A Collection of Strange Beliefs, Amusing Deceptions, and Dangerous Delusions

From Abracadabra to Zombies | View All

Gary Schwartz's Subjective Evaluation of Mediums

Veritas or Wishful Thinking?


Robert Todd Carroll


Gary Schwartz validates mediums with about as much care as Pope John Paul II validated saints. In his Afterlife Experiments, Schwartz anoints John Edward and Laurie Campbell on the basis of a few readings that dazzle the former Harvard professor. Never mind that Ray Hyman and Richard Wiseman don't see anything in the readings that can't be explained by cold or hot reading. Schwartz dismisses his critics as super-skeptics and is convinced that these mediums are the real thing. He is the self-proclaimed expert on the subject, so it is not surprising that some members of the media go to Schwartz as their point-man when they have a question about something important like a television series about a medium.

NBC's "Medium," starring Patricia Arquette as a psychic who helps police solve crimes - among other things - premiered January 3, 2005. Less than one month later, NBC announced that "Medium" will be back for a second season. The show averaged over 15 million viewers per episode in its first few weeks.

Arquette's character is referred to as a "research medium" and is said to be based on the work of Allison Dubois, who works out of Phoenix, Arizona. On her website, Dubois says:

I call things like I see them and I am not afraid to push the boundaries of my abilities under university research conditions. I pride myself on accuracy, consistency and easing the pain of those who have lost loved ones.

In other words, she is cut from the same cloth as John Edward, George Anderson, Laurie Campbell , James Van Praagh, and a host of other "grief counselors" who offer their services to the grieving and the bewildered. Like Edward, Anderson, and Campbell, Dubois has been tested by Gary Schwartz and declared by him to be a bona fide psychic.

"There is no question this is not a fraud - some people really can do this, and Allison is one of them," declared Schwartz. "Many people claim to do this, and there are clearly frauds out there. Allison was repeatedly tested and passed every test. As a scientist, I approach all this as an agnostic - I don't believe it; I don't disbelieve it. After testing her under conditions that ruled out the possibility of fraud, I came to the conclusion she's the real deal."

Whether she's the real deal, Allison is one busy psychic detective and research medium. According to her website, she's booked solid for the next three years. She claims she has a backlog of 200 murder cases to solve and 3,000 people on her waiting list.

According to Carla McClain of the Arizona Daily Star:

Dubois first called Schwartz four years ago, after seeing him on a "Dateline" NBC segment with John Edward on paranormal powers. She wanted to see how good her "gift" really was.

Schwartz first put Dubois through a direct, informal reading on himself. A beloved mentor of his had just died, but he told her nothing about that woman.

Among other things, Dubois told Schwartz "the deceased was telling me that I must share the following - I don't walk alone," a seemingly innocuous piece of information, but critical to him.

"My friend had been confined to a wheelchair in her last years - there is no way Allison could have known that," he said. ("Varied readings on Arizona psychic," January 17, 2005.)

One reason we should distrust Schwartz's evaluation of anyone's psychic ability is his persistent revelation that he has little or no understanding of how subjective validation works. His naiveté is exemplified by the way he interpreted Dubois's statement "I don't walk alone" to mean "confined to a wheelchair." The reason Dubois and others like her seem to have psychic powers is because their statements, phrases, words, even utterances of nothing more than initials, are given meaning by others. In a classic experiment that has been repeated many times in many different contexts, Bertram Forer gave a personality test to his students, ignored their answers, and gave each student an "evaluation" he had taken from a newsstand astrology column. He asked his students to rate the evaluation from 0 to 5, with "5" meaning the recipient felt the evaluation was an "excellent" assessment and "4" meaning the assessment was "good." The class average evaluation was 4.26. That was in 1948. The test has been repeated many times with psychology students and the average is still around 4.2. We might translate this to mean that it is quite common for people to be given strings of statements that are not based on any knowledge of the person and yet they commonly rate the statements as about 80% accurate. Similar experiments have been done with phony biorhythm charts, graphology readings, and astrological charts. (See Flim-Flam! by James Randi and the Nova production Secrets of the Psychics.)

In his book The Afterlife Experiments, Schwartz describes the Herculean efforts he went through to understand cold reading. He knows that many tricksters use cold reading to defraud others by pretending to know secrets about them and he wanted to make sure than none of the psychics he tested was a fraud. However, in his zeal to understand cold reading, he overlooked the most essential element in the process: the way subjective validation functions in the evaluation of any reading. Forer and others have been able to get a high rating of accuracy for phony readings without cold reading. Schwartz has been so diligent to make sure his subjects weren't cheating that he overlooked the obvious: the high ratings given psychic readings were probably due to subjective validation. Ruling out cold reading and cheating, while important, are not sufficient to ensure an accurate assessment of rater bias.

McClain continues:

After that, the formal, scientific experiments began under controlled conditions - some of them completely "blinded," so Dubois could not see or talk to the person she was reading, or vice versa. They were not even told each other's full names.

In that situation, it is virtually impossible to use tricks fake mediums use - throwing out streams of general information and following up on those that get visible reactions - methods known as "cold reading."

In some cases, fake mediums also have been known to tap phones and hire detectives to get vital information on people they are going to read. That is impossible if the medium does not know who the person is.

In one of these experiments, Dubois was asked to contact a deceased person close to a woman in England she had never met. She was told only the woman's first name and that she wanted to hear from her deceased husband. During the actual reading, Dubois was at the UA [University of Arizona] lab, and the woman was in England.

A transcript of the information Dubois got during the reading - supposedly from the dead husband - was sent to his wife in England, who scored it as 73 percent accurate.

"That's extraordinarily high accuracy, and Allison always scored in the near-80 percent range," Schwartz said. "That clearly puts her among the best of the best."

No psychic medium is 100 percent accurate, he said.

Schwartz asserts that a 73% accuracy rating by a "sitter" - the one for whom a reading is done - is "extraordinarily high accuracy." However, without a control, Schwartz has no way of knowing whether the English lady's rating of Dubois's reading is extraordinarily high or just average. Why didn't Schwartz send the English lady several readings without telling her which was the "real" reading? Why didn't he send the "real" reading to several different English ladies with similar interests and backgrounds and check their ratings against the sitter's? Had the English lady rated Dubois's reading as 73% accurate and several fake readings as very low in accuracy, Schwartz might have a case for crowing about her "extraordinarily high accuracy." Likewise, had several other English ladies who wanted to make contact with their dead husbands all given the thumbs down to the accuracy of Dubois's reading, then Schwartz might have reason for thinking that the reading was of "extraordinarily high accuracy." When he rates Dubois as among the best of the best, Schwartz is simply comparing her to other mediums he has tested and evaluated in the same way: without adequate controls and without an understanding of how subjective validation works.

McClain goes on:

Some of Dubois' best results were in one of her more famous UA experiments, when she read for celebrity physician-author-lecturer Dr. Deepak Chopra, just after the death of his father, a famous cardiologist in India.

During the reading, Chopra was in California, Dubois was in Arizona, and they were connected by phone. Dubois was not told who Chopra was. He could hear her, but he was not allowed to speak to her.

According to a summary of the reading done by Schwartz, she told him the deceased person was a man of great stature, extremely handsome, had beautiful women around him, was known to politicians and other well-known people, and was cremated - all accurate, according to Chopra's evaluation.

But she also told him his father was connected to the U.S. oil and steel industry, and there was a small dark terrier dog in his life - not true, Chopra said. Her accuracy score - 77 percent, according to Chopra's scoring, Schwartz said.

Schwartz loves to throw in the obvious misses as proof that no psychic is 100% accurate. Rather than take the obvious errors as signs that the psychic is not psychic, Schwartz takes it as evidence in support of the genuineness of the medium. Ian Rowland, in his book The Full Facts Book of Cold Reading, calls this the win-win gambit.

 Again, Schwartz has nothing to compare Chopra's rating to. For all Schwartz knows, there might be 10,000 phony readings that Chopra would rate at 75% or higher. He was only given one to rate. Schwartz seems to think it doesn't matter that Chopra is a believer and a highly motivated sitter. Chopra would want the medium to succeed, not only to verify his own personal beliefs but because he wants to make contact with his dead father.

After studying Schwartz's Afterlife Experiments, I have concluded that he would give up his belief that some mediums are genuine only if there were nobody on the planet who would rate a list of names, initials, dates, places, things, and the like, as being highly accurate for them personally. This is not going to happen. Thus, I deem it highly unlikely that Schwartz will ever give up validating mediums. And as long as Schwartz continues to have a professorship at a university and the support of an academic institution, it is highly likely that mediums like Allison Dubois will seek him out for validation.


Here is an example of how Schwartz works. In one session during the so-called "afterlife experiments," John Edward recited the following:

The first thing being shown to me is a male figure that I would say as being above, that would be to me some type of father image...Showing me the month of May. . . .They're telling me to talk about the Big H-um, the H connection. To me this is an H with an N sound. So what they are talking about is Henna, Henry, but there's an HN connection ... Very strong symbolism of teaching and books… The books come up where there may be something published. (Afterlife Experiments xix)

This list of items was validated by one of Schwartz’s sitters. To validate is not the same as to confirm that messages have truly come from a spirit. It means that the sitter in a reading can find meaning here, can connect the dots, so to speak. The ability to validate depends on several factors. The sitter must be willing to validate. The stronger the desire to make contact, the harder the sitter will work to find meaning and connections in the medium’s items. There may also be another mechanism at work here: the desire to please the medium. This may be due partly to the consideration that by pleasing the medium, the odds increase that the medium will make contact. But it may also have to do with a strange phenomenon that occurs in settings where a person gives up control of the situation to another, as in hypnosis or when being asked to assist a magician do a trick. There is sometimes a kind of loss of self in those situations, and combined with a desire to please, a kind of submission to the will of another, up to a point. If such a mechanism is at work in psychic readings, the sitter may acquiesce to the suggestions or items thrown out by the medium, not because they are true or truly significant, but out of a desire to please.

In one reading, for example, Edward got the sitter to validate the claim that her husband was dead. Schwartz admitted that the sitter's husband was alive at the time of the reading. However, since the husband was killed a few weeks later, Schwartz indicates he thinks it is more likely that Edward has precognitive powers than that the sitter made a false claim to please the medium.

Because the motivation of the sitter is so high that it might lead her to validate false or ambiguous statements, experiments should be designed in such a way that the experimenter always checks factual claims made by sitters. The word of the sitter should not be sufficient. Nor should the sitter's validation of the word of others be sufficient. Schwartz sometimes checks factual claims and sometimes he doesn’t. Sometimes he takes the word of the sitter. Sometimes he accepts the sitter's validation of the words of others. (See his "White Crow" experiment for an example of both.)

Even though the concern with factual accuracy is important in verifying the success of the medium, one should not lose sight of the importance of the studies that have been done on how the human mind works when it comes to making sense out of and giving significance to disparate data presented to it. The overall effect of subjective validation would show up in the way the sitters rate the accuracy of the mediums’ claims. There might well be a rater bias towards any reading done face to face or where the sitter can hear the medium do the reading (even if they are not visible to one another). To eliminate this kind of bias, Schwartz, in one experiment, had the readings done from remote locations, so that the sitter wasn’t present and didn’t hear what the medium was saying during the readings.

Schwartz is aware that sitters might be biased in their ratings and might give higher ratings of accuracy to items than they deserve, so he asked the sitters to rate down rather than up. Schwartz also has a procedure in some of his experiments and presents a challenge several times to the reader (or any skeptic) to see if they can connect the dots for a given reading. If he or the other experimenters can’t, or if subjects he calls “controls” in one experiment can’t, he takes that as strong evidence that the data is sitter-specific. He often works up conditional probability calculations, where he shows that the odds of several contingent conditions being related are millions or billions or trillions to one and he takes this as strong evidence that the data is sitter-specific.

The problem with this method is that the rest of us don’t have the interest in contacting the dead that the sitter does, but if even if we do, we don’t have any reason to believe that the items for one sitter would apply to us. Some sitters not only have a stronger will to succeed with the reading, but they are better suited for readings because of factors such as how many dead friends and relatives they have, which would be related to such things as age and size of extended family, whether one is a sociable type or not, being a gay male at a time when many gay males were dying of AIDS, and so on.

Schwartz seemed particularly impressed by the above quote from a John Edward reading because he couldn’t relate much of it to his own life. Ray Hyman has done a pretty good job of connecting the dots of this passage in his life.

When I put myself in the shoes of a possible sitter and try to fit the reading to my situation, I can find a good fit to my father, who was physically large, whose last name was Hyman, and for whom, like any human on this planet, experienced one or more notable events in the month of May. Other things in the reading also can easily be fitted to my father. Neither the original sitter nor anyone else would fit this cluster of facts! Schwartz makes much of the fact that the cluster of facts that a sitter extracts from a reading tend to be unique for that sitter. He even calculates the conditional probabilities of such a cluster occurring just by chance. Naturally, these conditional probabilities are extremely low--often with odds of over a trillion-to-one against chance. (Hyman 2003)

I can also connect these dots pretty well. I was born in the month of May and my father, who was stocky and often compared to a gentle bear, died about 30 years ago of the Big H: a heart attack. Henry was one of my high school buddies. Maybe Henry’s dead and is with my dad. Or maybe my dad is trying to let me know it’s him by bringing up the name of good friend from my youth. Of course, Hyman and I are both teachers and have been surrounded by books for most of our adult lives. Ray and I have published books and this reference could be our fathers’ way of letting us know that they know what we’ve been up to these past many years.

Schwartz relates the teaching and books to “literature and education.” I’m sure it could be related to several other things as well, such as libraries, bookstores, any kind of school, visiting anyone with a library (such as a lawyer or doctor), and so on. Schwartz asserts that “the probability of getting just this pattern of hits is on the order of a million to one.” How he knows this is not revealed. But my guess is that the odds are more favorable than he thinks. If there is a dead person, the odds are that the person is male, and if male a good guess would be older rather than younger, and if older, a father rather than a son. To throw out the notion that one senses a father figure does not seem to defy all odds, especially when the sitter recognizes her husband, not her father, as a father figure. I wouldn’t call the father image very specific. Nor would I call it correct. I’d call it common. (This point is more obvious in a reading with feedback from the sitter. If you get no positive response on the father figure, the medium can insist on the image and give the impression that the sitter isn’t trying hard enough. Or, the medium can change directions and hope for a better response, with initials for example, or two names, one male and one female, and see what response that generates.)

Schwartz dismisses the possibility that Edward was simply guessing, but I think the main issue isn’t whether Edward is guessing. He could be passing on things that are going through his mind, but that wouldn’t mean they necessarily came from outside his mind, either telepathically from the sitter or from the spirit world. He and many other “good” mediums might be people with very active imaginations, who generate words and images that originate in their own brains. What they do may be similar to dreaming out loud. But the focus shouldn’t be just on the medium and whether he or she is guessing or cheating; the focus should be on the sitter. Again, the focus shouldn’t be on whether the sitter is cheating, but on the dynamics of subjective validation. Schwartz says almost nothing about this well known psychological factor, except to dismiss the charge of cold reading on the part of the mediums on the ground that they are not using the standard tricks of magicians and mentalists. Maybe not. However, Schwartz seems to miss the point that although cold reading may sometimes be tricky, it doesn’t always involve trickery.

See also my review of the Afterlife Experiments;What if Gary Schwartz is Right?; and A Novel Way to Make an Ass of Yourself - Gary Schwartz Rides Again.

Last updated 12/09/10

More Essays

This page was designed by Cristian Popa.