The Myth of the Million Dollar Psychic Challenge

For ten years, the modern skeptical movement has wielded a cudgel against claims of the paranormal: the James Randi Million Dollar Challenge. In many debates over the possibility of psi abilities, the Challenge provides a final word for one side... "has so-and-so applied for the Challenge?"

The financial reward offered by the James Randi Educational Foundation is seen by many skeptics as providing an irresistible motivation for anybody with paranormal ability - after all, if someone could genuinely exhibit such powers, surely they would step forward to take the million?

However, after ten years, the James Randi Educational Foundation (JREF) says nobody has even got past their preliminary testing. Furthermore, none of the 'big fish' - medium John Edward, spoon-bender Uri Geller, psychic Sylvia Browne - have applied. And now, perhaps as a result of that fact, James Randi has announced that the Challenge will come to an end in two years, on March 6th, 2010.

But does the challenge really make a statement about the existence of the paranormal and/or psi abilities? According to paranormal investigator Loyd Auerbach (who, like Randi, is a member of the magic fraternity):

The suggestion that ending the Challenge after 10 years supports any statement that psi does not exist or someone would have won the challenge, is absurd on many levels."

The procedures for the Challenge included several hurdles in favor of, and multiple "outs" for Randi and the JREF that any discerning individual capable of any kind of extraordinary human performance would think twice about (and here I'm not just referring to psychics and the like).

What are these hurdles that Auerbach refers to?

Chances, of Anything...

First, and perhaps the most important, is the effect size required to win the challenge. While the JREF says that "all tests are designed with the participation and approval of the applicant", this does not mean that the tests are fair scientific tests.

The JREF need to protect a very large amount of money from possible "long-range shots", and as such they ask for extremely significant results before paying out - much higher than are generally accepted in scientific research (and if you don’t agree to terms, your application is rejected). In the case of parapsychological research, however, where effect size is often small (though apparently robust), this means most researchers would have to go to extraordinary lengths to win the million dollars. As one psi researcher pointed out to me:

In the ganzfeld telepathy test the meta-analytic hit rate with unselected subjects is 32% where chance expectation is 25%. If that 32% hit rate is the "real" telepathy effect, then for us to have a 99% chance of getting a significant effect at p <>

Furthermore, applicants must first pass a 'preliminary test', before they are allowed to progress to the actual 'formal' test which pays the million dollars. So an applicant must first show positive results in a preliminary test (yielding results against chance of at least 1000 to 1, apparently), then once through to the next stage they would then have to show positive results against much higher odds to claim the prize (by all reports, at odds of around 1 million to 1). Failure in either test means no cash prize, and a fail beside their name. It many respects it would be like telling a professional golfer to shoot 63 around Augusta National, then come back and shoot 59, to prove that he can play golf. In the words of Chris Carter, author of Parapsychology and the Skeptics:

If Randi were genuinely interested in testing unusual claims, then he would also not insist upon odds of at least one million to one against chance for the results. Anyone familiar with scientific studies will be aware that experimental results against chance of say, 800,000 to one would be considered extraordinary; but results this high would be, according to Randi, a “failure.”

Dr Michael Sudduth of San Francisco State University also pointed out to me a wonderful irony in one of the rules. Challenge rule #3 states: "We have no interest in theories nor explanations of how the claimed powers might work." As Sudduth puts it: “Curiously, Randi's challenge itself is saddled with assumptions of this very kind. The challenge makes little sense unless we assume that psi is the sort of thing that, if genuine, can be produced on demand, or at least is likely to manifest itself in some perspicuous manner under the conditions specified by the challenge.”