Golfy, have you inquired at challenge@randi.org if a polygraph would be acceptable in your proposed protocol and what was the response?
Assistance required for telepathy proof
- Thread starter golfy
- Start date
-
- Tags
- polygraphs telepathy
IXP
Graduate Poster
- Joined
- Oct 19, 2004
- Messages
- 1,395
I believe the test as described here could be passed without paranormal means. People can learn to fool a polygraph, i.e. manipulate the response that it gives. So the sender and receiver could develop a code that involves maniuplating the polygraph response based on characteristics of the word being tested. If the sender could then interpret the coded messages about what type of word each chart involved, he could probably not hit anything like 100%, but could increase the probability of correctly identifying the chart above chance. Do it enough times, and win $1M. I don't think this is very practical, but I don't think we could rule it out.
IXP
This all sounds very similar to the presentiment research done by Radin and others, in which physiologic changes (like those used for lie-detecting), rather than conscious awareness, are used to indicate anomalous cognition. Radin's results are too similar to noise to draw conclusions strong enough to pass the Million Dollar Challenge, but if you can see a strong pattern in your data, regardless of what happens here, maybe you should contact Radin.
It may be that people aren't lying, but that the word didn't reach conscious awareness. In that case, incentives and honesty won't help you.
Linda
It may be that people aren't lying, but that the word didn't reach conscious awareness. In that case, incentives and honesty won't help you.
Linda
Mojo
Mostly harmless
To get around this, they could use a list of words, some of which are agreed to cause the 'receiver' to produce a response and some of which don't, and randomise the order in which they are 'sent' so the receiver can't know when they are supposed to respond. Then see if the polygraph agrees with the times the words were 'sent'. If golfy can 'send' words, then it should be possible for the person reading the polygraph to identify, blind, which were the words that cause a response.
Golfy's claim really boils down to "I can remotely affect someone's polygraph results". Is that about it, Golfy?
This claim ought to be eligible and testable under the MDC rules, and can easily be controlled with appropriate blinding - maybe triple-blind, if a polygraph reader other than Golfy is used - to eliminate cheating.
This ought to be simple. It never is 'though. This will get haggled 'round and sidetracked and mired in irrelevant argument and somehow the two sides will not get a test done.
This claim ought to be eligible and testable under the MDC rules, and can easily be controlled with appropriate blinding - maybe triple-blind, if a polygraph reader other than Golfy is used - to eliminate cheating.
This ought to be simple. It never is 'though. This will get haggled 'round and sidetracked and mired in irrelevant argument and somehow the two sides will not get a test done.
Loss Leader
I would save the receptionist., Moderator
I understand what you and Rasmus are saying. And I agree with you or, at least, I would. But I don't think that you've correctly understood exactly what it is golfy wants to do. That's because I don't think golfy has correctly expressed exactly what it is he wants to do.
Every time he posts, his claims become more and more vague and disquieting. Witness:
It's shaping up to be:
1. I have a claim
2. But I can't prove it because people are lying
3. So I have a way to get around that problem
4. Unless they're accomplished liars.
If he succeeds, it means he is telepathic. If he fails, it means that people are conspiring to deny that he is telepathic.
Pup
Philosopher
- Joined
- Dec 21, 2004
- Messages
- 6,679
But the receiver doesn't know what word is being sent, so what difference does it make if he can manipulate the polygraph?
Here's what I'm proposing, if golfy thinks he could do it.
Sender in one room, receiver hooked up to polygraph in the other with a tester. No contact between sender and receiver. Neither knows what word will be sent until after they're separated, nor does the polygraph tester.
Sender is a given a random word to send, let's say, "blue." He sends it mentally.
Separately, at the same time, the polygraph tester, who also doesn't know what word is being sent, is given a list a ten words, one of which is blue. He asks the receiver, "Is he sending house? Is he sending hat? Is he sending blue? Is he sending chair?" (or whatever random ten words are on the list). He marks the polygraph print-out for each word.
Golfy believes that even if the receiver answers "no" to "is he sending blue?" there will be a spike in the chart which golfy will be able to recognize.
This is done x number of times, with a different sending word each time.
At the end, there are x number of polygraph print-outs, each with ten words written on them. Golfy believes each of those print-outs will have a spike connected to one word.
Golfy is given the print-outs in random order, and he marks which word has the spike.
If it's the word which was actually sent during that polygraph test, it's counted as a hit. If it's the wrong word, it's a miss. Period. No judging. Right or wrong.
Just need to make sure there's no way for golfy to signal the receiver during the test by non-paranormal means (same as for any telepathy test), and to agree on how many 1/10 chances golfy needs to get right to pass.
Edited to add: Golfy, do you think you can always make your receiver respond with something recognizable on a polygraph print-out, no matter how cleverly or skillfully he or she lies about the word you sent?
Last edited:
Loss Leader
I would save the receptionist., Moderator
I don't see any reason for Golfy to look at the polygraph results. The Polygrapher will ask the ten words; only two results are possible: The polygrapher will be able to determine which word the person received or the results of the test will not lead to a clear determination.
If the person received the word well, the polygrapher should be able to determine that he is either being truthful about hearing it or untruthful about not hearing it.
Why would Golfy have to see the polygraph results? Furthermore, how could he determine which word corresponded to which set of results? each set of results should (if the receiver answers no for every question) look like nine flat responses and one wide response. How could Golfy figure out which coresponded to what? Each set of results would look the same.
Hi Pup,
Your test is exactly as I intended it to be as it gets around any cheating. The JREF has suggested that it should be 1000:1 odds or greater – that would mean 3 exact hits in 3 tests = 1000:1 odds – if that is achieved then prize won. If one is a miss then another has to be a hit so 1 miss and 4 hits would be the same result, 1000:1 etc or so I was led to believe but 3 hits with 3 questions of 10 words each would be enough, 4 in 4 would be icing on the cake.
With a polygraph it depends on the person being tested and the stress level involved. If the receiver is an ideal subject then the lies would be obvious and the truths equally obvious i.e. calm responses produce little and the lies would produce large peaks on the graph.
I have done a number of tests with a GSR and it depends a lot on the persons guilt level when they lie. If they were to be arrested if found to be lying then the polygraph would have a virtual 100% accuracy, if it is about the winnings then if they lied and nothing was to happen to them if they did not produce a large polygraph deflection then the polygraph may not be so effective as it could be.
Quote from a polygraph examiner “The difficulty in this, is that for polygraph to work in the form it is designed for, the reaction may be minor due to the lack of fear at being found out lying.”
Stress attached to them lying would make the polygraph much more effective. How this could be done I don’t know but it certainly would increase my chances of winning the prize.
Regards
golfy
Your test is exactly as I intended it to be as it gets around any cheating. The JREF has suggested that it should be 1000:1 odds or greater – that would mean 3 exact hits in 3 tests = 1000:1 odds – if that is achieved then prize won. If one is a miss then another has to be a hit so 1 miss and 4 hits would be the same result, 1000:1 etc or so I was led to believe but 3 hits with 3 questions of 10 words each would be enough, 4 in 4 would be icing on the cake.
With a polygraph it depends on the person being tested and the stress level involved. If the receiver is an ideal subject then the lies would be obvious and the truths equally obvious i.e. calm responses produce little and the lies would produce large peaks on the graph.
I have done a number of tests with a GSR and it depends a lot on the persons guilt level when they lie. If they were to be arrested if found to be lying then the polygraph would have a virtual 100% accuracy, if it is about the winnings then if they lied and nothing was to happen to them if they did not produce a large polygraph deflection then the polygraph may not be so effective as it could be.
Quote from a polygraph examiner “The difficulty in this, is that for polygraph to work in the form it is designed for, the reaction may be minor due to the lack of fear at being found out lying.”
Stress attached to them lying would make the polygraph much more effective. How this could be done I don’t know but it certainly would increase my chances of winning the prize.
Regards
golfy
Rasmus
Philosopher
- Joined
- Jul 27, 2005
- Messages
- 6,372
I was under the impression that polygraphs do not perform better than chance when tested under scientific conditions?
Because that is what I understood his ability was. And we agree that wasn't spelled out very clearly. But if that is the claim we shouldn't be concerned with the "why" or "how" as long as we can rule out cheating.
Personally, I would have thought that the polygraph introduces an extra-chance of failure; but that was assuming that the receiver would be aware of the word he received. If he's not, then the polygraph might tap into his unconsciousness and thus reveal which word he received.
Great! Yes, it *should*. If so, that would almost get the test passed - if it weren't for the fact that someone would have to judge what is and isn't a 'wide response' as opposed to a 'flat' one.
If that determination is done by the applicant himself, the final result will be self-evident again.
No need to. There shouldn't be different read outs, since there is no (non-paranormal) way that the receiver could know which word was being send. If he does, the test is passed - right?
This is exactly why the JREF will (likely) refuse to allow a polygraph: There must not be any room for interpretation of the results, they have to be self-evident.
What do you think of the Achau Nguyen protocol, golfy?
NobbyNobbs
Gazerbeam's Protege
- Joined
- Apr 2, 2006
- Messages
- 5,617
For me to tell you your surname, wouldn't I have to know exactly when you are "thinking" it at me, so I know when to concentrate? Otherwise, any names that pop into my head all day are potential candidates. If you'd like to give me a day and time when you will think your name at me, I'll be open to receiving it.
As far as distance goes, since you don't know where I am, wouldn't it be difficult to determine if you can do this?
If you are still up for the conditions I set (i.e., get test conditions approved from JREF, get me to and from the test site, pay me $100,000 if you win) then I am still willing to be your volunteer.
I still don't see why a polygraph is necessary. I very much doubt JREF would allow it.
If by some miracle you get them to allow it, I'd be willing to offer some form of punishment to guarantee I won't lie. You said the more stress, the better the results. Fine. If I lie, I spend a week in jail. Will that work? (I really would not look forward to a week in jail, so the stress level ought to be more than adequate.)
Last edited:
Loss Leader
I would save the receptionist., Moderator
I don't know. I can't figure out what he's claiming, anymore. It just seems like he's adding an extra layer of procedures and people to blame when he fails.
But my guess is that his thinking will grow increasingly grandiose and disorganized and no agreement of any sort will ever be reached.
IXP
Graduate Poster
- Joined
- Oct 19, 2004
- Messages
- 1,395
Here is how you defeat this test:
Golfy finds a receiver who can manipulate the polygraph, showing either a high or low stress level. Golfy and the reciever have agreed beforehand that if the word begins with A-L he will show a low stress level and if it begins with M-Z he will show a high stress level. Assuming he can achieve this 100% and golfy can read it 100%, he can now guess which chart applies to the work with 2x the probabilty of doing it by chance. By breaking down the period of each reading into multiple parts, they might even achieve more bits of communication.
IXP
Last edited:
IXP
Graduate Poster
- Joined
- Oct 19, 2004
- Messages
- 1,395
No, you are not being dense, noone else seems to get it either. Go back and read the protocol that Pup posted (quoted in my post.)
The receiver is questioned as to whether each of the 10 words in the list is the one that was sent. So he sees all of the words. The polygraph outputs of the 10 separate questions are given to Golfy for him to determine which was the sent word. Basically the cheating method is for the receiver to mark the charts for each work with some information that allows Golfy to guess more accurately. This is a the kind of subtle information leak that has always plagued paranormal testing.
IXP
As you still haven't put the University plots on the forum than I'll have to wait untill I can in a few more posts. The intention and what the protocol has been made very clear and is only being confused by "too many cooks".
Pup described it as accurately as I could have.
golfy
Pup described it as accurately as I could have.
golfy
Rasmus
Philosopher
- Joined
- Jul 27, 2005
- Messages
- 6,372
No, the problem is that the procedure hasn't been explained in any detail at all.
Here is how I understand the procedure:
Stage 1: Telepathic transmission
Sender sends, receiver receives. After a set period of time or a signal of "success" etc. stage one comes to an end.
Stage 2: Polygraph phase
A neutral tester quizzes the receiver who is at that time hooked up to a polygraph:
Tester: Did you receive the word "monkey"?
Receiver: Yes/No.
Tester: Did you receive the word "cellphone"?
Receiver: Yes/No.
etc. pp.
Here it might indeed be possible for the receiver to encode some information about the words in the polygraph charts. Depending on how well this works, the chances of getting the right answer by non-paranormal means at a later stage could indeed increase drastically.
Imagine I could encode 3 bits of information in the polygraph chart for every word. I could narrow down the first letter of each word quite accurately. (I could receive far better results with a smarter algorithm, I guess, if I took into account the relative frequency of letters, etc.)
That would give me a *big* edge in the test. If I was even smarter and took into account that I could get lists with only words starting with the same letter things would be even better ...
So, if this is the intended procedure it would indeed be doubtful that it could go ahead ...