The Ganzfeld Experiments

[Dancing David]

There are ways that a subject could pick up on the fact that she was only one of a very few. Anyway, none of this explains why a larger number of trials should get a higher hit rate, least of all stating it's very obvious. So obvious that no-one on here has been able to explain it yet Perhaps you'd like to have a shot?

I believe the theory is that the effect would be very small, say 3% and that therefore the larger the number of trials the more likely the effect is to show.
A larger number of trials also would reduce the chance that the effect is just a product of random distribution skewing the results, the larger the sample, the more likely any effect found is not a sampling error.
And so when larger trials are held they should firm up the effect and not lower it. Lowering it would show that the effect was possibly due to random chance over the trial.

To control for the specialness factor and the lenth of time factor would require sperate trials to see if this impacted the effect. Against a control group of course.

I can see myself that to have a really good result you would want to screen large numbers of people and then select the ones who give good results for a followup study.

 

[Paul C. Anagnostopoulos]

Ian said:
Huh?? It was arranged by the experimenter for them to cheat?? You make the assertions, therefore back them up. Your task not mine.

No, the time of the calls was arranged by the experimenter. And I never said they cheated.

You can't make assertions and not back them up. It's just not acceptable.

I don't care.

Don't be absurd. One can almost certainly find what they consider to be flaws if one is determined enough. This goes for all of science. What I'm interested in is your insinuation that parapsychological research contains more potential flaws than other areas of research. You need to justify this allegation, or apologise.

I never said they do. But when your entire branch of science is based on a negative definition, flaws become the central thesis of the experiments. The experiments are nothing more than an attempt to reject the null hypothesis while controlling for all possible flaws. People just don't seem to grasp this.

Details please. I want evidence that those experiments were included in this ganzfeld meta-analysis. I sincerely hope you weren't lying about it Paul.

I never said they were. You asked for examples of experiments with flaws, so I gave you some.

~~ Paul

 

[amherst]

The original Honorton ganzfeld experiments did not involve a judge. The receiver rated the four photos shown to him after finishing up the ganzfeld.

Let's see, flaws in psi experiments. Well, the problem with not editing hints out of ganzfeld transcripts, which I mentioned above.

Hints? During the sending phase a completly target blind reciever describes the imagry he/she is seeing. This is recorded in a control room by an experimenter who is also completly blind to the target. After twenty minutes, the reciver is presented with four images on a computer screen, one of which is the one the sender was trying to transmit. Over a long series of runs, the reciever should guess the correct target 25% of the time, the average is 32.2%. And the same goes for an independent judge. What hints could be in the transcript except those provided by psi if everyone is blind to what the target is? This nonsense you keep spouting is becoming very tiring. You find these experiments important enough to post about, yet you seem to have an aversion for reading any of the articles.

The problem with new vs. used video tapes that occured in some other Ganzfeld experiments, along with the problem of tape rewind time.

After the Psychological Bulletin article came out in 93, Daryl Bem graciously gave some of his time to discuss the details of paper and other issues on sci.skeptic. This should answer your criticism:

"Not possible. First, the experimenter doesn't even know which of 4 videotape cassettes is in the player. Next, the segments are all equally long so rewind time is identical for all targets."
http://groups.google.com/groups?q=D...1ts01sINNplo@newsstand.cit.cornell.edu&rnum=1

The problem of subjective bias introduced by using people who know each other as sender and receiver.

What bias could occur? Everyone is blind to the target.

The sound from the video clips leaking into the receiver's headphones.

Bem again:
"This leakage is discussed in detail in the original Honorton et al.
(1990) report in the Journal of Parapsychology. In the actual
experiment, the white noice is very loud (about 68db so that the receiver
cannot hear his/her own voice). When the solder joint was
discovered--before the studies had been concluded. The potential problem
was assessed by turning off the white noise completely and seeing if any
soundtrack could be heard. A number of people put on the headphones and
were instructed to listen as carefully as possible for ANY noise. None of
them could hear anything. Next an amplifier was placed between the
soundtrack circuit and the headphones and turned to full gain; again the
white noise was turned off completely. Now a very dim background noise
could be heard. It was concluded that if nothing could be heard with the
white noise turned off completely, it was unlikely that anything could
have been detected in 68db of noise. A conclusion with which I concur.

But there is more. The leakage problem was fixed and the studies
continued. The results were then analyzed to see if performance declined
after the fix. Answer: No. In fact, performance improved. Moreover,
performance throughout was uncorrelated with the noise level used."

Statistical problems involving trials that are not independent.

You haven't got a clue as to what you're talking about Paul. Read the papers.


amherst

 

[thaiboxerken]

Hints? During the sending phase a completly target blind reciever describes the imagry he/she is seeing. This is recorded in a control room by an experimenter who is also completly blind to the target. After twenty minutes, the reciver is presented with four images on a computer screen, one of which is the one the sender was trying to transmit. Over a long series of runs, the reciever should guess the correct target 25% of the time, the average is 32.2%. And the same goes for an independent judge. What hints could be in the transcript except those provided by psi if everyone is blind to what the target is? This nonsense you keep spouting is becoming very tiring. You find these experiments important enough to post about, yet you seem to have an aversion for reading any of the articles.

The hard part to about this is that the reciever isn't guessing one of 4 targets at all, the reciever is simply describing something that they "see". It's the experimenter who is really doing the guessing with these protocols. The "hints" comes during the judgement phase, where the description has to be applied to one of 4 images, which could have nothing to do with the actual description the reciever gave. These ganzfield studies are complete BS with poor protocols.

If these people really have evidence of "psi", why aren't they funding their research with JREF money? They can easily apply for the challenge.

 

[Ersby]

Huh?? Transcript of the senders not receivers??

Sorry, I simply have no understanding of this. You must mean transcript of receivers, right?? But how can what the receiver say give any clue to the right target since she has no clue to the target either??

Sorry but I really have absolutely no idea what on earth you are talking about.

Yes, transcript of the senders. The experiemtns was by Schlitz and Gruber, "Transcontinental remote viewing". In the rejudging of the test, they said "in spite of the rigorous methods employed, a potential weakness existed in the study that involved inclusion of the agent's impressions of the target sites in the material given to judges" (the agent being the person who went to the targets to record his/her impresions, ie, the sender)

 

[Ersby]

So what about my point about individuls feeling more special if there are fewer of them. And why should the hit rate be greater with more subjects??

Are you suggesting the best psi experiments are those with a small number of individuals doing a small number of trials?

 

[Lucianarchy]

There are many labs, in many countries, doing many 'psi' effect type experiments. They have been doing this over many years. Even putting aside the phenomenaly huge wealth of personal testimony of , dreams, meditation sessions, altered states etc where the effect commonly appears, It is totally and completely irrational to suggest that each and every positive result in the lab is due to cheating or self delusion.

People seem to be quietly stepping over the above fact.

Amherst is right about people's somprehension skills in this thread. What has Dr Sheldrake's phone exp got to do with the Ganzfeld method??

 

[CFLarsen]

There are many labs, in many countries, doing many 'psi' effect type experiments. They have been doing this over many years.

And what are the results?

Even putting aside the phenomenaly huge wealth of personal testimony of , dreams, meditation sessions, altered states etc where the effect commonly appears, It is totally and completely irrational to suggest that each and every positive result in the lab is due to cheating or self delusion.

Why? Have you checked each and everyone of them?

People seem to be quietly stepping over the above fact.

It's not a fact, it's a claim. Please back it up with evidence.

Amherst is right about people's somprehension skills in this thread. What has Dr Sheldrake's phone exp got to do with the Ganzfeld method??

The Ganzfeld method can also be used to test RV.

 

[Zep]

Zep, you seem to miss the point. There are many labs, in many countries, doing many 'psi' effect type experiments. They have been doing this over many years. Even putting aside the phenomenaly huge wealth of personal testimony of , dreams, meditation sessions, altered states etc where the effect commonly appears, It is totally and completely irrational to suggest that each and every positive result in the lab is due to cheating or self delusion.

No, YOU have missed the point. At no time here have I, Zep, made any assertions, for or against the procedures, wherein PEAR gets the results of their own RV experiments. It is PEAR themselves who say the results are equivalent to chance, not me!

Wealth of personal testimony??? Anecdotes??? Are you serious? OK, I'll assume you are, and ask you what I asked Winston Wu, who also believed in the collective weight of anecdotes: How viable do YOU think would a particular phenomena be if it were seen and experienced by billions of people over tens of thousands of years, right up until today?

And by the way, there are NOT many labs in many countries doing "psi" experiments these days. May have been in the USSR 20 years ago, but not now. You would probably be really lucky to find a dozen reputable ones in total, and of these, PEAR are one of the premier facilities. Of course, there are lots of little back kitchens populated by dotty little people with tin-foil hats trying to create "thought rays" from potatoes and such. For example, Scoles, and Geller. You really want to count those as "psi researchers"?

 

[Interesting Ian]

The hard part to about this is that the reciever isn't guessing one of 4 targets at all, the reciever is simply describing something that they "see". It's the experimenter who is really doing the guessing with these protocols. The "hints" comes during the judgement phase, where the description has to be applied to one of 4 images, which could have nothing to do with the actual description the reciever gave. These ganzfield studies are complete BS with poor protocols.

I have absolutely no understanding of what you are saying. Please be specific. Where is the information coming which allows either receiver/judges to pick the target on average greater than 25%??

If these people really have evidence of "psi", why aren't they funding their research with JREF money? They can easily apply for the challenge. [/B]

They'll only be interested in properly scientifically conducted research.

 

[Interesting Ian]

Are you suggesting the best psi experiments are those with a small number of individuals doing a small number of trials?

I don't know if they are or not. What I want to know is why the opposite should be true. Dancing David mentioned it's not the hit rate that is being talked about but the likelihood. This would explain it. But it's not what you originally said.

 

[Zep]

Ian,

I haven't forgotten you, it's just that it's the other side of the world here and I sleep when you are awake, usually.

OK, here's a sample target picture that might be used for a ganzfield test.

Now consider the following list of words:

car
yellow
sign
red
road
door
wheel(s)
window
words
bricks
sunlight
shade
sloping
driving/drive
Europe
French
cover(ed)
park(ed)/(ing)

By scoring a "target hit" for ANY of these words appearing anywhere in the subject's spiel, it is easily possible to increase the likelihood of a higher ganzfield score overall. And allowing the subject to ramble on for a period of time increases the likelihood of at least one of these words appearing. In effect, if you wait long enough, a random guess of words will match sooner or later.

So there needs to be either a closer match to what the actual target scene is (eg. more words match), or there needs to be a grading of the scoring in some way, in order to obtain more accurate matching of subject spiel and target. In either case, a judgement will be required, UNLESS the target can be simplified so that it is either hit/miss without judgement OR there is the use of some numerical evaluation method.

PEAR went the numerical evaluation method - weighted scores of match criteria comparing subject and target. They even rescored it a few times independently. But lo and behold, chance results only.

By the way, what do you think the scoring might be using exactly the same words on this target image?

Or this one?

Even matching four or five words from the list, all these images are a "match" in a ganzfield test. And yet they are far from being the same image, don't you agree?

 

[amherst]

Ian,

I haven't forgotten you, it's just that it's the other side of the world here and I sleep when you are awake, usually.

OK, here's a sample target picture that might be used for a ganzfield test.

Now consider the following list of words:

car
yellow
sign
red
road
door
wheel(s)
window
words
bricks
sunlight
shade
sloping
driving/drive
Europe
French
cover(ed)
park(ed)/(ing)

By scoring a "target hit" for ANY of these words appearing anywhere in the subject's spiel, it is easily possible to increase the likelihood of a higher ganzfield score overall. And allowing the subject to ramble on for a period of time increases the likelihood of at least one of these words appearing. In effect, if you wait long enough, a random guess of words will match sooner or later.


So there needs to be either a closer match to what the actual target scene is (eg. more words match), or there needs to be a grading of the scoring in some way, in order to obtain more accurate matching of subject spiel and target. In either case, a judgement will be required, UNLESS the target can be simplified so that it is either hit/miss without judgement OR there is the use of some numerical evaluation method.

PEAR went the numerical evaluation method - weighted scores of match criteria comparing subject and target. They even rescored it a few times independently. But lo and behold, chance results only.

By the way, what do you think the scoring might be using exactly the same words on this target image?

Or this one?

Even matching four or five words from the list, all these images are a "match" in a ganzfield test. And yet they are far from being the same image, don't you agree?

That's why there is a 25% chance hit rate. If there is no psi involved a reciever is just as likely to mention images which are in a target as he is to mention images which are in a decoy. It is truly baffling that you don't understand this.

amherst

 

[Interesting Ian]

Originally posted by Interesting Ian
Huh?? Transcript of the senders not receivers??

Sorry, I simply have no understanding of this. You must mean transcript of receivers, right?? But how can what the receiver say give any clue to the right target since she has no clue to the target either??

Sorry but I really have absolutely no idea what on earth you are talking about.

--------------------------------------------------------------------------------



Yes, transcript of the senders.

Transcript of the senders??? Why on earth are the senders saying anything?? And given that they are saying anything why have the judges got access to what they're saying??? These are the ganzfeld experiments yes?? Paul was making out they were but then admitted he wasn't.

Look, I have absolutely no understanding of this at all. Is this mentioned in any of the 4 links that Amherst provided so I could read it up myself??

The experiemtns was by Schlitz and Gruber, "Transcontinental remote viewing". In the rejudging of the test, they said "in spite of the rigorous methods employed, a potential weakness existed in the study that involved inclusion of the agent's impressions of the target sites in the material given to judges" (the agent being the person who went to the targets to record his/her impresions, ie, the sender) [/B]

Who went to the targets??? But these are pictures and short video clips aren't they??? Please confirm this is the ganzfeld we are talking about included in the meta-analysis.

I have absolutely no understanding of why the Judges should know what the senders are saying. If they do I cannot begin to understand how the experiment could have any validity. It wouldn't be impressions, they would just say what they are seeing eg the Eiffel tower or whatever. And you sure as hell would not get glowing reports from prominent skeptics about how watertight the experimental protocol is!

Could you please provide me with a link so I have some clue what you're talking about.

 

[Interesting Ian]

Ian,

I haven't forgotten you, it's just that it's the other side of the world here and I sleep when you are awake, usually.

OK, here's a sample target picture that might be used for a ganzfield test.

Now consider the following list of words:

car
yellow
sign
red
road
door
wheel(s)
window
words
bricks
sunlight
shade
sloping
driving/drive
Europe
French
cover(ed)
park(ed)/(ing)

By scoring a "target hit" for ANY of these words appearing anywhere in the subject's spiel, it is easily possible to increase the likelihood of a higher ganzfield score overall. And allowing the subject to ramble on for a period of time increases the likelihood of at least one of these words appearing. In effect, if you wait long enough, a random guess of words will match sooner or later.

So there needs to be either a closer match to what the actual target scene is (eg. more words match), or there needs to be a grading of the scoring in some way, in order to obtain more accurate matching of subject spiel and target. In either case, a judgement will be required, UNLESS the target can be simplified so that it is either hit/miss without judgement OR there is the use of some numerical evaluation method.

PEAR went the numerical evaluation method - weighted scores of match criteria comparing subject and target. They even rescored it a few times independently. But lo and behold, chance results only.

By the way, what do you think the scoring might be using exactly the same words on this target image?

Or this one?

Even matching four or five words from the list, all these images are a "match" in a ganzfield test. And yet they are far from being the same image, don't you agree?

I've already addressed this in this very thread! I'll paste it in.

"This applies to all 4 stimuli. If we consider any one of the three control stimuli, the more the receiver rambles on, there will likewise be more chance of seeing similarities in the receivers impressions compared to what the control stimuli depict. Since the increase in the amount of similarities seen will, on average, be the same for each of the 4 stimuli, the chance of choosing the correct target must still only be 25%".

In the list of impressions given, and if these were 3 of the 4 stimuli and the 4th stimuli likewise fitted the impressions, then the judges are only going to have 25% of a chance in picking the target. So what is the problem???

 

[Interesting Ian]

I have absolutely no understanding of why the Judges should know what the senders are saying. If they do I cannot begin to understand how the experiment could have any validity. It wouldn't be impressions, they would just say what they are seeing eg the Eiffel tower or whatever. And you sure as hell would not get glowing reports from prominent skeptics about how watertight the experimental protocol is!

Could you please provide me with a link so I have some clue what you're talking about. [/B]

Also you would just get a 100% success rate. But it ain't got nothing to do with the paranormal!

 

[Zep]

Ian, have a crack at selecting which of my three pics is the one I selected first. Then tell me WHY you think it is the one I selected first.

 

[Interesting Ian]

Ian, have a crack at selecting which of my three pics is the one I selected first. Then tell me WHY you think it is the one I selected first.

Huh?? That you selected?? What do you mean??

 

[Ersby]

Transcript of the senders??? Why on earth are the senders saying anything?? And given that they are saying anything why have the judges got access to what they're saying??? These are the ganzfeld experiments yes?? Paul was making out they were but then admitted he wasn't.

Look, I have absolutely no understanding of this at all. Is this mentioned in any of the 4 links that Amherst provided so I could read it up myself??

Who went to the targets??? But these are pictures and short video clips aren't they??? Please confirm this is the ganzfeld we are talking about included in the meta-analysis.

I have absolutely no understanding of why the Judges should know what the senders are saying. If they do I cannot begin to understand how the experiment could have any validity. It wouldn't be impressions, they would just say what they are seeing eg the Eiffel tower or whatever. And you sure as hell would not get glowing reports from prominent skeptics about how watertight the experimental protocol is!

Could you please provide me with a link so I have some clue what you're talking about.

"The" ganzfeld experiments? The ganzfeld has been in use since 1974, so includes a much greater database than those included in the four links in amherst's first posts. And the ganzfeld isn't resctricted to a certain target type. Why wouldn't physical locations be an acceptable target?

And why shouldn't the sender say anything? Sometimes (as in the PRL tests) the sender is encouraged to verbalise their thoughts as they send their impressions to the target.

Here's the abstract of Shlitz's experiment: Schlitz, Marilyn; Gruber, Elmar. Transcontinental remote viewing. Journal of Parapsychology, 1980 Dec, v44 (n4):305-317. Abstract: Two experimenters carried out a long-distance remote-viewing experiment, with one of them, in Detroit, Michigan, acting as percipient and the other, in Rome, Italy, as the agent. From a pool of 40 geographical target locations in Rome, 10 were randomly chosen without replacement, and the agent visited them one at a time for 15 min on each of 10 consecutive days. The percipient, at the same time, recorded in words and sketches her impressions of the agent's location. Later, 5 independent judges received copies of these sketches, and the impressions translated into Italian. They visited the locations and judged the protocols with respect to their correspondence to the target sites. Analysis of the results by a direct-count-of-permutations method yielded a p of 4.7 * 10-super(-6 ) for judges' ratings and 5.8 * 10-super(-6) for rankings. The authors point out that free-response remote viewing may be a psi-conducive procedure, but that the results may also have been influenced by exceptionally high motivation on the part of the 2 experimenters.

Although (hilariously ) I got my experiments mixed up. The Schlitz experiment didn't have the temporal clues that I mentioned earlier: the fault was that the notes of the agent were in Italian, and were translated by someone who knew what the target was. Some consider this to be a possible source for a leak of information (refuted by Schlitz).

The temporal clues came from a Targ and Tart experiment as I recall. So many experiments, so many faults: no wonder I get them mixed up!

 

[Zep]

Huh?? That you selected?? What do you mean??

Tell me which picture out of the three that I sought out and found in the Internet first (they're not my pics). Tell me your reasons why you think I selected it first. Is that too simple a request?

[edit; piss-poor grammar]