Ganzfeld million dollar challange?

[Dancing David]

"

"In the judging procedure, the receiver is taken out of the ganzfeld state and given a set of possible targets, from which they must decide which one most resembled the images they witnessed. Most commonly there are three decoys along with a copy of the target itself, giving an expected overall hit rate of 25% over several dozens of trials."

And this even more prone to bias than asking the receiver to generate a word string and trying to score a hit. How are the picture of the decoy selected so that they are obviously different?
They are NOT decoys they are 'nulls'.

Why not use Rhine cards?

This is one of three ways to rate the success, this quote has mixed up the success rate between methodologies.

 

[Dancing David]

"Judging procedure"? You're kidding us, aren't you? How do you establish that a) the sender has sent; b) the receiver has received; c) the receiver has received what the sender sent? ie, how do you establish any communication between these two individuals is actually taking place? Shouldn't that be the first order of business? As written, it looks as though everyone is being asked to take on faith that there is anything happening here. Do you see how utterly ridiculous that is?

This reminds me of homeopathy's "memory of water" nonsense. The question is, if the water remembers, wouldn't it be filled to the brim (!) with "memory" of everything it's ever been in contact with? And if so, how do you establish that the memory of the ingredient you (as the homeopath) is applying, is what is actually affecting the patient? Pure twaddle!


M.

It gets worse. Assuming the match rate is 25% is bogus.

For the 'auto ganzfeld' where the receiver chose the picture they think was 'sent'.

First you would run a trial without a sender, just select the picture and then have receiver do the 'ganzfeld' state and then pick a picture.

This would then generate the 'raw pick probability' for each picture. Some pictures are going to be picked more often than others and some less. (A great study in and of itself.) So You Can Not Say that the Base Chance is 25%.

For some pictures the 'raw pick rate' is going to be higher.

If the 'sending' picture is always one of theose pictures in teh set, then the 'hit' rate is always going to be above twenty five percent.

To be a good protocol, every picture in the set would have to be matched and have the same 'raw pick rate' and in fact a second round of testing would be needed to assure that.

Then each 'sent' picture would have to be chosen at random from a matched set. That way you would have a better chance of the base rate being 25%

Did you read that Rodney?

Do you get it?

 

[Startz]

According to the Wikipedia article:

"...During this time a 'sender' observes a randomly chosen target and tries to mentally send this information to the receiver.

It gets worse. Assuming the match rate is 25% is bogus.

I don't really understand how these Ganzfield things are run, but as long as the sender's target is chosen randomly and everything is properly blinded, why wouldn't the match rate be 25%?

 

[Klimax]

I don't really understand how these Ganzfield things are run, but as long as the sender's target is chosen randomly and everything is properly blinded, why wouldn't the match rate be 25%?

Unless you get true random function where each thing has same chance being chose with uniform distribution ,you have pseudogenerator and it wil cause some things being more likely selected then others.(Like Gaus distribution and such) Quite interesting part of statistics.

 

[Startz]

I don't really understand how these Ganzfield things are run, but as long as the sender's target is chosen randomly and everything is properly blinded, why wouldn't the match rate be 25%?

Unless you get true random function where each thing has same chance being chose with uniform distribution ,you have pseudogenerator and it wil cause some things being more likely selected then others.(Like Gaus distribution and such) Quite interesting part of statistics.

I would think "chosen randomly" would mean each potential target would have a 1 in 4 chance of being chosen. Do the people running these experiments do something else?

 

[Rodney]

I would think "chosen randomly" would mean each potential target would have a 1 in 4 chance of being chosen. Do the people running these experiments do something else?

If you read the Wikipedia article (clearly most posters here have not), you will find that even skeptic Ray Hyman concedes that "the contemporary ganzfeld experiments display methodological and statistical sophistication well above previous parapsychological research. Despite better controls and careful use of statistical inference, the investigators seem to be getting significant results that do not appear to derive from the more obvious flaws of previous research."

 

[Jackalgirl]

If you read the Wikipedia article (clearly most posters here have not), you will find that even skeptic Ray Hyman concedes that "the contemporary ganzfeld experiments display methodological and statistical sophistication well above previous parapsychological research. Despite better controls and careful use of statistical inference, the investigators seem to be getting significant results that do not appear to derive from the more obvious flaws of previous research."

Okay, so it seems (at least via Wikipedia) that there may be something going on. Do you know what the methodology of this test is? If so, why not apply? If you don't, can you find out, and then apply?

Is it the media presence / academic affidavit part that's holding you up? Have your researcher friends published? Would they be willing to let you apply on their behalf and do all of the heavy lifting for them? If you provide a precise claim and a detailed protocol to JREF in an application, and if you satisfy all the other requirements of the Challenge, then you'll have an answer -- either the JREF will accept your application (in which case, the answer is "Yes, the JREF will accept this particular Ganzfield setup") or it will not (and will probably [hopefully] explain, at which point the answer will be "No, the JREF will not accept this particular Ganzfield setup" -- and, hopefully, we'll know why).

But given the JREF's unwillingness to play "what-if" games and enter into hypothetical discussions of imaginary Challenge applications, it really does look like you'll have to apply if you want to find out whether the particular Ganzfield protocol that you have in mind would be acceptable.

You could always go look at other Challenges (e.g., IIG), too. If you can pass theirs, it would satisfy the notoriety requirements of the Challenge. Might even get you straight to the formal demonstration of the MDC, IIRC.

 

[Rodney]

Okay, so it seems (at least via Wikipedia) that there may be something going on. Do you know what the methodology of this test is? If so, why not apply? If you don't, can you find out, and then apply?

Is it the media presence / academic affidavit part that's holding you up? Have your researcher friends published? Would they be willing to let you apply on their behalf and do all of the heavy lifting for them? If you provide a precise claim and a detailed protocol to JREF in an application, and if you satisfy all the other requirements of the Challenge, then you'll have an answer -- either the JREF will accept your application (in which case, the answer is "Yes, the JREF will accept this particular Ganzfield setup") or it will not (and will probably [hopefully] explain, at which point the answer will be "No, the JREF will not accept this particular Ganzfield setup" -- and, hopefully, we'll know why).

But given the JREF's unwillingness to play "what-if" games and enter into hypothetical discussions of imaginary Challenge applications, it really does look like you'll have to apply if you want to find out whether the particular Ganzfield protocol that you have in mind would be acceptable.

You could always go look at other Challenges (e.g., IIG), too. If you can pass theirs, it would satisfy the notoriety requirements of the Challenge. Might even get you straight to the formal demonstration of the MDC, IIRC.

Thanks for the advice, but I don't have the time to get involved in such a quixotic venture. I'm merely suggesting that it would be desirable for the JREF to clarify its position on Ganzfeld experiments, which clearly involve a significant expenditure of time.

 

[Jackalgirl]

Thanks for the advice, but I don't have the time to get involved in such a quixotic venture. I'm merely suggesting that it would be desirable for the JREF to clarify its position on Ganzfeld experiments, which clearly involve a significant expenditure of time.

I doubt they will, though. I'm fairly sure that they'll take it on a claim-by-claim basis. Some experiements might lend themselves better to the MDC than others depending on how their claims are worded and how their protocols are structured.

If JREF came back to you and said, "maybe; it depends on the claim", would you accept that?

 

[Jackalgirl]

...and would you accept it if -- if you were to start proposing hypothetical experiments and situations -- JREF replied with "we will not entertain hypotheticals; apply or go away"?

 

[GzuzKryzt]

Thanks for the advice, but I don't have the time to get involved in such a quixotic venture.
...

A Freudian slip or a satori?

 

[Dancing David]

I would think "chosen randomly" would mean each potential target would have a 1 in 4 chance of being chosen. Do the people running these experiments do something else?

There are two issues to start,

traditional Ganzfeld: receiver generates a 'word string' (IR talks about the mental pictures) this is then rated and matched as a hit.

auto-Ganzfeld: receiver choses the picture that they think was 'sent' from a set of four pictures.

As frequency statistics are based upon actual occurance,
traditional Ganzfeld: multiple pictures in a set can have 'hits' with a 'word string'. This can mean that a set could have a 'random word string match rate' of 25%, 50%, 75% or 100%.

auto-Ganzfeld: pictures can have higher rates of 'random pick' than others. For various reasons, certain pictures could get chosen at higher rates just because receivers 'like' them more than others. If this sort of picture appears in the sequence of 'sent' pictures more than 25% of the time, this could bias the hit rate.


And then there are many other confounding factors to the '25% base chance rate'.

Psi could exist, Ganzfeld is not structured to show that,

 

[Dancing David]

If you read the Wikipedia article (clearly most posters here have not), you will find that even skeptic Ray Hyman concedes that "the contemporary ganzfeld experiments display methodological and statistical sophistication well above previous parapsychological research. Despite better controls and careful use of statistical inference, the investigators seem to be getting significant results that do not appear to derive from the more obvious flaws of previous research."

What methods and protocols?

An improvement there may be, show us the actual procedures and protocols.

As stated in other threads Rodney, this is an issue in all areas of research, not just psi.

 

[Dancing David]

Thanks for the advice, but I don't have the time to get involved in such a quixotic venture. I'm merely suggesting that it would be desirable for the JREF to clarify its position on Ganzfeld experiments, which clearly involve a significant expenditure of time.

The procedure for the application is what it is. The issues of the claimants making the claim is upon the claimant. Not the JREF. Given the number of people interested in such things, you could run a study at low cost. You could probably even get donate space and materials.

 

[Startz]

There are two issues to start,

traditional Ganzfeld: receiver generates a 'word string' (IR talks about the mental pictures) this is then rated and matched as a hit.

auto-Ganzfeld: receiver choses the picture that they think was 'sent' from a set of four pictures.

As frequency statistics are based upon actual occurance,
traditional Ganzfeld: multiple pictures in a set can have 'hits' with a 'word string'. This can mean that a set could have a 'random word string match rate' of 25%, 50%, 75% or 100%.

auto-Ganzfeld: pictures can have higher rates of 'random pick' than others. For various reasons, certain pictures could get chosen at higher rates just because receivers 'like' them more than others. If this sort of picture appears in the sequence of 'sent' pictures more than 25% of the time, this could bias the hit rate.


And then there are many other confounding factors to the '25% base chance rate'.

Psi could exist, Ganzfeld is not structured to show that,

Very helpful. I certainly see why the "traditional Ganzfeld" doesn't have a 25 percent chance.

 

[RoboTimbo]

If it isn't likely that there will be a forthcoming challenge application, perhaps this thread should be merged with the one in General Skepticism. If, or when, a psi researcher gets their muddled testing protocols straightened out, then they could start their own thread here.

 

[Rodney]

I doubt they will, though. I'm fairly sure that they'll take it on a claim-by-claim basis. Some experiements might lend themselves better to the MDC than others depending on how their claims are worded and how their protocols are structured.

If JREF came back to you and said, "maybe; it depends on the claim", would you accept that?

Possibly -- if they confirmed that they would accept a 1000 hour+ Ganzfeld protocol.

 

[Rodney]

...and would you accept it if -- if you were to start proposing hypothetical experiments and situations -- JREF replied with "we will not entertain hypotheticals; apply or go away"?

I have no intention of proposing hypotheticals -- I just want to understand the JREF's position on whether Ganzfeld experiments qualify for the MDC.

 

[Rodney]

There are two issues to start,

traditional Ganzfeld: receiver generates a 'word string' (IR talks about the mental pictures) this is then rated and matched as a hit.

auto-Ganzfeld: receiver choses the picture that they think was 'sent' from a set of four pictures.

As frequency statistics are based upon actual occurance,
traditional Ganzfeld: multiple pictures in a set can have 'hits' with a 'word string'. This can mean that a set could have a 'random word string match rate' of 25%, 50%, 75% or 100%.

auto-Ganzfeld: pictures can have higher rates of 'random pick' than others. For various reasons, certain pictures could get chosen at higher rates just because receivers 'like' them more than others. If this sort of picture appears in the sequence of 'sent' pictures more than 25% of the time, this could bias the hit rate.


And then there are many other confounding factors to the '25% base chance rate'.

Psi could exist, Ganzfeld is not structured to show that,

Your concerns were addressed by more recent Ganzfeld experiments, as Ray Hyman concedes.

 

[Jackalgirl]

Possibly -- if they confirmed that they would accept a 1000 hour+ Ganzfeld protocol.

I think you misunderstood what I was saying. If JREF says "maybe, it depends on the claim", they would be saying "maybe, it depends on the claim".

In other words, someone would have to actually make the claim involving a 1000 hour+ protocol in order to find out whether JREF would accept it.