Evolution of telepathy?

[Pyrrho]

Originally posted by Interesting Ian
But I have already explained why it is not suggestive! :mad: You have just contradicted me but have failed to provide any reasons to call into question my original point.

If JREF has the power to refuse any applications what's to stop them from refusing any application that they feel has a tangible chance of winning the prize?
Perhaps there are some legal considerations? I don't know, but it would seem to me that Randi could not turn aside a legitimate application without incurring some problems.

[davidsmith73]

Originally posted by Pyrrho

Perhaps there are some legal considerations? I don't know, but it would seem to me that Randi could not turn aside a legitimate application without incurring some problems.

He could turn aside any application by saying that it is not a paranormal claim. He could then give an explanation as to why he thought so and not have to justify the validity of his explanation, but its not even clear from the application page that he is even required to do that.

[MRC_Hans]

Originally posted by Interesting Ian

But I have already explained why it is not suggestive! :mad: You have just contradicted me but have failed to provide any reasons to call into question my original point.

No, you have suggested why it is not suggestive. I dare to disagree.

If JREF has the power to refuse any applications what's to stop them from refusing any application that they feel has a tangible chance of winning the prize?

Because they would be open to scandal. If anybody were to publish a scientifically viable protocol plus documentation that the JREF had refused to accept it, the JREF would loose all credibility in the skeptic community. And forget about fan group or cult, there is nothing that enrages skeptics more than a false skeptic. Something you may have noticed reading this board.

Hans

[Interesting Ian]

Originally posted by MRC_Hans
Because they would be open to scandal. If anybody were to publish a scientifically viable protocol plus documentation that the JREF had refused to accept it, the JREF would loose all credibility in the skeptic community. And forget about fan group or cult, there is nothing that enrages skeptics more than a false skeptic. Something you may have noticed reading this board.



No, can't say I've noticed that at all!

But to the main point. How would people know about it if they don't publish details of the applications they have turned down??

[BillHoyt]

Originally posted by Nivek


This confuses me..:confused:
How can a phylosophy on the nature of knowledge allow one to deny knowledge? Can this truely be trustworthy? It seems like a contradiction to me.....

Knowledge, the last I looked, doesn't include things that are false. The methods of science are the gating mechanism that screens out the false and lets in the true.

Cheers,

[BillHoyt]

Originally posted by davidsmith73
You see, you have displayed the exact kind of bias I am talking about. If you were the JREF member that decided on the basis of a paranormal claim, you would have rejected a ganzfeld protocol. The claimant would then have to go home, the claim not tested, regardless of whether or not anybody else thought the claim was legitimately paranormal.

On the contrary, you are displaying the exact kind of nonsense for which JREF will not jeopardize a million. If your only ganzfeld design is so squishy, then you inherently have only pap to serve, sir. If ganzfeld falls apart under a more rigorous definition, then it was a house of cards to begin with.

What parameters are there for this "judging"? What good old fashioned hard rules? An apple can be called a Big Mac? That is permitted? Why? Can any food be substituted for any food? Only fruit for fruit? Only fruit of a different size? How about fruit pies? If I guess mince meat pie does that count as an apple hit? Why not open it up more broadly? An apple is made of molecules. Molecules are made of atoms. So anything made of atoms is a hit?

I don't think so.

Now lets go a bit further. Given you are unwilling to modify ganzfeld so that a "hit" must be a "hit", or even to define hard bounds around hit, how can anybody estimate the probabilities?

Give me a design. Hard and fast. Let's see if you can do something other than whine and smear.

[BillHoyt]

Originally posted by Interesting Ian


Now I find this extremely suspicious. What's to stop Randi from refusing any application which is likely to win?

Well, gee, we would records of correspondence, right? From both sides, right? So, how about you stop this smear campaign and cough up some evidence of all these horrid things that might be? Where are the letters from such an applicant and responses from JREF?

[davidsmith73]

Originally posted by BillHoyt


On the contrary, you are displaying the exact kind of nonsense for which JREF will not jeopardize a million. If your only ganzfeld design is so squishy, then you inherently have only pap to serve, sir. If ganzfeld falls apart under a more rigorous definition, then it was a house of cards to begin with.

What parameters are there for this "judging"? What good old fashioned hard rules? An apple can be called a Big Mac? That is permitted? Why? Can any food be substituted for any food? Only fruit for fruit? Only fruit of a different size? How about fruit pies? If I guess mince meat pie does that count as an apple hit? Why not open it up more broadly? An apple is made of molecules. Molecules are made of atoms. So anything made of atoms is a hit?

I don't think so.

Now lets go a bit further. Given you are unwilling to modify ganzfeld so that a "hit" must be a "hit", or even to define hard bounds around hit, how can anybody estimate the probabilities?

Give me a design. Hard and fast. Let's see if you can do something other than whine and smear.

Bill, you display the most pathetic understanding of the ganzfeld procedure I have heard for a while, wait no, Ken's is worse.

The judging procedure allows for a big mac to be called a hit when an apple is the target because the judge does not know which target is the actual target and which are the decoys. Thus the hit rate can be compared with chance over successive trials.

Are you really this stupid or are you deliberately winding me up ?

[davidsmith73]

Originally posted by BillHoyt


Well, gee, we would records of correspondence, right? From both sides, right? So, how about you stop this smear campaign and cough up some evidence of all these horrid things that might be? Where are the letters from such an applicant and responses from JREF?


Don't be stupid, he's talking about the procedure allowing the possibility of this happening. Just like if there is the possibility of fraud in a parapsychology experiment, that would decrease the credibility of the results. Same principle.

[thaiboxerken]

This is exactly my point. The JREF will not accept tests that try to replicate the scientific evidence for these phenomena, for example ganzfeld ESP, simply because they do not believe the evidence shows anything paranormal.

The ganzfield experiments only show that people can torture numbers to suggest whatever they feel like believing in. No science there.


Therefore the JREF should just stay out of the debate as to the existence of paranormal phenomena. Thats up to the scientific method to decide. The scientific method does deem any topic ouside its scope of investigation because it doesn't pass some biased and silly criteria.

That's your opinion, it's wrong, but you're welcome to it. One doesn't need to have a science research center to point out the obviousm, and the obvious is that there is no such thing as "psi" outside the context of fiction.



I was making sarcastic reference to your own pathetic understanding of paranormal phenomena :rolleyes:

Hardly, you are the one that wants to keep the lines between reality and "paranormal" blurry, that way you can make up any lie and call it evidence.

[thaiboxerken]

Do you know what a ganzfeld judging procedure is ?

The judge has to chose out of four possible targets which they think is the actual target for that trial. The judge does not know which is the correct target. The free response procedure is set up so that any possible ESP signals that produce mentation that is similar to the actual target choice is more likely to result in the correct target choice. This method is in good agreement with how normal psychology works. For example, if you watch a TV program about house fires before you go to sleep, you are likely to dream about things psychologically related to fire such as erupting volcanoes, firemen etc. You do not necessarily dream specifically about house fires.

Do you have anything reasonably intelligent to say ?

That judgement procedure makes the entire process subjective and not objective at all. This is not science, it's garbage. Do you have any real evidence of "psi" at all? I thought not, ladybrook.

[BillHoyt]

Originally posted by davidsmith73


Bill, you display the most pathetic understanding of the ganzfeld procedure I have heard for a while, wait no, Ken's is worse.

The judging procedure allows for a big mac to be called a hit when an apple is the target because the judge does not know which target is the actual target and which are the decoys. Thus the hit rate can be compared with chance over successive trials.

Are you really this stupid or are you deliberately winding me up ?

Really, david? A big mac is a hit? So if the target is an apple, what other subject responses constitute "hits"? Please exhaustively enumerate them.

[edited for clarity -bh]

[thaiboxerken]

Originally posted by BillHoyt


Really, david? A big mac is a hit? So if the target is an apple, what other subject responses constitute "hits"? Please exhaustively enumerate them.



Whatever it takes to "prove" there is psi, I guess.

:rolleyes:

It's about as increble as Luci's "ladybrook" evidence of psi. LOL.

[Nivek]

Originally posted by BillHoyt


Knowledge, the last I looked, doesn't include things that are false. The methods of science are the gating mechanism that screens out the false and lets in the true.

Cheers,

knowl·edge n.

1: The state or fact of knowing.

2: Familiarity, awareness, or understanding gained through experience or study.

3: The sum or range of what has been perceived, discovered, or learned.

4: Learning; erudition: teachers of great knowledge.

5: Specific information about something.

6: Carnal knowledge.

Middle English knoulech_: knouen, to know; see know + -leche, n. suff.]

Synonyms: information, learning, erudition, lore, scholarship

Source: The American Heritage® Dictionary of the English Language, Fourth Edition
Copyright © 2000 by Houghton Mifflin Company.
Published by Houghton Mifflin Company. All rights reserved

What I've been trying to say all along, there is nothing false about experiance, experiance is knowledge. People experiance telepathy and empathy and can attest their knowledge of it, producing evidence. Yet this evidence is completely rejected by those whom cannot attest to it, for they have yet the experiance. Like I have said before, there is no proof only perception.

[BillHoyt]

Originally posted by Nivek
What I've been trying to say all along, there is nothing false about experiance, experiance is knowledge. People experiance telepathy and empathy and can attest their knowledge of it, producing evidence. Yet this evidence is completely rejected by those whom cannot attest to it, for they have yet the experiance. Like I have said before, there is no proof only perception.

I outlined the definition used here. Amazing that science makes so much progress given that "there is no proof only perception."

[Pyrrho]

Originally posted by Nivek
What I've been trying to say all along, there is nothing false about experiance, experiance is knowledge. People experiance telepathy and empathy and can attest their knowledge of it, producing evidence. Yet this evidence is completely rejected by those whom cannot attest to it, for they have yet the experiance. Like I have said before, there is no proof only perception.
When people attest to their experiences, that is anecdotal evidence, which is just fine for parties, talk shows, and books aimed at lowest-common-denominator audiences. When it comes to science, anecdotal evidence isn't good enough. There can be plenty false about experience, and that's why personal experiences are regarded as unreliable and therefore useless evidence.

[Pyrrho]

quote:
--------------------------------------------------------------------------------
Originally posted by Pyrrho
Then there is this oft-ignored paragraph, which pretty much puts the lie to any nonsensical idea that the JREF Challenge is a scam:


quote:
--------------------------------------------------------------------------------

Please be advised that several claimants have suffered great personal embarrassment after failing these tests. I strongly advise you to conduct proper double-blind tests of any ability you believe you can demonstrate, before attempting to undergo a testing for this prize. This has saved me and many claimants much time and work, by showing that the powers were quite imaginary on the part of the would-be claimant. Please do this, and do not choose to ignore the need for such a precaution.
-- James Randi
--------------------------------------------------------------------------------

Originally posted by Interesting Ian:

How does it do that?


Well, in my opinion, Randi wouldn't have said this if he were running a scam. In that paragraph, he clearly recommends that people do some scientific tests of their supposed abilities before taking the JREF Challenge. This applies to all potential claimants, including tv psychic mediums.

To take this a bit further, those who don't like the rules for the JREF Challenge should easily be able to have their supposed abilities tested as Randi advises. If their abilities prove to be genuine, the JREF million-dollar-prize would be mere pocket change in comparison to the potential rewards for such genuine abilities. I should explain that by the term "abilities", I include miraculous things such as free energy machines and the like.

If, for example, the girl who claims to see through goggles had done as advised, she and her handlers could have avoided time, expense, and embarrassment.

[LightKnight]

Bill

LK,

Just so you know, under some general guidelines Randi posted to the board a while back, you are now refusing to support your claim. It is an arrogant refusal at that, considering this noblesse oblige feint of yours. Us unwashed masses will melt with this knowledge? In the absence of evidence to the contrary, I say: pah! It is more probable your claim will melt under scrutiny.



You would be correct in saying that I refuse to support my claim with a demonstration. I believed I pointed that out in what I said, regardless of how Mr. Randi's guidelines work. Again, I see someone take offense over my words. Please, take care to try and understand them. I point out that while some may not be able to deal with what I share in, I myself am incapable of dealing with situations that other people have no problem with. I am not singling anyone out. I am not saying that all skeptics are weak. You could be someone capable of dealing with it, just as I. I'm just saying that in our individuality, some are more capable than others in certain situations.



I believe I'm beginning to see why no one has ever won this test. For one, it singles out the greedy among us. Nearly all of those that I would consider legitamate would scoff at this test. It would be the equivalent of selling your soul... trading freely given gifts for vast amounts of money. Only those who lack understanding in themselves would seek such a challenge, only those who wish to prove themselves. Anyone who truly understands their gift, who has come far with it learns that there is but one being who we need to prove ourselves to.

While I came here looking for intellectually stimulating discussion, I will leave now, with disgust. The few of you honestly looking for debate are overshadowed by the idocy displayed by your 'peers.'

Tear this apart as you will. I must be running away because I cannot make a legitamate claim. Golly, what a coward I am.

Peace to you all

[thaiboxerken]

You would be correct in saying that I refuse to support my claim with a demonstration.

That's because you're afraid you'll find out that you really don't have superpowers, and your roleplaying would cease.

You could be someone capable of dealing with it, just as I. I'm just saying that in our individuality, some are more capable than others in certain situations.

I think most skeptics could easily deal with such a thing. I know I could. Just show me your superpowers, I would be forced to believe it then.


I believe I'm beginning to see why no one has ever won this test. For one, it singles out the greedy among us.

No one has won the test because people don't have superpowers. That's the point. Oh, and it's false to say that it singles out the greedy, as one can always give the money to charity. You have a superpower, so you claim, that means you have the power to send a million dollars to help feed starving children or other charitable causes.


Nearly all of those that I would consider legitamate would scoff at this test.

I agree.. except instead of scoffing, they are running and avoiding the test. They know it will expose their trickery.

It would be the equivalent of selling your soul... trading freely given gifts for vast amounts of money.

Again, this is false. The money is there and one can do whatever they want with it once they win the money. It would really be enlighting all of us skeptics if someone beat the test. You would be doing humankind a favor by showing the world what powers we all can have and use for good.


Only those who lack understanding in themselves would seek such a challenge, only those who wish to prove themselves.

Only those with geniune superpowers could win the test. Those that refuse the test are cowards.

Anyone who truly understands their gift, who has come far with it learns that there is but one being who we need to prove ourselves to.

Yea, this is an evasion. You don't have to prove it to anyone because you know you have the power. Yet.. you want to come in here and declare to us that you have the power. Why even tell us that you have superpowers if you aren't going to prove it to us? You are a liar, you don't have superpowers.


While I came here looking for intellectually stimulating discussion, I will leave now, with disgust. The few of you honestly looking for debate are overshadowed by the idocy displayed by your 'peers.'

Good riddance. And if you don't read this, maybe you can telepathically get the message.

[b]
Tear this apart as you will. I must be running away because I cannot make a legitamate claim. Golly, what a coward I am.
[B]

I agree, you are a coward.

[BillHoyt]

Originally posted by LightKnight
You would be correct in saying that I refuse to support my claim with a demonstration. I believed I pointed that out in what I said, regardless of how Mr. Randi's guidelines work. Again, I see someone take offense over my words.
If you wish to build a home with no foundation, expect it to blow over at the first hint of wind. We didn't even get rocking here. We didn't turn the ignition key. I certainly didn't take offense at your words. I find your pompous arrogance mildly amusing.
Please, take care to try and understand them. I point out that while some may not be able to deal with what I share in, I myself am incapable of dealing with situations that other people have no problem with. I am not singling anyone out. I am not saying that all skeptics are weak. You could be someone capable of dealing with it, just as I. I'm just saying that in our individuality, some are more capable than others in certain situations.
Uh-huh. Whatever. BTW, you mangled your second sentence beyond comprehension. Unless, of course, you mean to say what we already know. Whatever.
I believe I'm beginning to see why no one has ever won this test. For one, it singles out the greedy among us. Nearly all of those that I would consider legitamate would scoff at this test.
Right. Right. Above it all. Yes, that was clear from your "unwashed masses" poppycock, oh gummy guru.
It would be the equivalent of selling your soul... trading freely given gifts for vast amounts of money. Only those who lack understanding in themselves would seek such a challenge, only those who wish to prove themselves. Anyone who truly understands their gift, who has come far with it learns that there is but one being who we need to prove ourselves to.

While I came here looking for intellectually stimulating discussion, I will leave now, with disgust. The few of you honestly looking for debate are overshadowed by the idocy displayed by your 'peers.'

Tear this apart as you will. I must be running away because I cannot make a legitamate claim. Golly, what a coward I am.

Peace to you all
You're a put on, aren't you? Who are you really? Nobody is this big a nitwit. Fess up, dude.

[Nivek]

Originally posted by BillHoyt
You're a put on, aren't you? Who are you really? Nobody is this big a nitwit. Fess up, dude.

He's no put on, nor a niwit. Just a person looking for civilized discussion, unfortunately that is hard to find here.

[T'ai Chi]

Originally posted by Nivek
He's no put on, nor a niwit. Just a person looking for civilized discussion, unfortunately that is hard to find here.

Try not to speak so rationally Nivek, you might scare the spooks away.

It is far easier to put them on ignore. My only complaint is that there aren't enough spaces on the list for all the pseudo-skeptics (read: "believers of the 2nd kind").

[thaiboxerken]

Originally posted by T'ai Chi


Try not to speak so rationally Nivek, you might scare the spooks away.

It is far easier to put them on ignore. My only complaint is that there aren't enough spaces on the list for all the pseudo-skeptics (read: "believers of the 2nd kind").

:dl:

Nivek.. rationally.. what a riot!! You should've been a comedian.

[T'ai Chi]

I'm concerned about the argument, not the person. What nivek said in that particular post made sense.

[thaiboxerken]

Originally posted by T'ai Chi
I'm concerned about the argument, not the person. What nivek said in that particular post made sense.

Interesting.. you hold value in addressing the arguement and not attacking the person, right?

I wonder.. how does this post Try not to speak so rationally Nivek, you might scare the spooks away.

It is far easier to put them on ignore. My only complaint is that there aren't enough spaces on the list for all the pseudo-skeptics (read: "believers of the 2nd kind"). fit into your addressing the arguement values?

You shouldn't be crying ad-hominem when you are doing it yourself.

[T'ai Chi]

Clearly there is a difference Ken. You, and Bill, called out names at a specific person, when I was speaking in general of certain harmful groups and their thinking habits.

[BillHoyt]

Originally posted by T'ai Chi
I'm concerned about the argument, not the person. What nivek said in that particular post made sense.
O frapjous day! Then you wouldn't mind defeding your argument that we need to test each and every person before we can say there are no superpowers. Looking forward to it!

Cheers,

[thaiboxerken]

Originally posted by T'ai Chi
Clearly there is a difference Ken. You, and Bill, called out names at a specific person, when I was speaking in general of certain harmful groups and their thinking habits.


So, ad-hominem is ok if it's directed at groups of people, but not if it's directed at a single person? That clears it up for me.

Go harness chi or something.

[davidsmith73]

Originally posted by thaiboxerken


Do you have anything reasonably intelligent to say ?

That judgement procedure makes the entire process subjective and not objective at all. This is not science, it's garbage. Do you have any real evidence of "psi" at all? I thought not, ladybrook.

No Ken, the receivers mentation is subjective but the judging procedure is objective because the hit rate is comparable to chance. Like I said, anything intelligent to say ?

[davidsmith73]

Originally posted by BillHoyt


Really, david? A big mac is a hit? So if the target is an apple, what other subject responses constitute "hits"? Please exhaustively enumerate them.

[edited for clarity -bh]


You don't have to name them you fool. Thats the whole point of a free response followed by judging protocol. You can compare the hit rate to chance. If the receiver is obtaining any information relevant to the correct target, the judge will chose the correct target more often than 25% of the time.

I don't believe you don't understand this. I think you are just whining because you know you are wrong.

[BillHoyt]

David,

I repeat:

Really, david? A big mac is a hit? So if the target is an apple, what other subject responses constitute "hits"? Please exhaustively enumerate them.

[Pyrrho]

http://moebius.psy.ed.ac.uk/Text/Ganzfeld.html

http://moebius.psy.ed.ac.uk/Text/Ganzfeld.html#ov

Highlighted in bold are aspects which I find troubling:


The receiver remains in the Ganzfeld setting for about 30 minutes, reporting via a microphone, any mental imagery, thoughts or bodily sensations that they experience. After the impression period, the experimenter reviews the receiver's impressions with them, to remind them of their experiences.

The receiver then removes their eye shields and is shown video-clips via a monitor of four possible targets, one of which is a duplicate of the actual target for the session and the other three being decoys. The receiver and/or an independent judge ranks or rates each of the four targets according to the degree of similarity to the impressions. A successful session is when the actual target receives the highest rating/ranking.

As there is a one out of four chance of "guessing" the right target by chance alone (i.e. 25%), a successful study is one in which the correct target was chosen significantly more often than 25% of the time.


Why is this sort of thing valid at all? First, the experimenter is directly involved with the sender at some point; then, "the targets are rated according to the degree of similarity", which means that "hits" can be ambiguous. Because of the scoring methods and subjective judging, the statement "there is a one in four chance" is just not correct. Hits are counted if the guess is similar to the target but not exactly on target. It's like shooting an arrow at a target, hitting an outer ring, and calling it a bullseye.

Then there's this:


Sending Period: The experimenter then starts the session. First of all, relaxation instructions (heard by both the receiver and the sender) are played. After this, white noise is played over the audio channel for the duration of the session. During this, the sender will be shown the first playing of the computer-selected video clip and will attempt to silently communicate this to the receiver. This target video clip is repeated (typically 9 times) during the 30 minute sending period. The sender can hear the receiver's mentation (i.e. their verbal report of their ongoing thoughts, feelings and impressions)and is instructed to try to mentally reinforce those aspects which relate to the target. During this time, the receiver's mentation is also recorded and a summarised by the experimenter for use during the judging procedure.

Judging Period: At the end of the sending period, the experimenter reviews the mentation with the receiver over the audio link. The receiver is, at this point, allowed to add any further details or thoughts that they might have had. The receiver is then asked to remove the eye shields and watch their monitor. They are then shown four video clips, one at a time, and allowed to comment on each one. They are then asked to assign a rating from 1 to 99 to each clip, this representing the degree of correspondance of each video clip to the mentation. The highest rating is taken to be their first choice, the second highest their second choice, and so on.

Why all the involvement by the experimenter? Why isn't the mentation itself being scored, instead of having the receiver score the video clips as compared to the mentation -- after, it must be noted, discussion with the experimenter? The way this process is designed, the outcome depends upon subjective judgment.

[davidsmith73]

Originally posted by BillHoyt
David,

I repeat:

Really, david? A big mac is a hit? So if the target is an apple, what other subject responses constitute "hits"? Please exhaustively enumerate them.

Do you understand what I just wrote ?

A hit is a hit if the judge choses the correct target. The content of the mentation either contains information that will enable the correct target to be chosen more often than chance or it will not.

So a big mac is a hit only if the apple is the correct target. If the apple was the decoy and it was chosen then the "big mac" would be regarded as a miss.

get it yet ?

Getting back to the original point. The JREF has the power to decide if this protocol is "good enough" for the test.

[thaiboxerken]

Originally posted by davidsmith73



You don't have to name them you fool. Thats the whole point of a free response followed by judging protocol. You can compare the hit rate to chance. If the receiver is obtaining any information relevant to the correct target, the judge will chose the correct target more often than 25% of the time.

I don't believe you don't understand this. I think you are just whining because you know you are wrong.

LOL. The judge now chooses the target instead of the subject now. Interesting. How does the judge decide what is relative to the target?

The target is an apple. Big Mac is close enough because it's kind of the same shape (if you're a believer). Banana is relevant because it's a fruit. Wagon is red, so it's relevant. Candy is sweet, apples are sweet.. relevant. Jar.. applesauce is stored in jars, so it's relevant.

Can you not see that this judgement procedure is so subjective that any results are unreliable?

Wow, talk about junk-science.

[thaiboxerken]

Originally posted by davidsmith73


Getting back to the original point. The JREF has the power to decide if this protocol is "good enough" for the test.

Yes they sure do have that power. It doesn't take much expertise to realize that the Ganzfield protocol is flimsy, flexible and allows for whatever results the "scientists" have predetermined.

It's BS, it's crap, it's junk-science.

[Pyrrho]

Originally posted by thaiboxerken
Yes they sure do have that power. It doesn't take much expertise to realize that the Ganzfield protocol is flimsy, flexible and allows for whatever results the "scientists" have predetermined.

It's BS, it's crap, it's junk-science.
Is it just us, are we somehow not seeing clearly? Maybe we're not smart enough to overthink it enough. It sure looks over-intellectualized to me, and I cannot for the life of me figure out why people would even bother to examine all the statistical procedures when the very structure of these "experiments" is so clearly bogus. The experimenters have more or less created an amorphous "hit field" around each target. No two "hit fields" can be the same size, because each target has a different set of things that might be considered similar, which in turn depends upon what the judge thinks is similar. And that's the key here: it depends upon what the judge thinks is similar, based on what appears to be stream-of-consciousness "mentation" from the receiver.

[davidsmith73]

Originally posted by Pyrrho
http://moebius.psy.ed.ac.uk/Text/Ganzfeld.html

http://moebius.psy.ed.ac.uk/Text/Ganzfeld.html#ov

Highlighted in bold are aspects which I find troubling:


After the impression period, the experimenter reviews the receiver's impressions with them, to remind them of their experiences.


Nothing wrong here. The experimenter does not know the identity of the target.


The receiver and/or an independent judge ranks or rates each of the four targets according to the degree of similarity to the impressions.


Nothing wrong here. This is a subjective process. Each judge will have different responses to the mentation and will judge differently. However, randomisation of the targets will nullify this subjectivity.


As there is a one out of four chance of "guessing" the right target by chance alone (i.e. 25%), a successful study is one in which the correct target was chosen significantly more often than 25% of the time.


Nothing wrong here. Elementary statistics.


Why is this sort of thing valid at all? First, the experimenter is directly involved with the sender at some point;

before the session is started. The ratings are secured before the sender leaves the room. Thus the sender does not have any contact with anyone from the moment the target is revealed to them and until after the judging has been done.


then, "the targets are rated according to the degree of similarity", which means that "hits" can be ambiguous. Because of the scoring methods and subjective judging, the statement "there is a one in four chance" is just not correct.

yes it is because the targets are randomised.


Hits are counted if the guess is similar to the target but not exactly on target. It's like shooting an arrow at a target, hitting an outer ring, and calling it a bullseye.

That does not matter because randomisation of the targets means that if chance were alone operating then 25% hits would be expected.


The sender can hear the receiver's mentation (i.e. their verbal report of their ongoing thoughts, feelings and impressions)and is instructed to try to mentally reinforce those aspects which relate to the target. During this time, the receiver's mentation is also recorded and a summarised by the experimenter for use during the judging procedure.


Nothing wrong here. The mic is one way. the receiver does not hear anything from the sender. The experimenter does not know the target identity.


At the end of the sending period, the experimenter reviews the mentation with the receiver over the audio link. The receiver is, at this point, allowed to add any further details or thoughts that they might have had.


Nothing wrong here. The experimenter does not know the target identity.


They are then asked to assign a rating from 1 to 99 to each clip, this representing the degree of correspondance of each video clip to the mentation.


Nothing wrong here. The randomisation of targets eliminates the subjective aspect of the judges choice.


Why all the involvement by the experimenter? Why isn't the mentation itself being scored, instead of having the receiver score the video clips as compared to the mentation -- after, it must be noted, discussion with the experimenter?

What would scoring the mentation acheive ? You want to know if the target can be chosen correctly.

Also the experimenter does not know the target identity so discussion with them is fine. I guess its done to try to encourage the receiver to report as much detail as possible.


The way this process is designed, the outcome depends upon subjective judgment.

No it doesn't because the targets are randomised.

[davidsmith73]

Originally posted by thaiboxerken



Can you not see that this judgement procedure is so subjective that any results are unreliable?

Wow, talk about junk-science.


Randomisation of the targets Ken. Talk about not understanding the protocol.

[Pyrrho]

Originally posted by davidsmith73
Randomisation of the targets Ken. Talk about not understanding the protocol.
The benefit of randomization is negated by the judging procedure. You're still left with a purely subjective scoring system that renders the "1 in 4 chance" statement incorrect. The final statistical analysis is not done on a 1 in 4 chance of picking a target, it's done on a subjective score of 1 to 99 for similarity to each target. Because of the subjectivity, the true chance is ambiguous.

[BillHoyt]

Originally posted by davidsmith73


Do you understand what I just wrote ?

A hit is a hit if the judge choses the correct target. The content of the mentation either contains information that will enable the correct target to be chosen more often than chance or it will not.

So a big mac is a hit only if the apple is the correct target. If the apple was the decoy and it was chosen then the "big mac" would be regarded as a miss.

get it yet ?

Getting back to the original point. The JREF has the power to decide if this protocol is "good enough" for the test.
doh. doh. doh. Duhh, no, davie, still having just a bit of a problem here. The problem with this totally bogus procedure is exactly the same as it has always been. An apple is the target. The judge gets to call a "Big Mac" "mentation" a hit? No bounds, no rules, no parameters?

You say "its science" and I say "its nonsense",
You say "mentation" and I say "its mental",
let's call the whole thing off.

Give me the rules, clearly definable rules. You CANNOT measure probability with this fuzzy definition. get that yet?

[thaiboxerken]

Nothing wrong here. This is a subjective process. Each judge will have different responses to the mentation and will judge differently. However, randomisation of the targets will nullify this subjectivity.

This is false, random or not, the judge is still subjectively choosing whether the subject's guess is a hit or miss. This makes the test completely illegitimate. This is junk-science, and not matter how much jargon and BS you throw out, it is still junk-science.

[davidsmith73]

Originally posted by Pyrrho

The benefit of randomization is negated by the judging procedure. You're still left with a purely subjective scoring system that renders the "1 in 4 chance" statement incorrect. The final statistical analysis is not done on a 1 in 4 chance of picking a target, it's done on a subjective score of 1 to 99 for similarity to each target. Because of the subjectivity, the true chance is ambiguous.


If the targets are randomised then you would expect to pick the correct target 25% of the time regardless of any subjective biases you have towards particular images. The scoring of 1-99 still means that one image is chosen out of four. If that image has a one in four chance of being selected as the correct target then you would expect it to be a hit one time out of four, in the long run.

[davidsmith73]

Originally posted by BillHoyt

doh. doh. doh. Duhh, no, davie, still having just a bit of a problem here. The problem with this totally bogus procedure is exactly the same as it has always been. An apple is the target. The judge gets to call a "Big Mac" "mentation" a hit? No bounds, no rules, no parameters?

You say "its science" and I say "its nonsense",
You say "mentation" and I say "its mental",
let's call the whole thing off.

Give me the rules, clearly definable rules. You CANNOT measure probability with this fuzzy definition. get that yet?

I find it very ammusing that you still do not understand the ganzfeld protocol.

Actually, now that I think about it, its quite worrying. Are you representative of the standard of "sceptical" intelligence ? I seriously hope not.

[davidsmith73]

Originally posted by thaiboxerken


This is false, random or not, the judge is still subjectively choosing whether the subject's guess is a hit or miss. This makes the test completely illegitimate. This is junk-science, and not matter how much jargon and BS you throw out, it is still junk-science.

Of course the judge will be subjective in his choice of target. However, the randomisation of the targets means that any particular image has a one in four chance of being selected as the target. Over an adequate number of trials, there will be a 25% hit rate if information about the correct target is not obtained by the receiver.

[thaiboxerken]

Originally posted by davidsmith73


Of course the judge will be subjective in his choice of target. However, the randomisation of the targets means that any particular image has a one in four chance of being selected as the target. Over an adequate number of trials, there will be a 25% hit rate if information about the correct target is not obtained by the receiver.

This is also false, depending upon the subjectivity and how wide the judge decides to make the target, the fields can overlap and a correct hit can be procured for a guess that fits more than one target. It's junk science. Clearly, you do not understand the fallible nature of the protocol.

[davidsmith73]

Originally posted by thaiboxerken


This is also false, depending upon the subjectivity and how wide the judge decides to make the target, the fields can overlap and a correct hit can be procured for a guess that fits more than one target. It's junk science. Clearly, you do not understand the fallible nature of the protocol.


what do you mean by "how wide the judge decides to make the target" ? what overlaping fields ?

[thaiboxerken]

Originally posted by davidsmith73



what do you mean by "how wide the judge decides to make the target" ? what overlaping fields ?

Since the judgement is subjective, the targets are wider and can easily overlap.

Target #1 a ball
Target #2 an apple.

If the subject guesses, "round", the guess fits both targets. Do you understand now? Why is it so hard for believers to see the problems associated with the 'protocol'?

[davidsmith73]

Originally posted by thaiboxerken


Since the judgement is subjective, the targets are wider and can easily overlap.

Target #1 a ball
Target #2 an apple.

If the subject guesses, "round", the guess fits both targets. Do you understand now? Why is it so hard for believers to see the problems associated with the 'protocol'?


So what ? If the guess fits both targets then the judge will have to chose either the ball or apple. If the actual target is randomised so that its the ball half the time and the apple the other half the time, the judge will chose the correct target 50% of the time regardless of whether he prefers the apple or the ball.

That is unless the receiver obtains target relevent information in which case the target will be chosen correctly more than 50% of the time.

Why is it so hard for pseudosceptics to see the reasons behind ganzfeld protocol ?

[Pyrrho]

Originally posted by davidsmith73
If the targets are randomised then you would expect to pick the correct target 25% of the time regardless of any subjective biases you have towards particular images. The scoring of 1-99 still means that one image is chosen out of four. If that image has a one in four chance of being selected as the correct target then you would expect it to be a hit one time out of four, in the long run.
No, it doesn't. It means that a guess that is wide of the mark can still be scored as a hit if it falls within a range of similarity subjectively determined by the judge. There is no meaningful way to determine the true statistical chance of this.

[davidsmith73]

Originally posted by Pyrrho

No, it doesn't. It means that a guess that is wide of the mark can still be scored as a hit if it falls within a range of similarity subjectively determined by the judge. There is no meaningful way to determine the true statistical chance of this.


If a guess is wide of the mark but falls within a range of similarity subjectively determined by the judge, he will pick the image that this guess is similar to. That doesn't make it a hit. What makes it a hit is if the judge choses the correct target from the four possible choices. The image that the judge picks as being similar to the guess might not be the actual target. In fact, the image that the judge picks has a one in four chance of being the correct target.

[thaiboxerken]

So what ? If the guess fits both targets then the judge will have to chose either the ball or apple. If the actual target is randomised so that its the ball half the time and the apple the other half the time, the judge will chose the correct target 50% of the time regardless of whether he prefers the apple or the ball.

This makes the test invalid, as it's no longer testing the subject's abilities. Why can't you see that? Is your mind so closed that you can't see the fallibility of the test?

So now you have 1-4 chance changed to 1-4 chance with a 50% chance that the judge will pick the correct guess that the subject is guessing. What's that do for the statistics? You really, really need to get your logic circuits checked.

[thaiboxerken]

Originally posted by davidsmith73



If a guess is wide of the mark but falls within a range of similarity subjectively determined by the judge, he will pick the image that this guess is similar to. That doesn't make it a hit. What makes it a hit is if the judge choses the correct target from the four possible choices. The image that the judge picks as being similar to the guess might not be the actual target. In fact, the image that the judge picks has a one in four chance of being the correct target.

Can't you see that the subject is no longer being tested with this protocol, but the judge is instead? Now, it's not the subject picking the target, but the judge. That's junk-science.

[davidsmith73]

Originally posted by thaiboxerken

So what ? If the guess fits both targets then the judge will have to chose either the ball or apple. If the actual target is randomised so that its the ball half the time and the apple the other half the time, the judge will chose the correct target 50% of the time regardless of whether he prefers the apple or the ball.

This makes the test invalid, as it's no longer testing the subject's abilities. Why can't you see that? Is your mind so closed that you can't see the fallibility of the test?

So now you have 1-4 chance changed to 1-4 chance with a 50% chance that the judge will pick the correct guess that the subject is guessing. What's that do for the statistics? You really, really need to get your logic circuits checked.

I was talking about your example with two target choices - the ball and the apple. Two target choices = 50 % chance. 4 target choices = 25% chance. Anyone with even a remote amount if intelligence could have gathered that.

[thaiboxerken]

Originally posted by davidsmith73


I was talking about your example with two target choices - the ball and the apple. Two target choices = 50 % chance. 4 target choices = 25% chance. Anyone with even a remote amount if intelligence could have gathered that.

You still don't get it, the vagueness and subjectivity of the test renders any and all of the statistics null. The protocol isn't testing the subject, it's testing the judges ability to make the evidence fit a predetermined conclusion.

[Pyrrho]

Originally posted by davidsmith73
If a guess is wide of the mark but falls within a range of similarity subjectively determined by the judge, he will pick the image that this guess is similar to. That doesn't make it a hit. What makes it a hit is if the judge choses the correct target from the four possible choices. The image that the judge picks as being similar to the guess might not be the actual target. In fact, the image that the judge picks has a one in four chance of being the correct target.
As I understand this, to put it simply, the judge assigns a similarity rating of 1 to 99, and chooses images that have high similarity ratings. If the receiver says "ball", and the target is an apple, the judge would assign a high similarity score, and choose the apple, even though the receiver has not said "apple". If the target is in fact an apple, it is scored as a hit, even though the receiver never guessed "apple".

Now, because the receiver gives more detail than just a simple guess, the scoring system is by default based on fuzzy logic. You still cannot legitimately claim a 1 in 4 chance of choosing a target.

[davidsmith73]

Originally posted by thaiboxerken


Can't you see that the subject is no longer being tested with this protocol, but the judge is instead? Now, it's not the subject picking the target, but the judge. That's junk-science.

Firstly, the receiver is also allowed to be the judge in which case your above statement is simply BS.

Second, an independent judge uses the receievers mentation as information upon which to make their choice. If ESP is enabling the receiver to obtain information about the target, the judge will be able to use it.

Is there no end to your lack of understanding ?

[thaiboxerken]

Originally posted by davidsmith73


Firstly, the receiver is also allowed to be the judge in which case your above statement is simply BS.

Second, an independent judge uses the receievers mentation as information upon which to make their choice. If ESP is enabling the receiver to obtain information about the target, the judge will be able to use it.

Is there no end to your lack of understanding ?

Your belief has really closed your mind to the reality of the situation. The protocol is designed such that it is unreliable, unspecific and unobjective. The judge's subjectivity is introduced into the test no matter what the "randomness" of the targets is. Simply put, they can make up whatever results they like if they really need to. The ganzfield experiments show about as much evidence of psi as Luci's 'ladybrook' claim.

I understand the experiment and how it was conducted, I also understand that it's junk-science.

[davidsmith73]

Originally posted by Pyrrho

As I understand this, to put it simply, the judge assigns a similarity rating of 1 to 99, and chooses images that have high similarity ratings. If the receiver says "ball", and the target is an apple, the judge would assign a high similarity score, and choose the apple, even though the receiver has not said "apple". If the target is in fact an apple, it is scored as a hit, even though the receiver never guessed "apple".

Correct.


Now, because the receiver gives more detail than just a simple guess, the scoring system is by default based on fuzzy logic. You still cannot legitimately claim a 1 in 4 chance of choosing a target.

If the judge is given 4 images, he is not 25% likely to chose one over the other three. Is that what you are saying ?

[davidsmith73]

Originally posted by thaiboxerken


Your belief has really closed your mind to the reality of the situation. The protocol is designed such that it is unreliable, unspecific and unobjective. The judge's subjectivity is introduced into the test no matter what the "randomness" of the targets is. Simply put, they can make up whatever results they like if they really need to. The ganzfield experiments show about as much evidence of psi as Luci's 'ladybrook' claim.

I understand the experiment and how it was conducted, I also understand that it's junk-science.

You clearly don't understand the experiment. You have not explained in any way how the protocol is unreliable, unspecific and unobjective. You just keep whining like a little child. Got anything intelligent to say on the subject ?

[thaiboxerken]

The fallibility of the protocol has been pointed out throughout this thread. You just refuse to believe it. That's ok, this is why the experiments won't be winning any Nobel prizes. You can believe what you want, but these experiments have not withstood the scrutiny of the scientific community because of the flimsy protocols.

Now, if you have or know anyone with superpowers, feel free to have them demonstrate their powers for the JREF.

[davidsmith73]

Originally posted by thaiboxerken


Now, if you have or know anyone with superpowers, feel free to have them demonstrate their powers for the JREF.

I thought it wouldn't be long ! If there is real evidence out there deflect attention from it and appeal to the JREF prize !

[Interesting Ian]

Originally posted by Pyrrho

Is it just us, are we somehow not seeing clearly?

Quite a coincidence because I was just thinking "is it me, can it possibly be me that's incredibly thick, or is the so called skeptics"?? I absolutely am completely baffled about your criticisms of the ganzfeld procedure. What the hell is your problem with it??? :confused: You say:


Why is this sort of thing valid at all? First, the experimenter is directly involved with the sender at some point;


Yes what's wrong with this?? Obviously the experimenter is clueless as to the correct target, so what on earth is the problem??


then, "the targets are rated according to the degree of similarity", which means that "hits" can be ambiguous.


Oh right, so because ones impressions are not 100% the same as the picture that the sender is attempting to send this means that no anomalous cognition exists whatsoever! :eek:



Because of the scoring methods and subjective judging, the statement "there is a one in four chance" is just not correct.



Er . . .sorry I'm completely flummoxed by this assertion! Er . . . so what is the probability then if it isn't 25%?? 26%, 35%? 50% 95%? Let me know. Normally the chances of getting one target out of 4 is 25%. I'd be interested in your reasoning that it is actually higher than this! :eek:

[Interesting Ian]

Originally posted by thaiboxerken


Yes they sure do have that power. It doesn't take much expertise to realize that the Ganzfield protocol is flimsy, flexible and allows for whatever results the "scientists" have predetermined.

It's BS, it's crap, it's junk-science.

I'm sorry, but from what Pyrrho posted it looks absolutely sound. What on earth is wrong with it??

[Martin]

Originally posted by davidsmith73
If the judge is given 4 images, he is not 25% likely to chose one over the other three. Is that what you are saying ?It should be self-evident that if the images are, say, an apple, a brick, a car, and detailed blueprints of a nuclear reactor, then the probabilities that the receiver's mentations will correspond to each of these are manifestly not equal, psi or no psi.

[Interesting Ian]

Originally posted by Pyrrho

The benefit of randomization is negated by the judging procedure. You're still left with a purely subjective scoring system that renders the "1 in 4 chance" statement incorrect. The final statistical analysis is not done on a 1 in 4 chance of picking a target, it's done on a subjective score of 1 to 99 for similarity to each target. Because of the subjectivity, the true chance is ambiguous.

No, the chance must be 1 in 4 if the normally sensory channels are closed off. I mean it must be if there is no anomalous cognition. So is your claim that in fact the normal sensory channels are not completely closed off? If so could you expand on this?

I'm just trying to understand from where you've coming from. To be honest I don't think you're coming from anywhere. But people keep accusing me of being arrogant and not listening to other people.

[Interesting Ian]

Originally posted by Martinm
It should be self-evident that if the images are, say, an apple, a brick, a car, and detailed blueprints of a nuclear reactor, then the probabilities that the receiver's mentations will correspond to each of these are manifestly not equal, psi or no psi.

:confused: Even if there is a psychological pre-disposition to prefer certain images to others I cannot understand how the probability could be more than 25% unless these psychologically desirable images were deliberately chosen by somebody as a target. Sorry, I just don't understand. Enlighten me.

[Martin]

Originally posted by Interesting Ian
No, the chance must be 1 in 4 if the normally sensory channels are closed offThat would be true if we were talking about simply guessing which image was the target. But we're not looking at the probability of picking the right target. We're talking about the probability that, given the receiver's mentations and the judge's interpretation of these, the correct image will be chosen. That additional information is there, whether or not psi exists.

[Interesting Ian]

Originally posted by Martinm
That would be true if we were talking about simply guessing which image was the target. But we're not looking at the probability of picking the right target. We're talking about the probability that, given the receiver's mentations and the judge's interpretation of these, the correct image will be chosen. That additional information is there, whether or not psi exists.

So if psi doesn't exist what is this "additional information" then? I'm sorry, I still don't understand.

To recap, if the target that the sender will attempt to send is chosen wholly randomly out of the 4 images, then even though psi doesn't exist, and even though there is no sensory leakage whatsoever, then the probability that the judge will choose the right target will in fact be greater than 25% over a large number of trials. This is because of this "additional information".

Now what is this "additional information" precisely?

[thaiboxerken]

No, this is false, it's not a 1 in 4 chance. It's entirely up to the judge, the judge can say.. well close enough and make it virtually impossible to miss. If a bigmac is close enough to being an apple, then the chance of scoring goes well above and beyond 25%.

The nature of this BS experiment is that it's judgement procedure is 100% subjective.

That's not science, it's ********.

[thaiboxerken]

Originally posted by Interesting Ian

Oh right, so because ones impressions are not 100% the same as the picture that the sender is attempting to send this means that no anomalous cognition exists whatsoever! :eek:


Yes. This is true. If a sender is trying to send an image of an apple, but the reciever pictures a Big Mac instead.. there is no psi activity going on here. How hard is that to understand? When Big Macs and apples are "close enough", then the science has been tossed out of the door.

[Interesting Ian]

Originally posted by thaiboxerken
No, this is false, it's not a 1 in 4 chance. It's entirely up to the judge, the judge can say.. well close enough and make it virtually impossible to miss. If a bigmac is close enough to being an apple, then the chance of scoring goes well above and beyond 25%.

The nature of this BS experiment is that it's judgement procedure is 100% subjective.

That's not science, it's bullsh*t.

Huh?? But the judge doesn't know what the target is! He has the receivers impressions, and he has to choose one of the 4 targets which seems to fit most closely to those impressions. So he has 1 in 4 of a chance assuming no psi and no sensory leakage, right?? The only way it couldn't be was if the initial randomly chosen target was psychologically one which we might tend to choose. But there will be both images which we would psychogically tend to choose, and images which psychologically we would tend to avoid. So therefore over a number of trials, on average, the probability must be 25%, unless there is something wrong with the laws of probability! :eek:

Sorry, but what am I missing?? I repeat. I need 2 questions answered:

a) What is the nature of this additional information?

b) Where does it come from?

[thaiboxerken]

But the judge doesn't know what the target is! He has the receivers impressions, and he has to choose one of the 4 targets which seems to fit most closely to those impressions.

What impression? If the sender is sending an apple, and the reciever picks a big mac, and they say it's close enough.. something is definitely wrong.

[thaiboxerken]

So therefore over a number of trials, on average, the probability must be 25%, unless there is something wrong with the laws of probability

This might be the case in an objective study, but if the judgements are subjective, probabilities get thrown out the door.

[ceptimus]

Both the 'receiver', and the 'independant judge' have no knowledge (other than ESP) about which of the four video clips has been randomly chosen.

The long winded discussion/rating procedure between them is irrelevant - eventually they select one of the four videos.

If the correlation between their chosen video, and the randomly chosen video is better than chance (25%) then something funny is going on.

I would first suspect cheating and/or poor experimental conditions (maybe the judge or the receiver can hear what the sender is doing if the audio screening isn't good enough). However, if there is no cheating involved, and both receiver and judge are ignorant of which video was chosen, then this looks like good evidence for ESP to me, and ought to qualify for the JREF $1 million challenge.

[edited: spelling]

[thaiboxerken]

Bem, D.J., Palmer, J. & Broughton, R.S. (2001). Updating the Ganzfeld database: A victim of its own success? Journal of Parapsychology, 65, 207-218.

" The existence of psi-anomalous processes of information transfer such as telepathy or clairvoyance-continues to be controversial. Earlier meta-analyses of studies using the ganzfeld procedure appeared to provide replicable evidence for psi (D. J. Bern & C. Honorton, 1994), but a follow-up meta-analysis of 30 more recent ganzfeld studies did not (J. Milton & R. Wiseman, 1999). "

Sounds like the results of these Ganzfield tests are inconsistent and unreliable.

[Interesting Ian]

Originally posted by thaiboxerken

II
But the judge doesn't know what the target is! He has the receivers impressions, and he has to choose one of the 4 targets which seems to fit most closely to those impressions.

TBK

What impression? If the sender is sending an apple, and the reciever picks a big mac, and they say it's close enough.. something is definitely wrong.


Let's say the receiver gets a vague impression through psi of roundness from the sender trying to project the image of an apple, and the receiver says "I am getting an impression of a round object". Now if in the choice of the 4 pics there is an apple and a big mac, and the judge picks a big mac, then that is a miss! So here we have psi operating but it actually counts as a miss. So what are you whining about??

[thaiboxerken]

Originally posted by Interesting Ian



Let's say the receiver gets a vague impression through psi of roundness from the sender trying to project the image of an apple, and the receiver says "I am getting an impression of a round object". Now if in the choice of the 4 pics there is an apple and a big mac, and the judge picks a big mac, then that is a miss! So here we have psi operating but it actually counts as a miss. So what are you whining about??

No, because of the nature of the test, the judge can assign an equal number to both. Remember they assign 1-99 for each item. If the judge places a 80 on 2 of them, then both can be considered a hit by the experimentor. Simply put, they can make the results what they want based on "close enough".

Also, the validity of the Ganzfield experiments have been in question for a while now, as noted in the snapshot above.

[Interesting Ian]

Originally posted by thaiboxerken
Bem, D.J., Palmer, J. & Broughton, R.S. (2001). Updating the Ganzfeld database: A victim of its own success? Journal of Parapsychology, 65, 207-218.

" The existence of psi-anomalous processes of information transfer such as telepathy or clairvoyance-continues to be controversial. Earlier meta-analyses of studies using the ganzfeld procedure appeared to provide replicable evidence for psi (D. J. Bern & C. Honorton, 1994), but a follow-up meta-analysis of 30 more recent ganzfeld studies did not (J. Milton & R. Wiseman, 1999). "

Sounds like the results of these Ganzfield tests are inconsistent and unreliable.

I read something about that I think about a few weeks ago. I think they failed to include the most recent experiments or something, allowing the results to slide into non-significance. There was some other problem with analysis by Milton as well I think, but unable to recall or even remember where I saw it.

[thaiboxerken]

Originally posted by Interesting Ian


I read something about that I think about a few weeks ago. I think they failed to include the most recent experiments or something, allowing the results to slide into non-significance. There was some other problem with analysis by Milton as well I think, but unable to recall or even remember where I saw it.

There were 10 more experiments done after the Miltoin analysis. These other experiments shows much less signficant results. Other analysts included many early experiments which were already shown to be unreliable. To put it in perspective, there are those in the paranormal research world that are out there to prove that psi exists, not to see if it really does exist.

[Interesting Ian]

Originally posted by thaiboxerken


No, because of the nature of the test, the judge can assign an equal number to both. Remember they assign 1-99 for each item. If the judge places a 80 on 2 of them, then both can be considered a hit by the experimentor. Simply put, they can make the results what they want based on "close enough".



Well I find it difficult to believe this happens. I would have thought that the judge would be required to put them into a definite order. But obviously if they didn't that would seem to me to be a cheat, yes. Perhaps DavidSmith can confirm your assertion??



Also, the validity of the Ganzfield experiments have been in question for a while now, as noted in the snapshot above. [/B]

I don't think your quote demonstrates this. You'll need to provide details to back your allegation up.

[Interesting Ian]

Originally posted by thaiboxerken


Yes. This is true. If a sender is trying to send an image of an apple, but the reciever pictures a Big Mac instead.. there is no psi activity going on here.



I suspect it wouldn't be as definite as that. I suspect that the receiver is more likely to get an impression of roundness. But let's suppose you are correct. Yes there will be times when a guess is scored as a hit even though no psi is operating.

But this is why we can only conclude psi if the scores are consistently above 25% rather than simply consistently above 0%!!

[Interesting Ian]

Right we've had one skeptic (ceptimus) who agrees with me and DavidSmith. That is to say that in the absence of any sensory leakage (and of course psi), the target should only be selected 25% of the time. Now do any other skeptics agree with this, or are the rest of you still maintaining that even though there is no psi and no sensory leakage, the target, on average, will be chosen more than 25% of the time?

If so then then what is this "additional information"?

To recap, all skeptics apart from ceptimus are stating that if the target that the sender will attempt to send is chosen wholly randomly out of the 4 images, then even though psi doesn't exist, and even though there is no sensory leakage whatsoever, then the probability that the judge will choose the right target will in fact be greater than 25% over a large number of trials. This is because of some "additional information".

Now what is this "additional information" precisely?

[BillHoyt]

Originally posted by davidsmith73


I find it very ammusing that you still do not understand the ganzfeld protocol.

Actually, now that I think about it, its quite worrying. Are you representative of the standard of "sceptical" intelligence ? I seriously hope not.
Oh, I understand it perfectly. It is patent nonsense. Others have already pointed out its many flaws. The absence of sensory leakage is a sham when the judge steps in to judge the results. The a priori probability claims are also a sham with such an incredibly loose protocol.

If you cannot construct a tight ganzfeld, then the probability that ganzfeld is real is vanishingly small. If you cannot tell me, a prior what mumblings can be translated into what objects, then the priorities cannot be stated up front.

[Martin]

Originally posted by Interesting Ian
To recap, all skeptics apart from ceptimus are stating that if the target that the sender will attempt to send is chosen wholly randomly out of the 4 images, then even though psi doesn't exist, and even though there is no sensory leakage whatsoever, then the probability that the judge will choose the right target will in fact be greater than 25% over a large number of trials. This is because of some "additional information"I never said anything about a large number of trials. I'm still trying to work this through in my head, incidentally. I'm rather confused as to whether or not we have to make some assumptions about the nature of the targets to guarantee a 25% hit rate over a large number of trials. Either way, the subjective nature of the judging and the possibility that some targets are in some sense more appealing than others suggest to me that the results should be very variable. Which, in fact, they are.

[Interesting Ian]

Originally posted by Martinm
To recap, all skeptics apart from ceptimus are stating that if the target that the sender will attempt to send is chosen wholly randomly out of the 4 images, then even though psi doesn't exist, and even though there is no sensory leakage whatsoever, then the probability that the judge will choose the right target will in fact be greater than 25% over a large number of trials. This is because of some "additional information"
--------------------------------------------------------------------------------

I never said anything about a large number of trials.



Well, if you have in mind one single trial, and by chance a relatively psychologically desirable image (relative to the other 3) is randomly selected for the sender to send, then it would seem there will be a greater than 25% of a chance it will be chosen by the judge (even assuming no psi and sensory leakage). But even here the chance is only 25% before that image was randomly selected. Remember, there is also an equal chance that a relatively psychologically undesirable image will be randomly selected for the sender to attempt to send. But anyway, obviously we would have to have a fair number of trials before even the most tentative conclusions about any anomolous cognition could be drawn.




I'm still trying to work this through in my head, incidentally. I'm rather confused as to whether or not we have to make some assumptions about the nature of the targets to guarantee a 25% hit rate over a large number of trials.



Well let's think about this. First of all let's assume that psi doesn't exist and there is no sensory leakage. Now if all 4 images are all roughly of equal psychological appeal, then there is not much problem. But let's suppose that 1 image is of extreme psychological appeal and the other 3 images the absolute converse. Let's suppose the desirable image is randomly selected for the sender to attempt to send. The receiver obviously picks up nothing because psi doesn't exist. Nevertheless, psychologically s/he might be more disposed to feel the image is a pleasant one in nature and will therefore will say some appropriate comments reflecting this. The judge might then, on the whole, feel more disposed to pick the pleasant (ie correct) image. Thus if the pleasant image had been randomly selected, then a fair case can be made that from the moment after the random selection, the chance of picking the correct target will be greater than 25%.

But here is the crucial point. Assuming that the image is properly randomly selected at the beginning of the trial, it will be 3 times more likely that an unpleasant image will be the target. Obviously (given psi doesn't exist), the reciever will say the same things, and the judge will be more likely to pick the pleasant image (ie the wrong one). Conversely the chances of picking the correct target will now be less than 25%! Provided the original image is properly randomly selected it really doesn't matter about the nature of the images because it will even itself out so that the chances of picking the target will be 25%. This will be so even for a single trial before the image is initially randomly selected. And if you think about it, the chance must be 25% unless there is some leakage of information, or some additional information is available to the receiver/judge.



Either way, the subjective nature of the judging and the possibility that some targets are in some sense more appealing than others suggest to me that the results should be very variable. Which, in fact, they are.

Well I think I have refuted this just above, but let me know what you think.

Of course if psi does exist then it certainly seems to me the results will vary. So if for example all 4 images are of all large round objects then I would expect the judge to eventually pick the correct target than if all 4 images were radically different. But it must be 25% assuming no PSI or sensory leakage.

[ceptimus]

Originally posted by Interesting Ian
Now do any other skeptics agree with this, or are the rest of you still maintaining that even though there is no psi and no sensory leakage, the target, on average, will be chosen more than 25% of the time?This thread has quietened down a lot since this post. Where has everyone gone? Oh well, I suppose it was during the weekend...

[Ipecac]

What amazes me is that the same people who support this Ganzfeld procedure are the ones who complain that the JREF Challenge is biased.

[thaiboxerken]

Originally posted by Ipecac
What amazes me is that the same people who support this Ganzfeld procedure are the ones who complain that the JREF Challenge is biased.

They also happen to be believers of the paranormal. The Ganzfield experiments didn't convince them, they were already convinced, they see something that looks scientific and they have to cling on it so they can show something to skeptics and say "see it really is true". The fact is, the Ganzfield experiments are designed that the results have been unreliable. Oh well, thus is the mind of a believer, they'll cling onto anything that supports their beliefs including junk-science.

[Ipecac]

Originally posted by thaiboxerken
They also happen to be believers of the paranormal. The Ganzfield experiments didn't convince them, they were already convinced, they see something that looks scientific and they have to cling on it so they can show something to skeptics and say "see it really is true". The fact is, the Ganzfield experiments are designed that the results have been unreliable. Oh well, thus is the mind of a believer, they'll cling onto anything that supports their beliefs including junk-science.

Of course you're right. Perhaps "amazes me" wasn't the correct expression. :)

[ceptimus]

I don't believe in the paranormal. I don't believe that properly conducted ganzfeld protocols would show any better (or worse) results than those predicted by chance (25% in the version discussed here).

I also don't believe you have explained where is the 'flaw' in the experiments described. Just why would you expect better than 25% 'hits' just because two people, niether of whom know which target was chosen, go through some arcane process before making their selection?

That the ganzfeld protocol is unnecessarily complex seems to me beyond doubt, but I have not yet seen a convincing explanation of how it is unfair.

[Interesting Ian]

Originally posted by thaiboxerken
The fact is, the Ganzfield experiments are designed that the results have been unreliable.

Well I keep asking the skeptics but they keep ignoring me :confused: I repeat:

Right we've had one skeptic (ceptimus) who agrees with me and DavidSmith. That is to say that in the absence of any sensory leakage (and of course psi), the target should only be selected 25% of the time. Now do any other skeptics agree with this, or are the rest of you still maintaining that even though there is no psi and no sensory leakage, the target, on average, will be chosen more than 25% of the time?

If so then then what is this "additional information"?

To recap, all skeptics apart from ceptimus are stating that if the target that the sender will attempt to send is chosen wholly randomly out of the 4 images, then even though psi doesn't exist, and even though there is no sensory leakage whatsoever, then the probability that the judge will choose the right target will in fact be greater than 25% over a large number of trials. This is because of some "additional information".

Now what is this "additional information" precisely?

[thaiboxerken]

Milton, J. & Wiseman, R. (2001). Does Psi Exist? Reply to Storm and Ertel (2001). Psychological Bulletin, 127, 434-438.

The authors recently published a nonsignificant meta-analysis of 30 extrasensory perception ganzfeld studies. all conducted after the 1986 publication of important methodological guidelines aimed at reducing sources of artifact noted in earlier studies. In response, L. Storm and S. Ertel (2001) presented a meta-analysis of 79 studies published between 1974 and 1996. They argued that the positive and highly statistically significant overall outcome indicates a replicable paranormal effect. In doing so, they ignored the well-documented and widely recognized methodological problems in the early studies, which make it impossible to interpret the results as evidence of extrasensory perception. In addition, Storm and Ertel's meta-analysis is not an accurate quantitative summary of ganzfeld research because of methodological problems such as their use of an inconsistent method for calculating study outcomes and inconsistent inclusion criteria.

[davidsmith73]

Originally posted by Martinm
It should be self-evident that if the images are, say, an apple, a brick, a car, and detailed blueprints of a nuclear reactor, then the probabilities that the receiver's mentations will correspond to each of these are manifestly not equal, psi or no psi.

Absolutely. I have said this all along. But the randomisation of the targets nullifies this subjectivity.

[davidsmith73]

Originally posted by Interesting Ian


Well I keep asking the skeptics but they keep ignoring me :confused: I repeat:

Right we've had one skeptic (ceptimus) who agrees with me and DavidSmith. That is to say that in the absence of any sensory leakage (and of course psi), the target should only be selected 25% of the time. Now do any other skeptics agree with this, or are the rest of you still maintaining that even though there is no psi and no sensory leakage, the target, on average, will be chosen more than 25% of the time?

If so then then what is this "additional information"?

To recap, all skeptics apart from ceptimus are stating that if the target that the sender will attempt to send is chosen wholly randomly out of the 4 images, then even though psi doesn't exist, and even though there is no sensory leakage whatsoever, then the probability that the judge will choose the right target will in fact be greater than 25% over a large number of trials. This is because of some "additional information".

Now what is this "additional information" precisely?

I somehow doubt that either Ken or Bill Hoyt will give us an answer. They still clearly do not grasp the reasoning behind the ganzfeld protocol. I'm still holding out for Pyrrho to tell me what he mean't by his fuzzy logic thing though. Also, thanks ceptimus for speaking up against fellow sceptics, now there's a rarity !

But I have one objection to something ceptimus said:

"That the ganzfeld protocol is unnecessarily complex seems to me beyond doubt"


I don't see how the protocol is unnecessarily complex at all. Which particular part do you feel is unnecessarily complex ?

The free response element of the protocol enables the receiever to let their mind wander and report aloud any thoughts or images that they may receive via ESP. If the ESP signal is weak then it follows that it will be weakly incorporated into normal internal imagination. This weak interaction may be expressed as an increase in target relevant mentation rather than an exact description of the target. It just makes sense to use a free response protocol.

[ceptimus]

Originally posted by davidsmith73


I somehow doubt that either Ken or Bill Hoyt will give us an answer. They still clearly do not grasp the reasoning behind the ganzfeld protocol. I'm still holding out for Pyrrho to tell me what he mean't by his fuzzy logic thing though. Also, thanks ceptimus for speaking up against fellow sceptics, now there's a rarity !

But I have one objection to something ceptimus said:

"That the ganzfeld protocol is unnecessarily complex seems to me beyond doubt"


I don't see how the protocol is unnecessarily complex at all. Which particular part do you feel is unnecessarily complex ?I don't really see why the 'judge' is needed at all. Why not just let the 'receiver' choose the video, or image that seems the best fit to the 'impressions (s)he received?

[thaiboxerken]

The free response element of the protocol enables the receiever to let their mind wander and report aloud any thoughts or images that they may receive via ESP. If the ESP signal is weak then it follows that it will be weakly incorporated into normal internal imagination. This weak interaction may be expressed as an increase in target relevant mentation rather than an exact description of the target. It just makes sense to use a free response protocol.

LOL. It's alot of BS that give wiggling room to the judgement procedure. The assignment of the numbers 1-99 for each choice also gives more wiggling room. One of 4 targets is selected, but the results could also include the 2nd target chosen if it was close in proximity to the chosen target.

Target one recieves 99
Target 2 recieves 90.

Target 2 was the actual target, so they say it's still kind of a hit because it's close enough. I have to wonder if these "near misses" are included in the results. If not, why the 1-99 system in the first place?

That is the nature of these types of tests, it's complete and utter BS.


Why is the judge picking the targets, and not the reciever anyway?

[davidsmith73]

Originally posted by ceptimus
I don't really see why the 'judge' is needed at all. Why not just let the 'receiver' choose the video, or image that seems the best fit to the 'impressions (s)he received?

That is sometimes what happens. To me, it seems better for the receiver to do the judging because they have a much clearer idea of thier own conscious impressions. Perhaps an independent judge is sometimes used to test whether this makes any significant difference to the hit rate. Either way, it doesn't make the protocol any more complex surely ? No statistics have changed or anything.

[ceptimus]

Fair enough. I'm still waiting for an explanation from thaiboxerken or Ipecac or BillHoyt or Pyrrho as to why they think this protocol is so 'flawed'.

[ceptimus]

Originally posted by thaiboxerken
It's alot of BS that give wiggling room to the judgement procedure. The assignment of the numbers 1-99 for each choice also gives more wiggling room. One of 4 targets is selected, but the results could also include the 2nd target chosen if it was close in proximity to the chosen target.If the experimenters actually did this then I agree that their results are worthless. Can you site any examples of such cheating?

Don't misunderstand me - if the experimenters got positive results confirming PSI, then I would first expect that they were either cheating, or somehow deluding themselves with a defective procedure. However, the protocol as described seems fair enough to me. I would like to see properly controlled tests carried out using this protocol (as long as someone else pays). I would expect negative results, of course. If the results turned out positive then, hey!, something new for science to explain.

[thaiboxerken]

Originally posted by ceptimus
If the experimenters actually did this then I agree that their results are worthless. Can you site any examples of such cheating?


I've posted the results of a couple studies done on the Ganzfield experiments. It seems the experiments aren't faring so well when reviewed by the scientific community.

[davidsmith73]

Originally posted by thaiboxerken


I've posted the results of a couple studies done on the Ganzfield experiments. It seems the experiments aren't faring so well when reviewed by the scientific community.

So do you agree that the protocol does not contain any subjective "wiggle room" ?

[davidsmith73]

Originally posted by ceptimus
If the experimenters actually did this then I agree that their results are worthless. Can you site any examples of such cheating?


I do not think that the dozen or so independent experimenters deliberately fixed the analysis in this way. Including second guesses as hits is not mentioned in any of the papers I have read.


Don't misunderstand me - if the experimenters got positive results confirming PSI, then I would first expect that they were either cheating, or somehow deluding themselves with a defective procedure. However, the protocol as described seems fair enough to me. I would like to see properly controlled tests carried out using this protocol (as long as someone else pays). I would expect negative results, of course. If the results turned out positive then, hey!, something new for science to explain.

Why have you assumed that cheating or defective procedures were involved in the ganzfeld experiments done so far ? Who would conduct your proposed replication study ? Would it be have to be experimenters openly sceptical to the existence of psi ? and why would you accept the results of those experiments (positive or negative) but not the ones done so far since the exact same protocol would apply ?

[ceptimus]

Originally posted by davidsmith73
Why have you assumed that cheating or defective procedures were involved in the ganzfeld experiments done so far ? Who would conduct your proposed replication study ? Would it be have to be experimenters openly sceptical to the existence of psi ? and why would you accept the results of those experiments (positive or negative) but not the ones done so far since the exact same protocol would apply ? Perhaps I'm just too skeptical. What are (for me) amazing claims demand very convincing proof. I will take the time to look through the literature and see whether the experiments carried out so far convince me.

[thaiboxerken]

Originally posted by davidsmith73


So do you agree that the protocol does not contain any subjective "wiggle room" ?

How in the hell did read that from I've posted the results of a couple studies done on the Ganzfield experiments. It seems the experiments aren't faring so well when reviewed by the scientific community. ?

Wow, beleivers really do see what they believe and not what's real.

[Interesting Ian]

Originally posted by thaiboxerken


How in the hell did read that from I've posted the results of a couple studies done on the Ganzfield experiments. It seems the experiments aren't faring so well when reviewed by the scientific community. ?

Wow, beleivers really do see what they believe and not what's real.

Are they not faring so well because (or partially because) they do not realise the judge on average will pick the correct target more than 25% of the time in spite of the fact that the orignial image selected for the sender to send was wholly random, and there is no sensory leakage, and psi doesn't exist?? I ask again how it can be more than 25%? Where is your reasoning?? How can parapsychologists improve their protocol if you refuse to explain their error. I mean normally the chance of picking a target out of 4 possibilities is 1/4 or 25%. So why isn't it in this case??

This is after all the point which has been debated over the past couple of pages or so. And many peoiple have agreed with you that the chance is greater than 25%, yet no-one yet has explained why! :eek:

Why is that I wonder??

[ceptimus]

Originally posted by Interesting Ian
And many peoiple have agreed with you that the chance is greater than 25%, yet no-one yet has explained why! :eek:

Why is that I wonder??
Originally posted by ceptimus
I also don't believe you have explained where is the 'flaw' in the experiments described. Just why would you expect better than 25% 'hits' just because two people, niether of whom know which target was chosen, go through some arcane process before making their selection?
Originally posted by davidsmith73
I somehow doubt that either Ken or Bill Hoyt will give us an answer. They still clearly do not grasp the reasoning behind the ganzfeld protocol. I'm still holding out for Pyrrho to tell me what he mean't by his fuzzy logic thing though.Looks like nobody wants to admit they made a mistake? Or maybe we are all just stupid, and they can't be bothered to try and explain our error to us?

[BillHoyt]

Originally posted by Interesting Ian


Are they not faring so well because (or partially because) they do not realise the judge on average will pick the correct target more than 25% of the time in spite of the fact that the orignial image selected for the sender to send was wholly random, and there is no sensory leakage, and psi doesn't exist?? I ask again how it can be more than 25%? Where is your reasoning?? How can parapsychologists improve their protocol if you refuse to explain their error. I mean normally the chance of picking a target out of 4 possibilities is 1/4 or 25%. So why isn't it in this case??

This is after all the point which has been debated over the past couple of pages or so. And many peoiple have agreed with you that the chance is greater than 25%, yet no-one yet has explained why! :eek:

Why is that I wonder??

Apparently you are unaware of the reported differences between "static targets" and "dynamic targets"? You see, ganzfeld experiments done with dynamic targets yield higher hit rates than static targets. Psi guys usually conclude the greater sensory "richness" of dynamic targets cause the increase.

I think not.

The results get skewed by the targets and the number of things allowed to count as hits. It is no longer 25% when the target's "dynamics" come into the picture. How many images the judge may map to a dynamic target becomes the very important and very skewing issue here.

[Ossai]

davidsmith73
The free response element of the protocol enables the receiever to let their mind wander and report aloud any thoughts or images that they may receive via ESP. If the ESP signal is weak then it follows that it will be weakly incorporated into normal internal imagination. This weak interaction may be expressed as an increase in target relevant mentation rather than an exact description of the target. It just makes sense to use a free response protocol.
I'll ask again. Where are the controls?
Do the receiver and the judge interact?
Does the judge and/or receiver interact with the person running the video clips?
Who chooses the clips?
Are the clips different enough that the hit zones do not overlap? Who judges this?
Why are guesses rated on a scale of 1 to 99 instead of 1 positive and 3 negatives?
If ESP existed why aren't the receivers and senders shielded or in some manner isolated?
What would be the range?
Does the sender select the images sent or are they randomly selected beforehand?
Does the sender know of the other images in the set, or the other images that are to be sent?

Ossai

[Interesting Ian]

Originally posted by BillHoyt


The results get skewed by the targets and the number of things allowed to count as hits.



The fact that the targets are dynamic rather than static matters not one iota to my argument. If dynamic targets get greater hits this is suggestive that psi is responsible, especially as this would be what we might expect from our experience of psi anyway.



It is no longer 25% when the target's "dynamics" come into the picture.



How so?




How many images the judge may map to a dynamic target becomes the very important and very skewing issue here.

I think that the judge must put the images, videos or whatever in a definite order. If he chooses the wrong one it's a miss, it's as simple as that.

Sorry, you're going to have to do better than that.

[davidsmith73]

Originally posted by Ossai
davidsmith73

I'll ask again. Where are the controls?

The experimental sessions hit rate is most often compared to that expected by chance. In addition, if the receivers mentation is recorded it is sometimes re-judged at a later date with a mismatched set of targets, ie. 3 decoys and a second selected target. In this condition the psi mediated mentation should not be relevent since it should be relevant to the first target. However, one cannot rule out the possibility that the receiver is obtaining information about the control target simply because we do not know much about how psi operates. For this reason, I think a re-judging control analysis should not be performed. Its better just to compare to theoretical chance.


Do the receiver and the judge interact?

The receiver is sometimes the judge as well. If he is not then I don't know if they interact. If they did, it would not be a problem because the judge nor the receiver knows the target identity. It is made clear that the judge is allowed no contact with anyone who knows the target identity.


Does the judge and/or receiver interact with the person running the video clips?

Yes, through a two way mic. The experimenter selects the video clips at random from his room. However, they are not shown to the experimenter, only to the sender.


Who chooses the clips?

The experimenter


Are the clips different enough that the hit zones do not overlap?

Don't understand. If you mean that the clips are selected to be very different from one another then yes.

Who judges this?

The parapsychologists, taking information from normal psychology studies I guess.


Why are guesses rated on a scale of 1 to 99 instead of 1 positive and 3 negatives?

Don't really know. I suppose its to motivate the receiver to think carefully about their choice. Again, probably a technique borrowed from normal psychology.


If ESP existed why aren't the receivers and senders shielded or in some manner isolated?

They are. Read the Edinburgh parapsychology link posted here.


What would be the range?

There's a schematic of the laboratory in that link, I think.


Does the sender select the images sent or are they randomly selected beforehand?

randomly selected.


Does the sender know of the other images in the set, or the other images that are to be sent?


Most people participate only one. Sometimes twice. I think there are about 100 images in the pool. not sure though.

[thaiboxerken]

http://moebius.psy.ed.ac.uk/

If you click on "psi" in the frame. Then ESP on the top frame, you'll see a link for Ganzfield and "findings". If you read the findings, you'll find out that the results of these Ganzfield experiments are inconclusive at the most.

[Ossai]

davidsmith73
The experimental sessions hit rate is most often compared to that expected by chance.
(bold mine) At what point would it not be compared to chance?

In addition, if the receivers mentation is recorded it is sometimes re-judged at a later date with a mismatched set of targets, ie. 3 decoys and a second selected target. Why, what purpose does this serve?

In this condition the psi mediated mentation should not be relevent since it should be relevant to the first target. However, one cannot rule out the possibility that the receiver is obtaining information about the control target simply because we do not know much about how psi operates. Actually the hit rate should again equal change because of the dynamic targeting and judging process.

For this reason, I think a re-judging control analysis should not be performed. Its better just to compare to theoretical chance. Actually simply reevaluating in that manner should illustrate the absurdity of the test to begin with.

The receiver is sometimes the judge as well. If he is not then I don't know if they interact. I thought the judge was supposed impartially rate the item selected. This isn't possible if the receiver and the judge are the same person. Either separate the positions or remove the judge position entirely.

It is made clear that the judge is allowed no contact with anyone who knows the target identity. OK.

Yes, through a two way mic. The experimenter selects the video clips at random from his room. However, they are not shown to the experimenter, only to the sender. So the clips aren't randomly selected. Are there safeguards in place to ensure that the receiver/judge does not interact with the experimenter outside the setting? Knowledge of the experimenter could influence how the judge selects the item.

Don't really know. I suppose its to motivate the receiver to think carefully about their choice. Again, probably a technique borrowed from normal psychology. If the receiver is also the judge then there will be one positive answer and three negative answers. I know someone with a degree in psychology and I'll ask them for clarification later.

They are. Read the Edinburgh parapsychology link posted here.
What would be the range?
There's a schematic of the laboratory in that link, I think.
Link, it doesn't appear in the post.

Does the sender know of the other images in the set, or the other images that are to be sent?
Most people participate only one. Sometimes twice. I think there are about 100 images in the pool. not sure though. Slight clarification to my question. Does the sender and or the receiver know of the pool of images and what the targets are?


Ossai

[davidsmith73]

Originally posted by Ossai
davidsmith73

(bold mine) At what point would it not be compared to chance?

Its always compared to chance


Why, what purpose does this serve?

The second analysis would serve as a further control if you assumed that the receiver is not receiving information about the second "pseudo" target. Since you can't rule that out I would advise against it. It's rarely done anyway I think.

Actually the hit rate should again equal change because of the dynamic targeting and judging process.

How so ? We've already established that the judging process doesn't introduce any bias to the hit rate.

Actually simply reevaluating in that manner should illustrate the absurdity of the test to begin with.

If you think that, fine. I disagree.

I thought the judge was supposed impartially rate the item selected. This isn't possible if the receiver and the judge are the same person. Either separate the positions or remove the judge position entirely.

The judge rates the images on the basis of the receivers recorded mentation. If the receiver is the judge they have their mentation recorded in the form of memory and perhaps what they wrote/spoke about. There's no need to separate any "position".



So the clips aren't randomly selected.

they are randomly selected. Thats what I wrote originally wasn't it ?

Are there safeguards in place to ensure that the receiver/judge does not interact with the experimenter outside the setting? Knowledge of the experimenter could influence how the judge selects the item.

Since the experimenter does not know the identity of the target, this doesn't seem a problem to me. What are you getting at ?


Link, it doesn't appear in the post.

http://moebius.psy.ed.ac.uk/Ganzfeld_H.html

Slight clarification to my question. Does the sender and or the receiver know of the pool of images and what the targets are?

Not that I am aware of.

[Pyrrho]

Here's the flaw, as I see it:

For each target there is a set of possible similarities. The judge (who can also be the receiver, according to what I've read, but we'll assume this is an independent judge) is potentially capable of identifying any given similarity in that set. For an equal chance of 25% for the choosing of one of four targets, the ability of the judge to identify similarities between target and mentation must be equal for all targets.

Ideally, the judge is capable of identifying any and all similarities in the set; realistically, he is capable of identifying only a subset of all possible similarities to a given target, human memory being fallible, and because an individual's life experience is limited. It is possible that a judge may be able to know and identify all possible similarities between a single target and a mentation, but it is unlikely, and it is even more unlikely that a judge would be able to know and identify all possible similarities between all targets and all mentations.

This means that the set of similarities that a judge can identify is different for each target/mentation. The true set of similarties that the judge can identify is unknown, as is the total set of possible similarities. Therefore, the statement that there is an equal 25% chance that a given target will be chosen is incorrect. The true chance cannot be calculated.

If, over many trials, the set of targets used tends toward things that the judge is more able to identify similarities for, the result of meta-analysis could be skewed toward significance. Likewise, if the set of targets tends toward things that the judge is less able to identify similarities for, the results could be skewed toward non-significance.

The rating of similarities can be questionable. For example, suppose the target was a picture of a rabbit; the receiver reports a visualization of a TV antenna. There is no physical similarity, no functional similarity, no genetic similarity. A judge might be inclined to give it a low rating. However, the two are linked in a meaningful way by the colloquial U.S. expression for TV antennas: "rabbit ears". A judge familiar with the expression might assign a high rating.

As to the question of "fuzzy logic": this is essentially how a judge would identify similarities between seemingly unrelated targets/mentations. As I understand it, if the judge is only permitted to use a restricted set of similarities, the potential for finding a psi signal would be ruined. Possibilities outside the restricted set would be discarded -- improperly, in my opinion. Only if the judge is allowed to discern similarities from an unrestricted set of possibilites can the test be at all meaningful. Pure logic would require a given set of similarities for each target. Since this would negate the potential for finding a psi signal, it cannot be done.

However, because the experiment requires "fuzzy logic", or subjective choices by the judge, the potential for finding a psi signal is also rendered meaningless, because the chances for making a target choice are unequal, as explained above. One might argue that over many trials, the inequalities would tend to balance out, but that assumption would be unwarranted.

This could be informally tested by giving a single mentation to one judge who would score the mentation as compared to the same set of four targets once a week for several weeks. The judge must not be allowed access to previous scoring. At the end of the test period, the scores should be compared. If the chance for selecting the target is an equal 25%, then all scores should be the same each time, and the same target should be chosen each time. If the chance is unequal, the scores will be different each time, and a different target could potentially be chosen. Whether the same target is chosen each week is less important than the scoring, because it is on the basis of the scores that the target is chosen. This could easily be done electronically.

This of course requires a scrupulously honest judge.

[Ossai]

davidsmith73
The second analysis would serve as a further control if you assumed that the receiver is not receiving information about the second "pseudo" target. Since you can't rule that out I would advise against it. It's rarely done anyway I think.
So the sender sends, but may be sending more than one image. The receiver receives the image, but may be receiving more than one. There are no controls in place to prevent interference. There is no equipment in place to measure/monitor any type brain activity. Since the information is being passed in a nonphysical way, it should be assumed that some form of energy is being transmitted and received. But no energy dampeners are in place to prevent interference. There are no energy (wave frequency) devices to detect any sort of unusual energy or spikes/patterns/etc in the normal ambient energy.

Actually the hit rate should again equal change because of the dynamic targeting and judging process.
How so ? We've already established that the judging process doesn't introduce any bias to the hit rate. Only if the judge and the receiver are the same person. Once the data passes to someone else and then the second person makes the decision bias is introduced.

Actually simply reevaluating in that manner should illustrate the absurdity of the test to begin with.
If you think that, fine. I disagree. Why?
Using the same 1 to 99 scale, have someone unrelated to the first test rank four additional items. Compare the results. With enough results, the hit rates for both will match.

So the clips aren't randomly selected.
they are randomly selected. Thats what I wrote originally wasn't it ?
No, originally you wrote
Who chooses the clips?
The experimenter
And
Does the sender select the images sent or are they randomly selected beforehand?
randomly selected.
and
Does the judge and/or receiver interact with the person running the video clips?
Yes, through a two way mic. The experimenter selects the video clips at random from his room. However, they are not shown to the experimenter, only to the sender.
Now from the above responses the experimenter and the sender are the same person. You have implied that the clips are randomly selected before the experiment and then you state that the sender selects the clips at the time of the experiment.

Are there safeguards in place to ensure that the receiver/judge does not interact with the experimenter outside the setting? Knowledge of the experimenter could influence how the judge selects the item.
Since the experimenter does not know the identity of the target, this doesn't seem a problem to me. What are you getting at ? So the sender does not see the receiver. Are the images of the items just broadcast and the experimenter hopes the receiver somehow manages to pick them up?

Slight clarification to my question. Does the sender and or the receiver know of the pool of images and what the targets are?
Not that I am aware of. Is there a double blind set up so that the bank of images are unknown to the sender / experimenter until the time of the test? If so, who judges that the target zone of each item does not overlap?

Ossai

[davidsmith73]

Originally posted by Pyrrho
Here's the flaw, as I see it:

For each target there is a set of possible similarities. The judge (who can also be the receiver, according to what I've read, but we'll assume this is an independent judge) is potentially capable of identifying any given similarity in that set. For an equal chance of 25% for the choosing of one of four targets, the ability of the judge to identify similarities between target and mentation must be equal for all targets.

Ideally, the judge is capable of identifying any and all similarities in the set; realistically, he is capable of identifying only a subset of all possible similarities to a given target, human memory being fallible, and because an individual's life experience is limited. It is possible that a judge may be able to know and identify all possible similarities between a single target and a mentation, but it is unlikely, and it is even more unlikely that a judge would be able to know and identify all possible similarities between all targets and all mentations.

This means that the set of similarities that a judge can identify is different for each target/mentation. The true set of similarties that the judge can identify is unknown, as is the total set of possible similarities. Therefore, the statement that there is an equal 25% chance that a given target will be chosen is incorrect. The true chance cannot be calculated.
[

Absolutely agree.


If, over many trials, the set of targets used tends toward things that the judge is more able to identify similarities for, the result of meta-analysis could be skewed toward significance. Likewise, if the set of targets tends toward things that the judge is less able to identify similarities for, the results could be skewed toward non-significance.


This is the important point. This could potentially happen only if the targets were inadequiately randomised. If certain targets appeared more often than others and these targets conformed to the judges subjective bias then you might be right. However, the autoganzfeld studies included checks to safeguard against this. Each target appeared, on average, an equal number of times.

From "Does Psi exist" by Bem and Honorton:
--------------------------------------------------------------------
The random selection of the target and sequencing of the judging set were controlled by a noise-based random number generator interfaced to the computer. Extensive testing confirmed that the generator was providing a uniform distribution of values throughout the full target range (1-160). Tests on the actual frequencies observed during the experiments confirmed that targets were, on average, selected uniformly from among the 4 clips within each judging set and that the 4 judging sequences used were uniformly distributed across sessions. "
---------------------------------------------------------------------



As to the question of "fuzzy logic": this is essentially how a judge would identify similarities between seemingly unrelated targets/mentations.


So its another phrase for the judges subjective bias ?


However, because the experiment requires "fuzzy logic", or subjective choices by the judge, the potential for finding a psi signal is also rendered meaningless, because the chances for making a target choice are unequal, as explained above. One might argue that over many trials, the inequalities would tend to balance out, but that assumption would be unwarranted.

Thats a good point, but not a flaw issue. Also the potential for finding a psi signal is not meaningless, just possibly reduced. You are talking about the judges pre-disposition to particular targets acting to mask any possible psi effects about other targets. If you assume such masking occurs on a significant level then you would expect the hit rate to reduce to 25%. What can't happen is for the judges pre-disposition to increase the hit rate because of the randomisation of the targets.

[Interesting Ian]

Originally posted by Pyrrho
[B]Here's the flaw, as I see it:

For each target there is a set of possible similarities.



To the 4 videos or images that the receiver or judge views?



The judge (who can also be the receiver, according to what I've read, but we'll assume this is an independent judge) is potentially capable of identifying any given similarity in that set.



Remember what the skeptics such as yourself are stating. You're stating that the chance of correcting guessing the correct video or image will be greater than 25% in the absence of psi, and in the absence of any sensory leakage.

Bearing this in mind I have to ask what earth you are talking about? The receiver, lacking psi, cannot identify any similarities, therefore neither can the judge! You seem to be concentrating on problems assuming that psi does exist, therefore your argument is irrelevant.

Or am I just not understanding what you're saying??

If in the absence of any sensory leakage or psi, and providing the original target is wholly randomly selected, then if over a sufficiently large number of trials the chance of choosing the correct target is over 25%, then the receiver and/or judge must have access to additional information. Yes??

So what form does this additional information take??

[Pyrrho]

I don't see that randomization solves the problem of hypothetical limited-set of recognized similarities. Perhaps it does; I don't know, but for some reason I doubt it. For any given set of four targets, one cannot say that the chance is equal for each target. Randomization is irrelevant.

It is a given that the judge brings subjective bias to the process. This is part of human nature.

I believe I explained the difference between pure logic and fuzzy logic in recognizing similarities, which is not the same as subjective bias.

Subjective bias would be a problem if the judge scored something a certain way because the judge preferred to do so, not because the judge missed a certain set of similarities which would otherwise alter the score.

[Pyrrho]

Originally posted by Interesting Ian


Pyrrho: For each target there is a set of possible similarities.

To the 4 videos or images that the receiver or judge views?

Is it videos or static images? I admit that I'm a bit confused here -- I had thought that the sender was viewing static images.

As I understand the process, for each image viewed by the sender there is a set of possible similarities to each image viewed by the judge, who chooses a possible target based on the judge's rating of the mentation given by the receiver. There is a set of possible similarities for each images. As an illustration, consider the possible set of synonyms for a given word. Each word has a set of possible synonyms, all or some of which may be known by the judge. It is unlikely that the judge will be able to remember or identify each possible synonym in the set.

Remember what the skeptics such as yourself are stating. You're stating that the chance of correcting guessing the correct video or image will be greater than 25% in the absence of psi, and in the absence of any sensory leakage.

Bearing this in mind I have to ask what earth you are talking about? The receiver, lacking psi, cannot identify any similarities, therefore neither can the judge! You seem to be concentrating on problems assuming that psi does exist, therefore your argument is irrelevant.

I'm saying that the chance is unequal, because the judge's ability to recognize possible similarities is limited, and is limited differently for each of the four images. The argument for psi as detected by Ganzfeld studies seems to depend upon the chance being equal for all targets, but based on what I understand the process to be, this cannot be true. I am not saying that the correct image has a greater chance; in fact, depending on the judge, it may have a greater, lesser, or equal chance. My point is that the true chance is not knowable, or, at best, is knowable only within a certain deviation of error -- which is itself not knowable.

This has nothing to do with psi -- it is a problem inherent in the experimental design, as I understand it.

Or am I just not understanding what you're saying??

If in the absence of any sensory leakage or psi, and providing the original target is wholly randomly selected, then if over a sufficiently large number of trials the chance of choosing the correct target is over 25%, then the receiver and/or judge must have access to additional information. Yes??

So what form does this additional information take??
No additional information is required. If the total set of similarities for each of the four images is equal, the judge may not be able to identify all possible similarities for a given image. The judge may then assign a rating that is essentially incorrect. If anything, it is a case of less information being available to the judge. If the chance is indeed equal, then the test I proposed earlier should show identical results each time.

[BillHoyt]

Originally posted by Interesting Ian


The fact that the targets are dynamic rather than static matters not one iota to my argument. If dynamic targets get greater hits this is suggestive that psi is responsible, especially as this would be what we might expect from our experience of psi anyway.



How so?



I think that the judge must put the images, videos or whatever in a definite order. If he chooses the wrong one it's a miss, it's as simple as that.

Sorry, you're going to have to do better than that.
You didn't get my drift here. Ganzfeld collapses a universe of possible answers into four choices. The maps from all possible answers to the four allowable are not equal: for some of the targets, these maps are larger than for other targets. The probabilities are distorted in favor of the more dynamic targets. This has already been demonstrated with ganzfeld studies: more dynamic targets have higher hit rates. The easiest correction for this fatal flaw is to give the option "utter miss". Curious that this is not allowed even though the results demonstrate this skewed probability.

[davidsmith73]

Originally posted by Pyrrho
I don't see that randomization solves the problem of hypothetical limited-set of recognized similarities. Perhaps it does; I don't know, but for some reason I doubt it. For any given set of four targets, one cannot say that the chance is equal for each target. Randomization is irrelevant.

It is a given that the judge brings subjective bias to the process. This is part of human nature.

I believe I explained the difference between pure logic and fuzzy logic in recognizing similarities, which is not the same as subjective bias.

Subjective bias would be a problem if the judge scored something a certain way because the judge preferred to do so, not because the judge missed a certain set of similarities which would otherwise alter the score.

Ok I see the difference between subjective bias and fuzzy logic I think. However, its certain that the outcome of both subjective bias and fuzzy logic are the same here. Namely that they both result in an unequal probability of picking each image from the four.

I'll try to give an example of how I think randomisation nullifies this.

Lets say the target pool consists of just four images. One is of naked women and the other three are of non-sexual nature. Lets say that we have 100 judges at our disposal and each judge only takes part in one experiment. Lets also say that each judge has a predisposition towards finding similarities between mentation and sexual imagery. Perhaps they've read a bit too much Freud.

Now, if the naked women image is chosen as the target in every session then the judges will consistnetly pick it not because of psi but because of their predisposition. A high hit rate will result.

However, if the naked women target is randomised so that it has a 1 in 4 chance of being the correct target, the judges will still pick that image but they will only get it right 25% of the time.

I hope that explains it.

[davidsmith73]

Originally posted by BillHoyt

You didn't get my drift here. Ganzfeld collapses a universe of possible answers into four choices. The maps from all possible answers to the four allowable are not equal: for some of the targets, these maps are larger than for other targets. The probabilities are distorted in favor of the more dynamic targets. This has already been demonstrated with ganzfeld studies: more dynamic targets have higher hit rates. The easiest correction for this fatal flaw is to give the option "utter miss". Curious that this is not allowed even though the results demonstrate this skewed probability.

As I said, if each image, dynamic or static, has a 1 in 4 probability of being chosen as the target throughout the entire experiment then this point does not matter.

[Pyrrho]

Originally posted by davidsmith73


Ok I see the difference between subjective bias and fuzzy logic I think. However, its certain that the outcome of both subjective bias and fuzzy logic are the same here. Namely that they both result in an unequal probability of picking each image from the four.

I'll try to give an example of how I think randomisation nullifies this.

Lets say the target pool consists of just four images. One is of naked women and the other three are of non-sexual nature. Lets say that we have 100 judges at our disposal and each judge only takes part in one experiment. Lets also say that each judge has a predisposition towards finding similarities between mentation and sexual imagery. Perhaps they've read a bit too much Freud.

Now, if the naked women image is chosen as the target in every session then the judges will consistnetly pick it not because of psi but because of their predisposition. A high hit rate will result.

However, if the naked women target is randomised so that it has a 1 in 4 chance of being the correct target, the judges will still pick that image but they will only get it right 25% of the time.

I hope that explains it.
Strictly speaking, I am not talking about predisposition in terms of the judge favoring certain images above others; I am talking about the judge's ability to identify all possible similarities. The difference is subtle, but there is a difference between that and predisposition.

Hypothetically, let's say there are 1000 possible similarities the judge could identify for each of the four targets, but the judge is only capable of identifying 900 for one, 800 for another, all 1000 for the third, and 800 for the fourth. The chances are therefore unequal. Next week, given the same set of targets and the same mentation, the judge is able to identify 950, 900, 1000, and 800. The chances change. Some of this may be due to cultural effects, some of it due to the judge's education, or other factors. This is not predisposition, this is limitations of human memory and experience. Randomization will not affect this other than altering the set of targets.

I think there are enough unknown factors in the procedure to nullify legitimate detection of any psi signal.

[davidsmith73]

Originally posted by Pyrrho

Strictly speaking, I am not talking about predisposition in terms of the judge favoring certain images above others; I am talking about the judge's ability to identify all possible similarities. The difference is subtle, but there is a difference between that and predisposition.

Hypothetically, let's say there are 1000 possible similarities the judge could identify for each of the four targets, but the judge is only capable of identifying 900 for one, 800 for another, all 1000 for the third, and 800 for the fourth. The chances are therefore unequal. Next week, given the same set of targets and the same mentation, the judge is able to identify 950, 900, 1000, and 800. The chances change. Some of this may be due to cultural effects, some of it due to the judge's education, or other factors. This is not predisposition, this is limitations of human memory and experience. Randomization will not affect this other than altering the set of targets.

I think there are enough unknown factors in the procedure to nullify legitimate detection of any psi signal.

But assuming there is no psi signal these differences would not be expected to produce a hit rate greater than 25% over the long run yes ?

[Interesting Ian]

Originally posted by davidsmith73


But assuming there is no psi signal these differences would not be expected to produce a hit rate greater than 25% over the long run yes ?

This is what I can't understand because a lot of people are claiming that it would be higher than 25% despite proper randomisation procedures and no sensory leakage, which is really bizarre and I absolutely cannot understand at all.

Of course if psi DOES exist I understand there are many complexities in determining how large the effect is. But this is irrelevant if one is arguing that psi doesn't exist at all.

[davidsmith73]

Originally posted by Interesting Ian


This is what I can't understand because a lot of people are claiming that it would be higher than 25% despite proper randomisation procedures and no sensory leakage, which is really bizarre and I absolutely cannot understand at all.

Of course if psi DOES exist I understand there are many complexities in determining how large the effect is. But this is irrelevant if one is arguing that psi doesn't exist at all.

Which brings us right back to the start of this debate about ganzfeld. Why isn't it acceptable as evidence for the paranormal in the eyes of the JREF challenge if there is no subjective "wiggle room" as we have established ?

[Ipecac]

Originally posted by davidsmith73
Ok I see the difference between subjective bias and fuzzy logic I think. However, its certain that the outcome of both subjective bias and fuzzy logic are the same here. Namely that they both result in an unequal probability of picking each image from the four.

I'll try to give an example of how I think randomisation nullifies this.

Lets say the target pool consists of just four images. One is of naked women and the other three are of non-sexual nature. Lets say that we have 100 judges at our disposal and each judge only takes part in one experiment. Lets also say that each judge has a predisposition towards finding similarities between mentation and sexual imagery. Perhaps they've read a bit too much Freud.

Now, if the naked women image is chosen as the target in every session then the judges will consistnetly pick it not because of psi but because of their predisposition. A high hit rate will result.

However, if the naked women target is randomised so that it has a 1 in 4 chance of being the correct target, the judges will still pick that image but they will only get it right 25% of the time.

I hope that explains it.

It doesn't explain it to me. If the judge is predisposed to pick a certain image or images, then how are you measuring psi at all? You're measuring the judge's disposition. Is the mentation supposed to be influencing the judge in spite of his predispositions? This makes no sense to me.

[ceptimus]

Right. Let's test the idea. Now who would have thought of that?

Here are four images, and some of them have (I think) more emotional appeal than others.

Image 1: http://www.blogjam.pwp.blueyonder.co.uk/kittens/0137.jpg

Image 2: http://www.steam-in-south-africa.com/siberia/Peterhof4.jpg

Image 3: http://www.masterplumbers.com/plumbviews/images/potty.jpg

Image 4: http://www.cypresssoaring.org/GFX/90AUG-%7E1.JPG

I have used a random number generator to select one of the above four images, and I have done this one hundred times and recorded the results.

The challenge is now for you to make your selections, and PM them to me. If you can't be bothered to do all one hundred trials, you can just send 20 or 50 choices - whatever you like. I'll wait a day or two so everyone who wants to has had a chance to respond, and then I'll post the results (if any).

Note: You can use any method you wish to make your selections - you can assign weightings of 1 to 99, or talk it over with a judge, wear a blindfold if you wish - whatever. All I'm interested in is your choices, which you can PM like this example:


trial choice
1 2
2 1
3 2
4 4
5 2
6 3
. .
. .
etc. etc.
If I understand Pyrrho, BillHoyt, thaiboxerken, etc. correctly, they would expect to get 'hits' at a better than 25%, as some of the pictures are more appealing than others. I would only expect 25% hits regardless. If I've misunderstood what Pyrrho, Ken and Bill are saying, maybe this example will allow them to help me understand.

[davidsmith73]

[QUOTE]Originally posted by Ipecac


It doesn't explain it to me. If the judge is predisposed to pick a certain image or images, then how are you measuring psi at all? You're measuring the judge's disposition. Is the mentation supposed to be influencing the judge in spite of his predispositions? This makes no sense to me. [/QUOTE

The presence of a judge’s subjective bias should still result in an overall hit rate of 25% if no psi is operating. If their disposition is as strong as I assumed in the example I gave then I could imagine that a psi signal would have a hard time overcoming it. But I gave that example as a deliberate exageration to prove a point. I don’t think a judge in reality would be that pre-disposed.
]

[BNiles]

Originally posted by Pyrrho

Strictly speaking, I am not talking about predisposition in terms of the judge favoring certain images above others; I am talking about the judge's ability to identify all possible similarities. The difference is subtle, but there is a difference between that and predisposition.

Hypothetically, let's say there are 1000 possible similarities the judge could identify for each of the four targets, but the judge is only capable of identifying 900 for one, 800 for another, all 1000 for the third, and 800 for the fourth. The chances are therefore unequal. Next week, given the same set of targets and the same mentation, the judge is able to identify 950, 900, 1000, and 800. The chances change. Some of this may be due to cultural effects, some of it due to the judge's education, or other factors. This is not predisposition, this is limitations of human memory and experience. Randomization will not affect this other than altering the set of targets.

I think there are enough unknown factors in the procedure to nullify legitimate detection of any psi signal.

Pyrrho,
I understand what you're getting at here, and agree with you. A judges understanding of the target will have a big effect on his rating. I also see where Ian is coming from when he says that this is only a problem if Psi exists.

I think what needs to be clarified is that this poses a problem whether Psi exists or not.

If a target is randomly selected, and it happens to be something that the judge is very familiar with, and the sender is a fraud (no Psi) he will still grant a higher rating on responses.

If a target is randomly selected, and it happens to be something that the judge is not familiar with, and the sender is not a fraud (Psi exists) he will still grant a lower rating on responses.

One way around this may be this:
All known similarities of the targets are generated and agreed upon by a large group and then entered into a database before experiments take place. Targets are selected randomly and shown only to the sender. Receiver then writes down descriptions. Descriptions are entered into computer that compares against database of known similarities and generates a score.
This removes any subjective bias while still allowing for fuzzy logic. It also removes a judge’s personal bias and understanding of a target because the larger the group that generates & agrees on the similarities the more accurate the scoring can be.

[BillHoyt]

Originally posted by ceptimus
If I understand Pyrrho, BillHoyt, thaiboxerken, etc. correctly, they would expect to get 'hits' at a better than 25%, as some of the pictures are more appealing than others. I would only expect 25% hits regardless. If I've misunderstood what Pyrrho, Ken and Bill are saying, maybe this example will allow them to help me understand.

It has nothing to do with appeal, and everything to do with "richness". But it first starts with fallacious probability reasoning of the sort that some Lotto players use. Let's look at this first.

Ask yourself: is it better to play the same number over and over or better to pick numbers at random? Some people's instincts tell them to stay put because that way they don't have two different probabilities multiplying together and reducing their chances of hitting Lotto. In fact, either way you play, your chances are the same.

Now how can that be? Isn't your random number picking a probability generator and the Lotto random number picking also a probability generator? Shouldn't you multiply those two probabilities together and get a horribly small probability of hitting the Lotto? Doesn't that mean you should pick a number and stick with it?

This is a mistaken analysis that misses the essential point: the target is the target. It simply is. There is no probability to the target Lotto number. Today it is 4387. Period. The only probability is whether you bought a ticket for 4387.

Now if this clicks, you begin to see how horridly flawed ganzfeld is. The ganzfeld target has no probability. It is what it is. The probability shifts to the chances that the judge matches the mentation to the target. The target "richness" skews this horribly. If one of the possible targets has many more matching possibilites, this raises its chances of being picked by the judge.

The mistake in saying the ganzfeld probability is .25 is the same as the Lotto mistake of assuming you improve your chances by sticking with the same number. The Lotto target is chosen and has a probability = 1.0. The remaining probability lies in matching your Lotto pick to the Lotto target. Likewise, the ganzfeld target has a probability = 1.0. The remaining probability lies in matching the mentation to the target. Richer targets have higher probabilities of being matched, and therefore skew the results.

[Interesting Ian]

Originally posted by BillHoyt


It has nothing to do with appeal, and everything to do with "richness". But it first starts with fallacious probability reasoning of the sort that some Lotto players use. Let's look at this first.

Ask yourself: is it better to play the same number over and over or better to pick numbers at random?



Assuming the original numbers was also picked randomly then in terms of probability the chances will be the same whether one changes or not (DUH!) as will the size of the payout. Of course if the original nubers had not been random, but were say 11, 31,32,33,34,46, then although the chances of getting all six numbers correct would be the same, the original numbers are liable to have a bigger payout.

That's the sort of numbers I used to pick before Camelot introduced that bloody lucky dip! :mad: So I never go on the national lottery now.




Some people's instincts tell them to stay put because that way they don't have two different probabilities multiplying together and reducing their chances of hitting Lotto.



I have absolutely no idea what you're talking about. What 2 different probabilities?? :confused:



In fact, either way you play, your chances are the same.



How could anyone conceivably conclude otherwise??



Now how can that be?



What do you mean?? How could it conceivably not be??



Isn't your random number picking a probability generator and the Lotto random number picking also a probability generator?



Ah! I see what you're getting at. But what's this got to do with the chance of picking the correct target magically being more than 25%??



Shouldn't you multiply those two probabilities together and get a horribly small probability of hitting the Lotto? Doesn't that mean you should pick a number and stick with it?



Oh don't be so absurd! :rolleyes:



This is a mistaken analysis that misses the essential point: the target is the target. It simply is. There is no probability to the target Lotto number. Today it is 4387. Period. The only probability is whether you bought a ticket for 4387.

Now if this clicks, you begin to see how horridly flawed ganzfeld is.



Er think this "lotto" is different from the lottery I'm familiar with.

Anyway, No it doesn't. I am wholly unable to see how any intelligent person could conclude it's flawed.



The ganzfeld target has no probability. It is what it is. The probability shifts to the chances that the judge matches the mentation to the target. The target "richness" skews this horribly.



But which cannot conceivably make any difference whatsoever since the original target was selected wholly randomly.



If one of the possible targets has many more matching possibilites, this raises its chances of being picked by the judge.



Perhaps so, but this will be precisely balanced out by those targets which have many fewer matching possibilities.



The mistake in saying the ganzfeld probability is .25 is the same as the Lotto mistake of assuming you improve your chances by sticking with the same number.



I sincerely doubt you could point to anyone who actually believes this. And I cannot understand any reasoning which would make anyone conclude so. And I am wholly unable to see how this relates to the ganzfeld experiments.



The Lotto target is chosen and has a probability = 1.0. The remaining probability lies in matching your Lotto pick to the Lotto target.



Huh?? WTF???



Likewise, the ganzfeld target has a probability = 1.0.



Er . . no . . it's 25% or 1/4.




The remaining probability lies in matching the mentation to the target. Richer targets have higher probabilities of being matched, and therefore skew the results. [/B]

Perhaps so, but if so less detailed images have lower probabilities of being matched, and therefore skew the results in the precise opposite direction. This results in a chance of precisely 25% over a sufficiently large number of trials.

[ceptimus]

Originally posted by BillHoyt
It has nothing to do with appeal, and everything to do with "richness".OK, but some of the images I posted are also richer than others? No? I still don't understand your position - would you expect people to score better than 25% hits in my test?

I believe I understand your Lotto reasoning, but I don't see how that applies here. The chances of scoring hits in my test are 25%? Yes or No? Do you intend to PM an entry to my test?

[thaiboxerken]

Originally posted by ceptimus
Right. Let's test the idea. Now who would have thought of that?

Here are four images, and some of them have (I think) more emotional appeal than others.
I have used a random number generator to select one of the above four images, and I have done this one hundred times and recorded the results.

The challenge is now for you to make your selections, and PM them to me. If you can't be bothered to do all one hundred trials, you can just send 20 or 50 choices - whatever you like. I'll wait a day or two so everyone who wants to has had a chance to respond, and then I'll post the results (if any).


Ok, I'll do it the Ganzfield way. I'm getting a mentation of a Big Mac. Which of the 4 pictures describes the mentation I recieved? Now the judge is gonna pick the picture that best fits my mentation. Can you not see the BS behind this test now?

[davidsmith73]

Originally posted by BillHoyt

It has nothing to do with appeal, and everything to do with "richness". But it first starts with fallacious probability reasoning of the sort that some Lotto players use. Let's look at this first.

Ask yourself: is it better to play the same number over and over or better to pick numbers at random? Some people's instincts tell them to stay put because that way they don't have two different probabilities multiplying together and reducing their chances of hitting Lotto. In fact, either way you play, your chances are the same.

Now how can that be? Isn't your random number picking a probability generator and the Lotto random number picking also a probability generator? Shouldn't you multiply those two probabilities together and get a horribly small probability of hitting the Lotto? Doesn't that mean you should pick a number and stick with it?

This is a mistaken analysis that misses the essential point: the target is the target. It simply is. There is no probability to the target Lotto number. Today it is 4387. Period. The only probability is whether you bought a ticket for 4387.




The lotto number has a certain probability of being spat out by the machine. Each lotto number has an equal probability of being spat out. Likewise, each ganzfeld image has a certain probability of being chosen as the target by the random number generator that selects them. On average, each image will be selected as the target by the random number generator an equal number of times.

So just as the lotto player does not fare any better by sticking with his numbers or using any other strategy, the ganzfeld judge does not fare any better by going with his predisposition or application of any kind of fuzzy logic simply because each ganzfeld image has an equal chance of being spat out by the random generation process, just like the lotto situation.



Now if this clicks, you begin to see how horridly flawed ganzfeld is. The ganzfeld target has no probability. It is what it is. The probability shifts to the chances that the judge matches the mentation to the target. The target "richness" skews this horribly. If one of the possible targets has many more matching possibilites, this raises its chances of being picked by the judge.


Yes but this is a single trial. If the image that was randomly picked as the target has many more matching fuzzy logic similarities or whatever you want to call it, then yes it will be picked.

However, that image has the same probability of being chosen as the target as all the other images. Over many trials this subjective bias will be nullified because it would not be possible even for a changing fuzzy logic/subjective bias to be correlated with the random target selection.

Does anyone know of a simulation of this type ?

[ceptimus]

Originally posted by thaiboxerken


Ok, I'll do it the Ganzfield way. I'm getting a mentation of a Big Mac. Which of the 4 pictures describes the mentation I recieved? Now the judge is gonna pick the picture that best fits my mentation. Can you not see the BS behind this test now? I suppose you'd pick the potty for the Big Mac? I don't know, and I don't care. I think the BS is that you believe you can expect to score significant deviation from 25% hits. Are you going to PM me an entry to my test? I'll be happy to tell you, and everyone else how well you score.

Surely this simple test is a way for you and others to prove your theory?

No one has entered yet. If that's a clue then you are welcome to it.

[BillHoyt]

Two essential questions:

1. Why is the Lotto random generator no different from sticking with one number?

2. How do you calculate the probability of the ganzfeld target? (Why is it not .125?)

[edited to clarify question 2]

[davidsmith73]

Originally posted by BillHoyt
Two essential questions:

1. Why is the Lotto random generator no different from sticking with one number?


2. How do you calculate the probability of the ganzfeld target?

Don't really understnad the relevance of the first question. The lotto random generator is a machine that picks a number at random. Sticking with one number is not random.

second question I don't understand either. The probablility of the random event generator picking a target or the judge chosing a target ?

[BillHoyt]

Originally posted by davidsmith73


Don't really understnad the relevance of the first question. The lotto random generator is a machine that picks a number at random. Sticking with one number is not random.

That's right. You just identified differences between the two. And yet the probabilities of winning Lotto are the same whether one sticks with one number or uses a random generator. What's wrong here?

[Ipecac]

Originally posted by ceptimus
I suppose you'd pick the potty for the Big Mac? I don't know, and I don't care. I think the BS is that you believe you can expect to score significant deviation from 25% hits. Are you going to PM me an entry to my test? I'll be happy to tell you, and everyone else how well you score.

Surely this simple test is a way for you and others to prove your theory?

No one has entered yet. If that's a clue then you are welcome to it.

While I appreciate your posting the pictures and setting up the random numbers, aren't we missing a sender? Isn't someone else supposed to pick the image and then the judge sees if they can get a mental image of what the tester picked?

[Interesting Ian]

Originally posted by Ipecac


While I appreciate your posting the pictures and setting up the random numbers, aren't we missing a sender? Isn't someone else supposed to pick the image and then the judge sees if they can get a mental image of what the tester picked?

We don't want any esp involved do we? We're assuming that esp doesn't exist and seeing if any of the targets get more than 25% to a significant statistical extent.

[BillHoyt]

Originally posted by ceptimus
I believe I understand your Lotto reasoning, but I don't see how that applies here. The chances of scoring hits in my test are 25%? Yes or No? Do you intend to PM an entry to my test?
In general, 25% would be expected. But the appeal of a picture could skew the choices. More importantly, this has nothing to do with the ganzfeld problem. The ganzfeld problem is "richness". That is, the probability of finding more "mentations" that can be mapped to a particular target. This is very different from simply picking a picture.

[davidsmith73]

Originally posted by BillHoyt


That's right. You just identified differences between the two. And yet the probabilities of winning Lotto are the same whether one sticks with one number or uses a random generator. What's wrong here?

I don't understand. Whats your point :confused:

[davidsmith73]

Originally posted by BillHoyt

In general, 25% would be expected. But the appeal of a picture could skew the choices. More importantly, this has nothing to do with the ganzfeld problem. The ganzfeld problem is "richness". That is, the probability of finding more "mentations" that can be mapped to a particular target. This is very different from simply picking a picture.


Not really. The outcome is the same in both situations. A picture is chosen. Whether the choice is due to the appeal of the picture or their similarity richness does not matter. As long as the targets are randomly selected from the whole pool of images and each target is chosen on average an equal number of times throughout the whole experiement, the mechanism by which the image is chosen by the judge is irrelevant. All methods by which the target is chosen by the judge will result in chance hit rate over a sufficient number of trials. Unless psi is operating ;)

[thaiboxerken]

Originally posted by ceptimus
I suppose you'd pick the potty for the Big Mac? I don't know, and I don't care. I think the BS is that you believe you can expect to score significant deviation from 25% hits. Are you going to PM me an entry to my test? I'll be happy to tell you, and everyone else how well you score.

Surely this simple test is a way for you and others to prove your theory?

No one has entered yet. If that's a clue then you are welcome to it.

Oh, but let me do it the Ganzfield way. First you pick what you think I was mentating. Then tell me the target. Then I'll tell you that I was really thinking about the target.

[ceptimus]

Originally posted by thaiboxerken
Oh, but let me do it the Ganzfield way. First you pick what you think I was mentating. Then tell me the target. Then I'll tell you that I was really thinking about the target. You're (I assume deliberately) missing the point.

You and others have said that that because of the 'appeal' or 'richness' of certain images, the test gives skewed, and therefore invalid results. You think that better than 25% hits are to be expected, even given the total absence of PSI, ESP or whatever. A few posters here, myself included, think that is BS.

So to test your theory, I have issued a challenge. I assume that posters to the JREF forums understand the nature of a challenge to test whether a claim is BS? I'm not offering a $1 million prize or anything, just the chance for you to prove that you are right.

I'm still waiting for your entry.

[thaiboxerken]

You're (I assume deliberately) missing the point.

You and others have said that that because of the 'appeal' or 'richness' of certain images, the test gives skewed, and therefore invalid results. You think that better than 25% hits are to be expected, even given the total absence of PSI, ESP or whatever. A few posters here, myself included, think that is BS.

I have not made any such arguement. I find the protocols BS because they depend on subjectivity of the judge, the experimenter and the sitters... so much subjectivity that there is no objectivity.


So to test your theory, I have issued a challenge. I assume that posters to the JREF forums understand the nature of a challenge to test whether a claim is BS?

Your challenge is invalid, as it's not even close to being a replica of Ganzfield protocol or procedures. I simply gave you one example of a mentation as it would occur in a Ganzfield experiment, and you didn't like that answer.

The Ganzfield experiments have been found wanting. Scrutiny by the scientific community have found flaws in the early protocols and insignificance in the recent studies. In other words, there isn't any conclusive data produced by the Ganzfield experiments.

[Pyrrho]

Originally posted by davidsmith73
But assuming there is no psi signal these differences would not be expected to produce a hit rate greater than 25% over the long run yes ?
For the sake of discussion, I'll say yes, but the data is suspect, thus rendering the investigation moot.

[Pyrrho]

Originally posted by ceptimus
You're (I assume deliberately) missing the point.

You and others have said that that because of the 'appeal' or 'richness' of certain images, the test gives skewed, and therefore invalid results. You think that better than 25% hits are to be expected, even given the total absence of PSI, ESP or whatever. A few posters here, myself included, think that is BS.

So to test your theory, I have issued a challenge. I assume that posters to the JREF forums understand the nature of a challenge to test whether a claim is BS? I'm not offering a $1 million prize or anything, just the chance for you to prove that you are right.

I'm still waiting for your entry.
I'm willing to give it a shot. It'll have to be later this evening, as I can't spend time on it now.

I'd like to mention that I included a possible test of my hypothesis in one of my earlier posts.

[davidsmith73]

Originally posted by Pyrrho

For the sake of discussion, I'll say yes, but the data is suspect, thus rendering the investigation moot.

Which data and how is it suspect ? Do you mean suspect in ways other than we'vew been discussing ?

[ceptimus]

Originally posted by thaiboxerken
You're (I assume deliberately) missing the point.

You and others have said that that because of the 'appeal' or 'richness' of certain images, the test gives skewed, and therefore invalid results. You think that better than 25% hits are to be expected, even given the total absence of PSI, ESP or whatever. A few posters here, myself included, think that is BS.

I have not made any such arguement.
In that case, I apologise, I must have confused your posts with those of others.
I find the protocols BS because they depend on subjectivity of the judge, the experimenter and the sitters... so much subjectivity that there is no objectivity.


So to test your theory, I have issued a challenge. I assume that posters to the JREF forums understand the nature of a challenge to test whether a claim is BS?

Your challenge is invalid, as it's not even close to being a replica of Ganzfield protocol or procedures. I simply gave you one example of a mentation as it would occur in a Ganzfield experiment, and you didn't like that answer.

The Ganzfield experiments have been found wanting. Scrutiny by the scientific community have found flaws in the early protocols and insignificance in the recent studies. In other words, there isn't any conclusive data produced by the Ganzfield experiments. My argument is that any amount of subjectivity on the part of the receiver or judge could not change the expected hit rate from 25%. My challenge is for those who believe that more than 25% hits are to be expected, even in the total absence of PSI, ESP, whatever. It still sounds to me as though you are one of the believers of this theory.

I am quite willing to be proved wrong - it wouldn't be the first time. This is your chance to convince me.

For those too stupid to understand long arguments I will try to put it in a nutshell.

My position: No amount of 'subjectivity' or 'richness' or anything else could be expected to make the hit rate in the tests described here to deviate from 25%. If the hit rate deviates from 25%, then that is proof of either PSI, or cheating. (I include 'sensory leakage' and other such flaws in the test procedure as cheating).

If you believe the hit rates will deviate from 25% in the absence of PSI or cheating, you can try to convince me, should you wish, by taking part in my test.

I hope that's clear now. [sigh]

[davidsmith73]

Originally posted by Pyrrho

I'm willing to give it a shot. It'll have to be later this evening, as I can't spend time on it now.

I'd like to mention that I included a possible test of my hypothesis in one of my earlier posts.

I'll be eating my words if this test turns up a high hit rate :(
Don't worry though, I won't be rationalising the results by saying that psi was involved, as long as nobody puts themselves into a ganzfeld state ;)

edited to add:

If it does turn up a high hit rate does that mean we can claim the million ! :D

[BillHoyt]

david,

The lotto random generator is a machine that picks a number at random. Sticking with one number is not random.

You pointed out a difference in the two scenarios. One has two sources of probability operating and the other doesn't. Yet the probability of a win is the same. How can this be?

Now there are two probabilities operating in ganzfeld as well, and yet you cite only one as the probability of a "hit".

Please give me the reason for this. Not just your answer, please. Give me the probability argument. Let me illustrate it a bit more before you bite, however. With two more questions:

In a fair coin toss, what is the probability of getting "HH"? What is the probability of getting a second H if the first toss yielded an H?

[davidsmith73]

Originally posted by BillHoyt
david,



You pointed out a difference in the two scenarios. One has two sources of probability operating and the other doesn't. Yet the probability of a win is the same. How can this be?

Now there are two probabilities operating in ganzfeld as well, and yet you cite only one as the probability of a "hit".

Please give me the reason for this. Not just your answer, please. Give me the probability argument. Let me illustrate it a bit more before you bite, however. With two more questions:

In a fair coin toss, what is the probability of getting "HH"? What is the probability of getting a second H if the first toss yielded an H?

In answer to your first question the probability is 0.5 x 0.5 = 0.25
The second question is 0.5

I still have no idea what you are on about though. You're going to have to explain a bit more I'm afraid :confused:

[Pyrrho]

Originally posted by davidsmith73
Which data and how is it suspect ? Do you mean suspect in ways other than we'vew been discussing ?
Based on what I perceive to be a flaw in the judging/scoring process, as explained in my earlier posts. I think the data is suspect; I think that the chance for the judge choosing a potential target based on the judge's interpretation of a mentation is not 25% in all cases because the judge is not uniformly capable of identifying all possible similarities in a given mentation as they relate to the target objects. This has nothing to do with personal preferences, emotional responses to given images, sensory leakage, or cheating.

[thaiboxerken]

My argument is that any amount of subjectivity on the part of the receiver or judge could not change the expected hit rate from 25%. My challenge is for those who believe that more than 25% hits are to be expected, even in the total absence of PSI, ESP, whatever. It still sounds to me as though you are one of the believers of this theory.

I am quite willing to be proved wrong - it wouldn't be the first time. This is your chance to convince me.

The subjectivity does matter though because of the subjectivity of the judgement, they can and probably do go back and repick the targets after they figure out what the correct one is.

They pick one of 4 based on the judge's subjective interpretation of what the "mentation" means. It was wrong, but they see the correct target and decide that they were wrong in picking the first error target because they're subjective interpretation better fits the correct target. It's this kind of stuff that has drawn criticism in the early Ganzfield experiments. After the second-guess stuff has been taken out, the experiment results have become about 25%.

The nature and subjectivity of the protocol is such that it can justify such "second" guesses and exceptions to their own rules.

When a science experiment uses judgements such that Big Macs and Apples are close enough to being the same, something is wrong.

Peer review by the scientific community has found that the Ganzfield experiments as a whole have produced inconclusive data.

[BillHoyt]

Originally posted by davidsmith73


In answer to your first question the probability is 0.5 x 0.5 = 0.25
The second question is 0.5

I still have no idea what you are on about though. You're going to have to explain a bit more I'm afraid :confused:

In your first answer you show an analysis for two independent events happening together. p(A+B) = p(A)*p(B).

What formula did you use for the second answer?

[BillHoyt]

Hint: the first formula is a special case of the second formula.

[davidsmith73]

Originally posted by BillHoyt
Hint: the first formula is a special case of the second formula.

Bill, stop being so bloody cryptic and get to the point. What has all of this got to do with the ganzfeld protocol ?

[BillHoyt]

Bill, stop being so bloody cryptic and get to the point. What has all of this got to do with the ganzfeld protocol ?

It has everything to do with ganzfeld. I am getting to the point.

In the second case, where we already had a head, you used conditional probability to calculate the odds of getting a second head. In this case, the two coins are totally independent, and the probability is simply the same as that of getting a head. Period.

Conditional probability is the probability of A given that B has occurred:

p(A|B) = p(A) * p(B|A) / p(B)

For ganzfeld, two strange things apply: the universe of mentations is infinite, and the universe of targets is simply four. The mapping of that infinite universe is foisted upon the judge because he simply cannot say "none of those targets are a Big Mac". He must choose the apple, because that is closest. But we know there are "dynamic" targets that ganzfeld experiments have demonstrated to increase hit rates. Does this point to psi or does this point to an underlying problem?

Let us say the field of 4 targets has one such "dynamic". The object is so rich in possibilities that the judge has a .7 probability of mapping something to it. p(A) is the prob. the judge maps the mentation to the rich target. p(B) is the prob. the rich target is the randomly selected target.

p(A|B) = .7 * .25 / .25 = .7

Now if you reused the same targets over and over, and had a true random target choice, you would see the probabilities fall back to .25. But what if the target choices are a much larger pool from which we select 4 at at time. If we do this, without replacement, and if the targets have unequal "dynamism" or "richness", then I suspect we'll see artificially inflated hit rates that are purely a function of the mix of rich targets and the artificial requirement that a mentation be mapped to one of the choices.

[ceptimus]

So far, two people have entered my test.

Remember what they were trying to do: they were trying to score 'hits' by guessing integers from 1 to 4, which I then compared against my list of randomly generated numbers.

Obviously we would expect them to score one hit in four or 25%. However, there are some nutty people posting here, who maintain that hits at a different (better) rate can be obtained by the following method:

1. Assign pictures with different degrees of 'richness' to the 4 numbers.

2. Have one person guess things, call them out.

3. Have another person decide which image best matches the guess, and assign a 'score' to how well a picture matches the guess (1 to 99, or 0 to 100 etc.)

4. After a set period of time, total the scores, and chose the image with the highest score.

5. By this roundabout method you have arrived at your guess of the random number. :rolleyes:

Anyway, as I was saying, two people have entered, so far.

One person scored 5 hits out of 30 guesses. This is worse than expected by chance, (7.5 hits).

A second person did much better, scoring 30 hits out of 100 guesses. This is a fairly significant result, and would only be expected to occur in 3.39% of 100-guess trials.

I've not named the people who entered here - they may not wish others to know who they are, but they can, of course, post here if they wish to identify themselves. If anyone else wishes to enter they still can.

I'll post the spreadsheet in a couple of days time if anyone wants it. Obviously, once the random numbers have been published, I'll not be able to accept any more entries. ;)

[Ossai]

Waiting on the numbers.

To generate the numbers I sent in, I used the random number generator in excel. Copy and paste.

Ossai

[ceptimus]

Originally posted by Ossai
Waiting on the numbers.

To generate the numbers I sent in, I used the random number generator in excel. Copy and paste.

Ossai I used Excel too. PM me if you want the spreadsheet emailing. In the table below, T is trial, R is Random, and the columns for davidsmith73 and yourself, show the guesses, and the cumulative number of hits.

Here you go:

T R davidsmith73 Ossai
1 4 1 0 1 0
2 1 3 0 2 0
3 3 1 0 1 0
4 4 2 0 2 0
5 2 4 0 1 0
6 2 4 0 2 1
7 4 4 1 2 1
8 4 1 1 2 1
9 3 2 1 2 1
10 1 4 1 2 1
11 2 1 1 3 1
12 1 1 2 2 1
13 2 3 2 3 1
14 1 4 2 2 1
15 1 4 2 1 2
16 3 1 2 3 3
17 3 2 2 3 4
18 3 4 2 1 4
19 1 4 2 1 5
20 2 4 2 3 5
21 2 2 3 4 5
22 1 1 4 1 6
23 1 4 4 3 6
24 2 1 4 1 6
25 2 4 4 2 7
26 4 3 4 3 7
27 1 4 4 1 8
28 4 4 5 2 8
29 2 1 5 3 8
30 3 4 5 2 8
31 2 2 9
32 2 3 9
33 4 3 9
34 1 1 10
35 3 3 11
36 1 4 11
37 1 1 12
38 2 4 12
39 1 2 12
40 2 4 12
41 3 2 12
42 4 3 12
43 1 1 13
44 3 3 14
45 1 3 14
46 3 4 14
47 1 1 15
48 3 3 16
49 3 4 16
50 3 3 17
51 2 4 17
52 4 1 17
53 3 3 18
54 3 2 18
55 4 1 18
56 2 2 19
57 3 3 20
58 1 2 20
59 4 2 20
60 4 2 20
61 2 2 21
62 4 4 22
63 1 1 23
64 3 2 23
65 1 3 23
66 2 3 23
67 4 4 24
68 1 3 24
69 2 3 24
70 1 4 24
71 1 2 24
72 3 1 24
73 3 4 24
74 4 2 24
75 2 1 24
76 4 4 25
77 4 2 25
78 3 2 25
79 4 2 25
80 1 4 25
81 2 2 26
82 2 2 27
83 4 3 27
84 4 3 27
85 3 2 27
86 4 3 27
87 2 2 28
88 2 1 28
89 2 1 28
90 4 2 28
91 3 3 29
92 4 2 29
93 2 3 29
94 3 2 29
95 4 3 29
96 1 3 29
97 1 1 30
98 4 3 30
99 4 3 30
100 1 2 30
The test is, of course, now closed. I thought I'd killed this thread already. :D

[Ossai]

Ceptimus
So now that random guesses have a better than average hit rate (25%), even without a dynamic target and only two respondents, do you understand why the Ganzfield test if flawed?

A hit should be a hit, either yes or no. By not allowing a miss (a choice is forced) you are disrupting the test from the beginning. For example if the correct answer had been 5 yet only 1 thru 4 were available an flaw is deliberately included which will invalidate the gathered data.

Ossai

[T'ai Chi]

Originally posted by Ossai
Waiting on the numbers.

To generate the numbers I sent in, I used the random number generator in excel. Copy and paste.

Ossai

Can you tell us what time you generated those numbers?

;)

[BillHoyt]

Originally posted by T'ai Chi


Can you tell us what time you generated those numbers?

;)

Oh, this should be rich...

[ceptimus]

Originally posted by Ossai
CeptimusYes? I'm here
So now that random guesses have a better than average hit rate (25%), That's not true. David did worse than expected by chance. You did better. But if we were to repeat this test many times, I would expect results as good as yours, in 3.39% of trials....even without a dynamic target and only two respondents, do you understand why the Ganzfield test if flawed?No. Sorry, I don't understand what you mean. :confused:A hit should be a hit, either yes or no. By not allowing a miss (a choice is forced) you are disrupting the test from the beginning. For example if the correct answer had been 5 yet only 1 thru 4 were available an flaw is deliberately included which will invalidate the gathered data.

Ossai I'm sorry, that just went over my head. Someone will have to explain it to me in terms I can comprehend.

[T'ai Chi]

Originally posted by BillHoyt


Oh, this should be rich...

Oh, this should be more gilb comments...

[Ossai]

T'ai Chi
Can you tell us what time you generated those numbers?
Not exactly, but if Ceptimus will look at when(day and time) he received the message and subtract approximately 2 minutes that should be close enough. :rolleyes:

Ossai

[BillHoyt]

Originally posted by Ossai
Ceptimus
So now that random guesses have a better than average hit rate (25%), even without a dynamic target and only two respondents, do you understand why the Ganzfield test if flawed?

A hit should be a hit, either yes or no. By not allowing a miss (a choice is forced) you are disrupting the test from the beginning. For example if the correct answer had been 5 yet only 1 thru 4 were available an flaw is deliberately included which will invalidate the gathered data.

Ossai

Ossai,

This "collapsing probability" effect is exacerbated by "dynamic" targets. By offering more opportunities to map "mentations" to targets, the dynamic targets give an additional shift to the probabilities.

[ceptimus]

Originally posted by Ossai
T'ai Chi

Not exactly, but if Ceptimus will look at when(day and time) he received the message and subtract approximately 2 minutes that should be close enough.I received Ossai's message on 08-29-2003 at 01:54 PM