Merged New telepathy test: which number did I write ?

[Nay_Sayer]

Keep doing the same thing and hoping for a different result?

Shirley nobody would do such a thing.

3!

and don't call me shirley

 

[Michel H]

why is it invalid?

Right you only accept certain answers, gotcha

Because, with the protocol used in this test, you may not post the "guessed" number in your first post. You should have posted:

xx, Lets keep the trend going

Also invalid test is still invalid

instead of:

3, Lets keep the trend going

Also invalid test is still invalid

,
and sent your full answer to Agatha.

 

[fromdownunder]

And it's not indispensible to be an incredible genius to deliberately answer badly (or not answer at all)...

So anybody who gives the correct answer is correct, and anybody who answers incorrectly is deliberately doing it just to trick you into thinking you are not telepathic?

Got it!

If you are so certain that we are all getting the correct answer, you don't have to do it by poll. You are a true believer, and are convinced that 75% of the forum are deliberately lying to you. So a poll, where 75% of the responses lie does not help you at all.

Norm

 

[Turgor]

I think you picked the wrong number Michel H, because I am getting a strong impression that it is a suffusion of yellow.

 

[Michel H]

I think you picked the wrong number Michel H, because I am getting a strong impression that it is a suffusion of yellow.

Very good, Turgor, am I supposed to laugh?
Your post is still not an answer to my test.

 

[Akhenaten]

Very good, Turgor, am I supposed to laugh?
Your post is still not an answer to my test.

That's a bit cruel and dismissive.

How do you know poor Turgor doesn't have synesthesia?

 

[Ladewig]

I don't know, I am afraid this protocol may seem too complicated for many people, .

Actually, I am still a bit confused by the protocol. If I sent a single, unambiguous number to Agatha, what is gained by the hash ? If I tried to change my response, wouldn't Agatha be able to point out that I was violating the rules?

 

[Ladewig]

Also, Michel H., if you don't mind answering a few questions while we wait for more responses,

Does everyone have this power and you are merely better than most?

Or are there very few people able to transmit information and you are one of them?

Or are you the only known example of someone having this power?

Or something else?

 

[Kid Eager]

The last time I partcipated in this guessing game I was accused of being in a mental institution and therefore my response was invalid.

I'm now in a high-sided elastic banjo with an eskimo parasol, so rest assured that my response is both fluffy and perky.

The number I'm seeing is XX.

 

[Sherlock]

I do not know. And I know I do not know because I do not know.

 

[tuxcat]

Michel if I was you I'd stop the test and use a larger spread of numbers (say from -infinity to +infinity) in order to remove the participants guessing correctly through chance alone. As it stands, we can expect 25% of people to guess correctly the number so I don't see what it proves other than that with favourable odds people are going to come up trumps.

 

[Akhenaten]

The last time I partcipated in this guessing game I was accused of being in a mental institution and therefore my response was invalid.

I'm now in a high-sided elastic banjo with an eskimo parasol, so rest assured that my response is both fluffy and perky.

The number I'm seeing is XX.

You are so invalidated.

 

[Iamme]

This thread is a scam. Any schoolkid knows nothing can be learned from this whether 70% say 3 or 25.6% say 3. You cant prove telepathy from chance. With infinite number pick if just ONE person guesses correct number, than you test that one person again and again...duh. So here`s the deal. The scam is this. After he puts all number replies in a computer, a random generator will come up with some posters credit card number. Lol!

 

[Iamme]

This thread is a scam. Any schoolkid knows nothing can be learned from this whether 70% say 3 or 25.6% say 3. You cant prove its only telepathy from only 4 choices, duh. What if he DID choose 3, and a bunvh of people here are giving reasons its likely 3, so they all jump on the 3 bandwagon? But with say an infinite numbers pick... if just ONE person guesses the correct number, then you test that one person again and again...duh. So here`s the deal. The scam is this. After he puts all number replies in a computer, a random generator will come up with some posters credit card number. Lol!

 

[Emily's Cat]

So if your objective is to try to prove or disprove telepathy, this approach isn't going to do anything for you. As has been mentioned a couple of times: 4 choices is too few, a single round is too little, and even if you get everyone on the board to respond, there's too broad a margin for natural randomness. Statistically speaking, none of your results are going to be valid.

If, however, your objective is to design a valid test with which to find evidence of telepathy, using the interwebs, then you might get a good discussion. I can make a suggestion.

Don't use numbers, letters, or cards. Humans aren't very good at being random, and IIRC there has been a fair bit of research showing that we have a tendency to choose numbers in a non-random fashion, just because we happen to like certain numbers for no good reason. Similarly, we tend to choose either very commonly used letters (rstlne, compliments of Wheel of Fortune ) or very rarely used letters (zqj) so that the in-between letters are very underrepresented. With cards it's similar - we have a tendency to choose face cards and low cards. So all of those schemas are bad juju for your test, because you won't get a uniform distribution of guesses.

Don't use a single guess. The reason these things have distributions is that randomness exists. There will be absolutely no reasonable way for you to distinguish between coincidence and telepathy. There's no way for you to determine whether my answer, for example, is right because I'm telepathic or because I just got lucky.

So... If you were doing this in person, you could repeat the process with slips of paper, many times, and maybe get a test that is statistically valid. Using the interwebs, there's too much opportunity for cheating. Even if you're a completely honest person, and you absolutely never cheat... the fact that there is opportunity for it to happen means that your results aren't valid.

So... If I were bored enough to design a test that is limited to an interwebs discussion forum... Here's what I would do to get as valid and accurate a result as I could.

Select a set of colored shapes (go find images of them, from an approved hotlinking source, and provide all of us with the links for them). Stick with a simple palette of colors - I'd recommend the primaries, plus black and white, for a total of 5 possible colors. Use simple, even polygons - circle, square, triangle. All told you're looking at about 15 items.

Choose a trusted moderator for the test. You put together a sequence of 30 items. You can repeat items however many times you want, you can generate the sequence randomly, it doesn't matter. You provide your sequence to the trusted moderator before opening the thread.

Then simply ask each poster to likewise make a sequence of 30 items.

Compare position in the sequence for each item, and look at the distribution of how many each person gets correct.

In this fashion, you treat each item in the list as if it were a repeated guess. For this test, your null hypothesis is that the number of correct guesses per poster is normally distributed. If the distribution is fairly normal in shape, then there is no evidence of any sort of telepathy.

Important: If a skew exists in your distribution, that is not evidence that telepathy exists, it will only tell you that your null hypothesis fails - that there is not enough statistical evidence to conclude that telepathy does NOT exist.

You could probably design something more sophisticated if you spent a long time at it... but I would be comfortable accepting this as a reasonably valid test, provided you understand what the results actually mean, and you understand the null hypothesis.

 

[stanfr]

Here:
the number i came up with is ##

what's the point? what does my statement have to do with credibility?
As many have pointed out, your protocol is flawed. If I said for example:

I chose ## because that is the number most mentalists will chose.

How is this going to do anything to demonstrate 'telepathy'? All it will test is how versed you are in mentailism.

 

[Nay_Sayer]

Here:
the number i came up with is ##

what's the point? what does my statement have to do with credibility?
As many have pointed out, your protocol is flawed. If I said for example:

I chose ## because that is the number most mentalists will chose.

How is this going to do anything to demonstrate 'telepathy'? All it will test is how versed you are in mentailism.

It won't. various people have explained why "Pick a number 1,2,3 of 4 and I will guess it, If I do, I am psychic!" is flawed, it has not worked.

 

[Ladewig]

Perhaps five or six valid answers would be a good number, but this is maybe somewhat unrealistic, I don't know..

First, let me say that I mean this post to be respectful and I hope it doesn't not hurt my credibility rating.

Let me address the issue with a question.
If someone approached you and said, I have telekinetic powers such that I can influence the roll of an ordinary six-sided die. And that person said, please roll the die six times while I concentrate on the number 5. And the results were 5,2,6,4,5,3. And the claimant said "ah ha, my results are double what one would expect from chance, therefore I have proved my telekinetic power." Would you believe the claimant actually had that power? If you were not convinced, then what would you want to see before you were convinced?

 

[Michel H]

First, let me say that I mean this post to be respectful and I hope it doesn't not hurt my credibility rating.

Let me address the issue with a question.
If someone approached you and said, I have telekinetic powers such that I can influence the roll of an ordinary six-sided die. And that person said, please roll the die six times while I concentrate on the number 5. And the results were 5,2,6,4,5,3. And the claimant said "ah ha, my results are double what one would expect from chance, therefore I have proved my telekinetic power." Would you believe the claimant actually had that power? If you were not convinced, then what would you want to see before you were convinced?

Ladewig, the result of the telekinesis experiment you mention is two successes after six trials. The binomial probability of getting two or more successes after six trials, (when the probability of success on a single trial is equal to 1/6 = 0.16666666666), is equal to p = 26.3% (see here). This is a fairly large probability, so the result of your hypothetical experiment is not statistically significant (see Exemple 1 on this webpage). Telekinesis is a (physically) almost impossible phenomenon, so you would need a much smaller p-value for your "evidence" to be considered convincing.

 

[Ladewig]

Ladewig, the result of the telekinesis experiment you mention is two successes after six trials. The binomial probability of getting two or more successes after six trials, (when the probability of success on a single trial is equal to 1/6 = 0.16666666666), is equal to p = 26.3% (see here). This is a fairly large probability, so the result of your hypothetical experiment is not statistically significant (see Exemple 1 on this webpage). Telekinesis is a (physically) almost impossible phenomenon, so you would need a much smaller p-value for your "evidence" to be considered convincing.

So why did you think "Perhaps five or six valid answers would be a good number"? You obviously know statistics well enough to know that five or six answers would be completely and unmistakably meaningless in such a test as this one.