What a Fair Test Looks Like
Using Charles Tart's Miss Z experiment, and what it teaches
I want to show you what a fair test looks like, and I’m going to do it with a case from my own field. It’s also a lesson in why even a good-looking result isn’t the end of the story. Especially a good-looking result.

Miss Z
In 1968, the psychologist Charles Tart ran a study in his sleep laboratory. His subject, known as Miss Z, said she often left her body while asleep. He wired her to an EEG and placed a five-digit number on a shelf above her bed, out of reach. On the fourth night, she woke and called out the number, 25132. It was correct. Tart worked out the odds of guessing a five-digit number at 100,000 to 1.
If you’re on my side of this debate, that sounds like the end of the argument. If you’re on the other side, you’re already reaching for your objections. Both reactions are too quick.
Why it settled nothing
Tart himself listed the holes, which is the most admirable thing about the whole episode. There was a clock on the wall above the shelf, and the number could have been reflected in its glass. He was watching from the next room through a window, and he admitted he had dozed during the night. There was no video. Nobody could rule out her slipping a mirror up to look. He changed the room afterwards so no reflection was possible, but Miss Z moved away and it was never repeated. Tart wrote that the result could not be taken as conclusive.
I respect him enormously for that. Most people defend their best result like a dog with a bone. Tart pointed at the weak spots himself, in print, about his own work. That is what a researcher does and what a salesman doesn’t.
The checklist for a test
From cases like that, here is what I’d want in any test of an unusual claim, whether it’s a dream, a trance or a hospital room.
The target is chosen at random, by someone who isn’t present, after the claim has been made.
There is no ordinary way to see or hear it. That means checking reflections, sounds, and who is in the room.
Everyone present, and everyone checking the result, is blind to the answer.
Everything is recorded at the time, by video or independent witnesses, not remembered afterwards.
The plan is written down and filed before the study starts, so nobody can change what counts as a hit.
The result is repeated by someone else.
Failures are published as well as successes.
Notice that this list is not designed to keep strange claims out. It’s designed to let a strange claim in, and to make sure that if it gets in, nobody can later say it sneaked past security.
Why the last two matter
In 2011 a respected psychologist, Daryl Bem, published nine experiments in a leading journal that seemed to show people could sense the future. It looked rigorous. It passed peer review. When other teams tried to repeat the work, including one led by Stuart Ritchie, they didn’t get the result.
Then, in 2015, a large project tried to repeat 100 published psychology studies. About 97% of the originals had reported a statistically significant result. Only about 36% of the repeats did.
Sit with that for a moment. This isn’t a story about fringe science. It’s a story about the mainstream, and about how often a striking, careful, published result evaporates when someone else tries it. A single result, however good, is a lead and not a conclusion. That is true of Miss Z, and it’s true of results I’d dearly like to be true.
Two corrections of my own
I can’t ask this standard of others and walk around it myself, so here are two things I need to put right.
In some earlier posts I wrote that information gathered in trance had, in a number of documented cases, been verified by means the traveller couldn’t have had. I attributed that to Michael Harner’s research. Having gone back to check, I can’t support that wording. What Harner documented well is that people from different cultures describe consistent structures in these states. That is an interesting finding in itself. But I haven’t seen verified cases that meet the checklist above, and I shouldn’t have implied I had.
I also wrote that accounts from the van Lommel study were corroborated by hospital staff. I now can’t find that in the paper itself, and I shouldn’t have written it as settled.
I’m correcting both here, in public, because I’d be a hypocrite to do anything else.
Kathy Postelle Rixon is Chair of The Philosophical Society: Oxford, an independent researcher in consciousness studies, and a shamanic practitioner of more than forty years, working in Q’ero, Norse and Celtic traditions. You can contact her at kathy@magicinharmony.com or visit her website at www.magicinharmony.com.






Comments