parapsychology and the challenge

[davidsmith73]

Can anyone think of a particular parapsychology experiment that's been done already that, if performed successfully in the company of JREF, would qualify for the million?

 

[drkitten]

Can anyone think of a particular parapsychology experiment that's been done already that, if performed successfully in the company of JREF, would qualify for the million?

I'm not quite sure what you're asking. Part of why the JREF exists (and why Randi got into this business) is because he was investigating the claims that were made by parapsychology researchers --- claims that, if reproducible, would have shattered the scientific understanding of the day (and of today, for that matter).

Uri Geller would be a good example. If he had actually been able to perform as he claimed, he would certainly have qualified for the million long ago. A number of scientists performed various experiments on him, that if they had been performed "successfully" in the company of JREF would have qualified. But, of course, the reason that they couldn't be performed successfully is because the results had been obtained through trickery and deception which the original researcher had not noticed.

My working assumption, then, would be that any "successful" experiment that has been done outside of the supervision of the JREF or similar group is successful only because the trickery hasn't been caught. Of course, we don't know what sort of trickery was used, because it wasn't caught. But I'm willing to operate on the basis that it was trickery until someone can come up with a more credible hypothesis.

 

[Carn]

Re: Re: parapsychology and the challenge

My working assumption, then, would be that any "successful" experiment that has been done outside of the supervision of the JREF or similar group is successful only because the trickery hasn't been caught. Of course, we don't know what sort of trickery was used, because it wasn't caught. But I'm willing to operate on the basis that it was trickery until someone can come up with a more credible hypothesis.

I think you are a bit to radical, since the general scientific standard, which of course was devised for sensible things with plausible theoretical explanations, allows them to go on with "successful" experiments and all they need is a little delusion:
The success criteria with random data.

Generally speaking a experiment is a success if its result has less than 5% to be obtained by chance.

Now if you have 20 parapsychologists doing 1 experiment per year, they will get 1 experiment, which showed something.
They exchange info and during next year they try to replicate the successes of the rest.
The experiment will be copied by the 19 others and therefore likely there will be a further one this year. Gives news of a replicable experiment. Many retries will fail, but as they have no theory behind it, they can select any minute difference(such are always there) as possible suspect that the many were unable to repeat the experiment.
This gives them much to do the following year. Again everyone does the experiment again, again getting one success. Now even they will get suspicious, as the guys who had a success the first year and the second year are unable to reproduce their sucess. In the next year they'll do again and fianally realize the sucess is purely random.

So they spend 3 years realizing that one of them just got lucky.
No take the bias in that unsucessful experiments will not be published as often and you can imagine, that they chase luck for decades that way.

And JREF challenge is always out of reach, because none of their experiments beats the 1 in 1000 chance, much less the 1 in 1000000.

Carn

 

[MRC_Hans]

Quite right, Carn. And this is central, and something paranormal proponents need to mull over (they might even complain about it): The JREF challenge requires a much greater certainty than a normal statistical analysis.

Hans

 

[Rolfe]

Re: Re: Re: parapsychology and the challenge

Generally speaking a experiment is a success if its result has less than 5% to be obtained by chance.

Generally yes, for biological experiments. However, that doesn't mean anyone is declaring there's definitely something there, it just means it looks like there's a fair probability that there's something there, and it's worth a closer look.

Then the same researchers and/or others try to replicate the result and track it down to its lair, and in the end a consensus emerges as to what is or isn't going on.

However, in contrast,

The field of physics is full of examples of extraordinary effects with p values much lower than 0.05 that failed to be replicated and never became part of established knowledge. The accepted criterion for publication in most physics journals is p=10<SUP> -4</SUP>. That is, as study is published only when one in ten thousand experiments or a smaller fraction would produce the observation as a statistical artefact. Such a strict threshold would be useful in evaluating extraordinary claims, such as may be made for CAVM.

Source, Complementary and Alternative Veterinary Medicine Considered, by David Ramey and Bernard Rollin.

Essentially, when dealing with sciences as "hard" as physics, you have to be very sure indeed before you concede that a new effect has been observed. A similar standard of proof should be required for any claim which appears to run contrary to "science as we know it".

Studies with weak statistical significance simply point to potentially fruitful areas for further research. If there's anything there, it will eventually become clear. Unfortunately, with psi (as with most alternative medicine) nothing ever bcomes clear except that they never seem to give up trying.

Rolfe.

 

[Tricky]

Can anyone think of a particular parapsychology experiment that's been done already that, if performed successfully in the company of JREF, would qualify for the million?

Absolutely! Randi has repeatedly offered Dr. Gary Schwartz the million just to turn over his raw data (a deal that Schwartz agreed to, then reneged). I don't have the reference, but I believe that Randi has also offered him the prize if he can repeat his results with proper controls in place.

 

[davidsmith73]

Re: Re: Re: parapsychology and the challenge

I think you are a bit to radical, since the general scientific standard, which of course was devised for sensible things with plausible theoretical explanations, allows them to go on with "successful" experiments and all they need is a little delusion:
The success criteria with random data.

Generally speaking a experiment is a success if its result has less than 5% to be obtained by chance.

Now if you have 20 parapsychologists doing 1 experiment per year, they will get 1 experiment, which showed something.
They exchange info and during next year they try to replicate the successes of the rest.
The experiment will be copied by the 19 others and therefore likely there will be a further one this year. Gives news of a replicable experiment. Many retries will fail, but as they have no theory behind it, they can select any minute difference(such are always there) as possible suspect that the many were unable to repeat the experiment.
This gives them much to do the following year. Again everyone does the experiment again, again getting one success. Now even they will get suspicious, as the guys who had a success the first year and the second year are unable to reproduce their sucess. In the next year they'll do again and fianally realize the sucess is purely random.

So they spend 3 years realizing that one of them just got lucky.
No take the bias in that unsucessful experiments will not be published as often and you can imagine, that they chase luck for decades that way.

And JREF challenge is always out of reach, because none of their experiments beats the 1 in 1000 chance, much less the 1 in 1000000.

Carn

I'm sure nobody minds speculation as long as you acknowledge it as so. Having the evidence that 19 failed psi experiments are performed for every successful 1 is a different matter.#

So is your answer to my question - none ?

 

[davidsmith73]

Quite right, Carn. And this is central, and something paranormal proponents need to mull over (they might even complain about it): The JREF challenge requires a much greater certainty than a normal statistical analysis.

Hans

I agree that psi experiments should require a greater certainty than more conventional experiments that are based on solid theory, but I don't see any indication as to what objective degree of certainty the JREF adheres to, in the rules that is. It seems to depend upon an agreement between the JREF and the claimant.

That aside, I wonder how many negative psi experiments would have to have been performed to negate all the positive ones so far acheived? Is it really plausable that the time and resources have been available to so few researchers?

 

[davidsmith73]

Re: Re: parapsychology and the challenge

Absolutely! Randi has repeatedly offered Dr. Gary Schwartz the million just to turn over his raw data (a deal that Schwartz agreed to, then reneged).

I guess that Randi knows that Schwartz's methods are invalid and his million is safe. I don't think Schwartz's experiments, as they stand, qualify for a psi experiment that would win the prize.

I don't have the reference, but I believe that Randi has also offered him the prize if he can repeat his results with proper controls in place.

That's interesting. I wonder what experimental design and what level of results Randi requires.

 

[MRC_Hans]

I agree that psi experiments should require a greater certainty than more conventional experiments that are based on solid theory, but I don't see any indication as to what objective degree of certainty the JREF adheres to, in the rules that is. It seems to depend upon an agreement between the JREF and the claimant.

It has been my impression that they are aiming at the one in a million certainty. Quite appropriate, as a million bucks is at stake.

That aside, I wonder how many negative psi experiments would have to have been performed to negate all the positive ones so far acheived?

All the positive ones?? Herein lies trouble; because many of the positive trials have been severely challenged, protocol-wise. So once we look at only those with impeccable protocols, how many remain, if any?

Is it really plausable that the time and resources have been available to so few researchers?

I don't understand that question.

Hans

 

[Zep]

davidsmith73,

What would you suggest is the situation if the first analysis of 25 years of some psi research data suggested that some particular psi phenomenon existed. But when that self same data was analyzed more closely, the phenomenon became less and less obvious. And when strong analysis was done to remove all biases and influences, the data yielded nothing at all.

What do you think would cause a situation like that?

 

[Darat]

Re: Re: Re: parapsychology and the challenge

...snip...

That's interesting. I wonder what experimental design and what level of results Randi requires.

He has certainly put the money on the line for experiments for homeopathy in the past so I don’t think there is any reason to assume that the same wouldn’t be the case for a specific parapsychology experiment.

Would you consider dowsing to be a "power" that is covered by parapsychology like say telepathy is? If so then again there are plenty of examples of Randi putting the million dollars on the line.

I’d suggest the quickest way to get an answer would be to drop Randi or Kramer an email – they are after all the only ones who can give you a definitive answer.

(Edited for a to.)

 

[davidsmith73]

davidsmith73,

What would you suggest is the situation if the first analysis of 25 years of some psi research data suggested that some particular psi phenomenon existed. But when that self same data was analyzed more closely, the phenomenon became less and less obvious. And when strong analysis was done to remove all biases and influences, the data yielded nothing at all.

What do you think would cause a situation like that?

The PEAR paper is certainly a tough read. I am trying to figure out, what is it specifically about the later analytical methods that removes bias and influences? From what I can understand, the difference between the earlier and later methods is that the later ones used more options for responses by the RVers.

eg, Q1) is the scene indoors?

answers: 1- absent
2- unsure
3- present
4- dominant

 

[davidsmith73]

Re: Re: Re: Re: parapsychology and the challenge

Would you consider dowsing to be a "power" that is covered by parapsychology like say telepathy is? If so then again there are plenty of examples of Randi putting the million dollars on the line.

I wouldn't consider dowsing in the same category as other more likely phenomena like telepathy simply because dowsing isn't something that's spontaneously experienced by normal people in a variety of contexts. Instead it's something that people have "tried" to do for reasons I can't really understand. I don't think its a natural phenomena, in fact I think it's just some foolish types walking about with sticks and dangling crystals.

 

[Dr Adequate]

Re: Re: Re: Re: parapsychology and the challenge

I'm sure nobody minds speculation as long as you acknowledge it as so. Having the evidence that 19 failed psi experiments are performed for every successful 1 is a different matter.#

That is not what was claimed. You've missed the point entirely. Let's try again.

The point is not that there are nineteen failures for every success, but that if you take 5% as success, then there will be at least one success for every nineteen failures. If you take 5% significance as a criterion of success then even if there's no psi involved, you'll succeed 5% of the time. Right? 'Cos that's what it means.

This is why we normally ask for a repeatable experiment. One success at the 5% level might be fluke.

Randi, on the other hand only gets one experiment, so it must be designed large enough that there is a very very low probability of the phenomenon occurring by chance.

Got it?

 

[Psiload]

Re: Re: Re: Re: Re: parapsychology and the challenge

I wouldn't consider dowsing in the same category as other more likely phenomena like telepathy simply because dowsing isn't something that's spontaneously experienced by normal people in a variety of contexts. Instead it's something that people have "tried" to do for reasons I can't really understand. I don't think its a natural phenomena, in fact I think it's just some foolish types walking about with sticks and dangling crystals.

And trying to influence the output of a random number generator is something that IS spontaneously experienced by normal people?

Seems like some foolish types sitting around trying to make a computer do their psychic bidding....

doesn't seem any more rational to me than what the crystal danglers are up to.

 

[davidsmith73]

Re: Re: Re: Re: Re: parapsychology and the challenge

That is not what was claimed. You've missed the point entirely. Let's try again.

The point is not that there are nineteen failures for every success, but that if you take 5% as success, then there will be at least one success for every nineteen failures. If you take 5% significance as a criterion of success then even if there's no psi involved, you'll succeed 5% of the time. Right? 'Cos that's what it means.

This is why we normally ask for a repeatable experiment. One success at the 5% level might be fluke.

Randi, on the other hand only gets one experiment, so it must be designed large enough that there is a very very low probability of the phenomenon occurring by chance.

Got it?

Oh ok, I was under the impression that Carn was implying that the success rate already achieved for psi research is a fluke. Hence his words:

This gives them much to do the following year. Again everyone does the experiment again, again getting one success. Now even they will get suspicious, as the guys who had a success the first year and the second year are unable to reproduce their sucess. In the next year they'll do again and fianally realize the sucess is purely random.

So they spend 3 years realizing that one of them just got lucky.
No take the bias in that unsucessful experiments will not be published as often and you can imagine, that they chase luck for decades that way.

And JREF challenge is always out of reach, because none of their experiments beats the 1 in 1000 chance, much less the 1 in 1000000.

Significance of the results aside, is there any psi experiment that would qualify on methodological grounds?

 

[davidsmith73]

Re: Re: Re: Re: Re: Re: parapsychology and the challenge

And trying to influence the output of a random number generator is something that IS spontaneously experienced by normal people?

Seems like some foolish types sitting around trying to make a computer do their psychic bidding....

doesn't seem any more rational to me than what the crystal danglers are up to.

The sense of free will as something that can effect the physical world is something experienced by everyone, for example volitional movement. So it would be an investigation into whether this feeling is really an illusion or whether it can really effect physical processes beyond the workings of the brain. Conventional science usually performs experiments on free will under the assumption that the feeling is a generated, non-causal phenomena based on normal physical processes occuring in the brain. PK tests are testing the alternative hypothesis. So yes, I would say it's a pretty obvious and rational thing to test.

 

[Tricky]

Re: Re: Re: parapsychology and the challenge

I guess that Randi knows that Schwartz's methods are invalid and his million is safe. I don't think Schwartz's experiments, as they stand, qualify for a psi experiment that would win the prize.

They are some of the most well known and frequently cited experiments available today. They have the material, if not philosophical support of a large, state-run university where Schwarts is a full professor. They have tested some of the most well-known "psychics" in the world.

I think you would be hard pressed to find another research project with such credentials, and while I realize that such visibility does not make his methods valid, you would think that the visibility would at least attract the best psi researchers in the business. After all, it's hard to get a grant to do psi research. The University of Arizona ought to be a Mecca for such people.

Which raises another question. Why is it so hard for psi researchers to agree on a methodology? In every other line of research, one superior technique will eventually become predominant because it consistantly works better than others. But this is not the case with psi. None of them seem to work better consistantly. It is not as if they haven't tried lots of different methods. Years of constant tweaking of the paramaters have yielded no better results, and in fact, have reduced "psi effects" practically to noise level, so much so that it requires torturing the data with things like meta-analysis to reveal any effect at all. If this trend continues, such effects should disappear entirely.

 

[davidsmith73]

Re: Re: Re: Re: parapsychology and the challenge

They are some of the most well known and frequently cited experiments available today. They have the material, if not philosophical support of a large, state-run university where Schwarts is a full professor. They have tested some of the most well-known "psychics" in the world.

Who have they been cited by?

I think you would be hard pressed to find another research project with such credentials, and while I realize that such visibility does not make his methods valid, you would think that the visibility would at least attract the best psi researchers in the business. After all, it's hard to get a grant to do psi research. The University of Arizona ought to be a Mecca for such people.

Why would serious psi researchers attempt to carry out research as sloppy as Schwartz's? Methods are everything in psi research, we both know that.

Which raises another question. Why is it so hard for psi researchers to agree on a methodology? In every other line of research, one superior technique will eventually become predominant because it consistantly works better than others. But this is not the case with psi. None of them work better consistantly. It is not as if they haven't tried lots of different methods. Years of constant tweaking of the paramaters have yielded no better results, and in fact, have reduced "psi effect" practically to noise level, so much so that it requires torturing the data with things like meta-analysise to reveal any effect at all. If this trend continues, such effects should disappear entirely.

Which experiments are you talking about?