the deck of cards the same number of times. 

 

In a test such as the one shown above, a subject would attempt to guess what was the symbol of a card at the top of the deck. The pack of cards would be like a deck of playing cards, in that all of the cards have an identical appearance on one side, but different symbols on the other side. By looking at the top of the deck, you could not see what was the symbol you would see when the card was turned the other way. 

On page 74 Rhine discusses tests he did with A. J. Linzmayer. We read that in card-guessing tests using Zener cards in which the chance of guessing the card symbol correctly is 1 in 5, Linzmayer guessed nine consecutive cards correctly. We read that he did the same feat the next day. The chance of getting such a result with nine consecutive guesses of these cards is 1 in 5 to the ninth power, or 1 in 1,953,125. 

On page 76 we read that instead of guessing about 60 cards correctly in a series of 300 card guesses (about the result expected by chance), Linzmayer guessed 119 cards correctly. The probability of a result that good can be calculated using a binomial probability calculator such as the Wolfram Alpha binomial probability calculator. Below are the inputs you can use. That calculator uses the confusing term "endpoint." It would be better if it used the clearer term "number of successes."

As the screen above indicates, the probability of getting a result as good as the result reported for Linzmayer is less than 1 in 10 to the 14th power, less than 1 in 100 trillion. 

On page 78 Rhine says that he tested Linzmayer and saw him guess correctly 15 consecutive times, using the Zener ESP test cards. The probability of getting this result by chance is 1 in 5 to the 15th power, or 1 in 30,517,578,125, about 1 in 30 billion. When continued for a total of 25 guesses, the result was 21 correct out of 25. The probability of getting a result that good by chance is about 1 in 100 billion.

On pages 94 to 95 of his book, Rhine discusses an experiment with Zener cards done with Hubert Pearce. Pearce correctly guessed 25 consecutive times the symbol of the top card of a deck of Zener cards that was cut each time. The probability of getting this result by chance is 1 in 5 to the 25th power, or 1 in 298,023,223,876,953,125. Rhine says it was the most phenomenal thing he ever witnessed. On page 104 Rhine says that George Zirkle duplicated Pearce's feat of making 25 consecutive correct guesses when being tested with Zener cards. 

On page 100 Rhine says that June Bailey averaged between 8 and 10 correct guesses on "thousands of trials" in which the expected chance result is only 5 correct guesses. On the same page we read that T. Coleman Cooper "was able to score reliably about 8 hits per 25 over thousands of trials."

In a set of 2000 guesses with Zener cards, the expected chance result is about 400 correct guesses. Getting 8 hits per 25 in a set of 2000 guesses would mean a number of successes of about 640. (8 out of 25 is .32, and 2000 multiplied by .32 is 640.)  To calculate the probability of getting the results discussed above for Bailey and Cooper, you can use the inputs below in the Wolfram Alpha binomial probability calculator:

The calculator tells us that the likelihood of getting the results reported for Bailey and Cooper is less than 1 in 1,000,000,000,000,000,000,000,000,000.

On the next page we read this about May Frances Turner: "Her averages through many thousands of trials were in the neighborhood of 9." The probability of getting such a result by chance would be much smaller than the tiny probability just mentioned. 

The account below appeared on page 90 of  the April 15, 1940 edition of Life magazine, which during its heyday was one of the three or four leading weekly magazines in the United States. We read of tests with Zener cards that have five possible symbols, on one side of the card. The chance probability of guessing one of the cards correctly is 1 in 5. 

On page 92 the same Life magazine article tells us of the remarkable success in a long-distance test carried out over a span of 250 miles. It is a test that can be considered either a test of clairvoyance or a test of telepathy:

The wording is ambiguous, leaving us in doubt whether the 16 out of 25 result was obtained on two consecutive days, or whether it was one result of 16 out of 25 spanning two days. I cleared up the ambiguity by searching in the writings of Joseph Rhine, to find his account. It occurs on page 60 of his book Extra-sensory Perception. We read this:

So the results on the first three tests were:

Test 1: 19 out of 25 correct

Test 2: 16 out of 25 correct

Test 3: 16 out of 25 correct

Rhine does not do a good job of explaining how unlikely this result is. But by using what is called a binomial probability calculator, we can estimate that. The first three tests add up to being 51 successes out of 75, in a test in which the expected chance result is about 15 (which is one fifth of 75). Using the Wolfram Alpha binomial probability calculator, we can calculate the chance of that level of success, by using the inputs below:

The probability of getting a result as good as this by chance is calculated above. It is roughly 1 in 10 to the 19th power or 1 in 10,000,000,000,000,000,000.

The Life magazine article also discusses on page 95 tests of clairvoyance and ESP that Joseph Rhine did with the medium Eileen Garrett.  These were very unusual, in that the medium would go into a trance, and then start speaking as if she were another personality called Uvani. Often when this occurs with mediums, there is an impression of communication with some unearthly realm, such as an afterlife realm.  The original report of these tests appear at the beginning of this edition of the scientific journal Character and Personality, the December 1934 edition (Volume 3, Issue 2). On page 100 we read this report by Rhine:

"The experimentation with the Mrs. Garrett personality began on April 10 and lasted until April 28, approximately three weeks. During this period 14,425 tests or trials were given her in the normal state in clairvoyance and telepathy combined.

The work with the Uvani personality in clairvoyant and telepathic perception began on the 17th of April and lasted until the 25th. The amount of work per day, as well as the number of days, was limited by Uvani’s disinclination toward the experiments, which was in contrast to Mrs. Garrett’s willingness and patience. Only 1,575 trials were obtained with Uvani.

In all there were performed 16,000 trials at clairvoyant and telepathic perception. This number includes all the results, high scores and low. The most probable number of hits expected by chance for this number of trials would be 3,200 but the actual results were 4,018, or 818 hits above the chance mean. This gives an average per 25 of 6.3, and when evaluated for anti-chance significance, a value of X (i.e., deviation divided by probable error) equal to 24.0. This gives odds against the chance hypothesis of such a huge number—one with well over 50 digits—that it is beyond a moment’s question that chance is not the explanation."

Although he states very exactly the experiment's results, Rhine is not expressing very clearly how above chance these results are. But with a binomial probability calculator, a more clear idea of the improbability can be revealed. Using the Wolfram Alpha binomial probability calculator, the improbability can be calculated using the inputs shown below:

The tests involving guessing the symbol shown on a Zener card with 5 possible symbols. The tests were done with a deck of cards with an equal number of each of the five symbols. The probability of guessing correctly 4018 or more out of 16,000 cards is 1 in 7.7  times 10 to the 56th power. This is a probability of less than 1 in 100,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000. We would never expect chance to produce such a result, even if you spent half of every person's life doing telepathy tests on them. 

The reported test result is one of the best test results ever achieved in a test of telepathy. I can recall only one better result: the result of the Riess test, reported here. 

Rhine's results were very convincing evidence for ESP.  I think he may have got results far more convincing if he had varied his procedural methods more. In the book we read of years using the same type of Zener cards, containing only abstract symbols with no particular emotional associations. There are endless other ways in which ESP could be tested. A method involving a more emotional component might be much more successful. For example, tests could be done with cards having emotionally powerful photos. A test of telepathy with such cards might produce much higher scores than tests using cards with abstract symbols. 

The 1925 New York Times article below (which you can read here) has the large headline "Balfour Says Telepathy Is Proved Fact." The Balfour referred to (Earl Balfour) was once a Prime Minister.