Merged New telepathy test: which number did I write ?

Status
Not open for further replies.

Michel H

Banned
Joined
May 25, 2012
Messages
3,398
Location
Belgium
Hi, I invite you to participate in a new telepathy test.

At about 20:17 on this Monday October 21 (Brussels, Belgium time), I wrote carefully one of the four numbers: "1", "2", "3", "4" on my sheet of paper, and I surrounded it with a circle. Then, I wrote it again twice.

I shall repeat this number from time to time during this test.

It was selected using this random number generator: http://www.random.org/integers/ .

Please note that the number I wrote has no meaning, it was just produced by the random generator.

I ask you to write it here (if you think you know it, even with a doubt). You may also answer "I don't know".

In this test, however, unlike the other tests I have done so far on this forum, I ask you to not write immediately the number explicitly (more on this below).

I also ask you to write a comment, together with your numerical answer, or at least a small sentence, like, for exemple:
I'll say 3. :) ,
I answer 1. Good luck. ,
or I say 4 and this is a ridiculous test. (if you want to be more aggressive).

In your comment, you may describe (or attempt to describe) the (assumed) telepathic, extra-sensory perceptions which possibly led you to give the answer you chose, and/or say a few words about telepathy in general, and/or tell us how confident you are that your number is the correct one, and/or say why you picked this particular number, and/or express an opinion on this test (good or bad), with perhaps suggestions for improvement, and so on. For inspiration, you may take a look at the 32 answers with text which have already been given on this forum in one of my telepathy tests, see analysis1 and analysis2.

My hope is that the comment or text you will write will give me some valuable (and essential) insight into your state of mind when you answered. Then, I shall use your text to rate your answer, on a credibility scale between -10 and 10, like I did in the previous tests on this forum.

In this thread however, I want to evaluate credibilities without any knowledge of the number you picked ("in a blind way"), to make sure that I don't get influenced or biased by the number you chose. This should make this test more rigorous, although at the cost of additional complexity.

To achieve this greater rigor, I ask you to give your answer in two stages. In your first post, you should write your normal, complete answer, with the "guessed" number (1, 2, 3 or 4) replaced by "xx" . So, if your normal, complete answer is, for exemple:

I am seeing a 4 very clearly. It's almost as though I had written it myself.

(this answer was given by Loss Leader in my previous test, the number 4 he gave was correct),

post instead:

I am seeing a xx very clearly. It's almost as though I had written it myself.

I also ask you to post the MD5 hash of your normal, full answer, using for exemple this website: http://www.md5hashgenerator.com/ . If your answer is very simple, I ask you to make it more complex by adding additional characters and to post the MD5 hash of this more complicated text (this is to make sure your hash cannot be decrypted, see for exemple this website: http://www.md5online.org/ , or this one: http://www.md5decrypter.co.uk/. In most cases, additional characters will probably not be necessary) . For exemple, if your answer is just "I answer 1", instead of posting the hash of "I answer 1", post the hash of "I answer 1 oibyytrtzerazewoytitrxeuytepiurezqaytiyttuecrtexruog". People who answer "I don't know" (possibly with a text) should not introduce xx's in their answers (and there is no need for MD5 hashes either).

After a reasonable number of forum members have validly answered (if this "reasonable number" is ever achieved ), I should post my "credibility ratings" for all valid answers to this test (and also the number I wrote and circled), and then (and only then) you should post your actual, full answers (with guessed numbers now visible), and also (possibly) the "complexified" text you used for the MD5. The MD5 hash of your full answer (or of your "complexified" answer) should match the hash you have given previously, in your (normally) first post. When this is done, for all answerers, I should then post an analysis of the results, like I have done for my two previous tests.

It may also be useful (I recommend it) that you send your (full) answer, in the form of a private message, to either Agatha, or Femke, or to both (click on "Contact Info", and then on "Send a private message..." on their profile pages. On these pages, look also under their names to see their "last activity" dates and times). Agatha and Femke are active members on this forum who have kindly offered to help, in order to make my tests more rigorous:
...

I reiterate what I have said before: you will not get meaningful results from a test such as this. If you insist on rating answers subjectively for 'credibility', then you would need to introduce blinding so that you do not know which answers are right or wrong. If you want to do such a test, I am willing to be the person to whom everyone sends their answers, once a protocol is established.
...
,
Michel,
I offer myself as a blinder: you can PM me the hash, and I am willing to receive the answers from members who want to participate. After a previously arranged period of time I will provide a list of their answers, verbatim with an X for the number. You can then post the credibility rating, and afterwards I will provide the actual numbers.

Agatha and Femke should then post the full answers after I have posted my credibility ratings. If Femke and Agatha cannot, or no longer want to cooperate as an assistant, a "helper", then I may have to request the help from another volunteer. One possibility would be that each answerer sends his/her answer to the previous answerer, and perhaps the first answerer sends his/her answer to a Moderator (perhaps Loss Leader), or an Administrator (perhaps LashL).

Femke, if you want to participate in this telepathy test (you are welcome to do so), it might be somewhat useful you send your answer to Agatha (similarly for Agatha, although Agatha has not answered in the telepathy tests so far, she has only commented in the threads).

Sending your answers to a "helper" may be useful for several reasons. First, some answerers in the test might never return to post their full answers (or might return only after several weeks or months). This may be particularly frustrating if their answers look very interesting. Also, some people may have lost their answers, or may post them inaccurately. The redundancy in the people who can post full answers (after I have posted my credibility ratings) may be useful to get full answers quickly. Of course, if a complete answer has already be posted by either Agatha, or Femke, or the member himself/herself, it is not indispensable to post it a second time (I can check its correctness using the MD5 hash).

I shall now explain again the protocol of this test, using a simple exemple (if some of you think this protocol is not valid, or not good, I am of course willing to consider reasonable changes).

Let's assume again that your normal ("unmasked"), full answer is:

I am seeing a 4 very clearly. It's almost as though I had written it myself.

With the protocol, the procedure used in this test, the first thing you should (ideally) do is to send a private message to Agatha and/or Femke, for exemple:

Hi Agatha/Femke,

My answer to Michel H's test is:

I am seeing a 4 very clearly. It's almost as though I had written it myself.

Sincerely

(Of course, in this new test, the correct number is not necessarily equal to 4, 4 has only a 25% probability.)

Then, your first answer, your first post on this thread should (normally) be (if it is an answer to the test, not a comment), for exemple:

I am seeing a xx very clearly. It's almost as though I had written it myself.

MD5 hash of full answer: 3f429870bc6ff08d79d49ee471004b44

Complete answer sent to Agatha/Femke.



You should then wait until I post all credibility ratings (with my number). When this is done, you should post e.g. (if neither Agatha nor Femke has already posted your full answer):

{My full answer to the test is:

I am seeing a 4 very clearly. It's almost as though I had written it myself.}

You may also just say:

The number I chose was 4.

Thank you for participating.
 
Statistical probability doesn't imply telepathy.

Let's play a variation on your game. I'll flip a coin, and you tell me what you think the land is: heads or tails. If you're right more than 50% of the time, I'll say you're telepathic.

Here's the rub though: the 50% probability on coin tosses is based on a normal distribution, with a normal error bound. The number of heads in a series of tosses will be higher than 50% heads about half the time. The number of times you're right in a random guess will be about 50%... but as that is also a normal distribution, it will be higher than that about half the time as well.

Now, if I do a series of 5 throws, and you get every single one right, then that's pretty neat. But reality is that you've got around a 3% chance of being exactly right on every single toss in a series of 5 tosses. That 3% doesn't imply that you have any special abilities. It just means that statistical variance exists.
 
Why not make this new post a Vote?

:cool:
A Vote? You probably mean a poll? The texts or comments people write are important to me in these tests. If I post a poll, some people might answer anonymously in the poll without posting any text in the thread, I'm not interested in that. In addition, the security would be low, because (although I wouldn't do it) I could say that I wrote and circled the number which was answered most.
 
A Vote? You probably mean a poll? The texts or comments people write are important to me in these tests. If I post a poll, some people might answer anonymously in the poll without posting any text in the thread, I'm not interested in that. In addition, the security would be low, because (although I wouldn't do it) I could say that I wrote and circled the number which was answered most.

What keeps you from doing that if people type their answers in anyway?
 
What keeps you from doing that if people type their answers in anyway?
Well, Emily, I explained in the opening post that participants in the test should write xx instead of their chosen numbers in their first posts (ruling out just comments). Only later should you reveal your chosen number.
 
I do not understand any aspect of the point of this thread.

The answer is 3 because it just is.

A few years ago on these forums I correctly guessed a poster's four digit security code because I am cool and telepathic and stuff.
 
I think this is a silly experiment, but...

You should ask people to put their answers in spoiler tags, and not to look at anyone else's answers before giving their own, to avoid being influenced by them.

3.

 
why no blinding of the guesses?
Is this so you can only count the answers you need to give the result you want?
 
Actually, if you read the full text of this new proposal, it really is better than the previous ones. He is making a step towards blinding. If you are answering, you are supposed to hide the number you selected, (For example, "I am clearly seeing xx.")and PM the actual response to either Agatha or Femke.

Michel, I will wait to see if either of them are still willing to volunteer, and then post my answer here. Thank you for taking the first step to improving your test. :)
 
I am happy to volunteer :D Just send me the numbers you perceive/guess/whatever and I will keep them secret until the appropriate time.
 
If people post their answer, and also the hash of their answer, can the hash be recreated by someone adding a number in place of the xx that is posted in the thread? If so, that would be a problem with the blinding. Edit: I see you've specified adding some more unguessable characters to the answer prior to hashing it, so that should be fine.
 
Last edited:
If people post their answer, and also the hash of their answer, can the hash be recreated by someone adding a number in place of the xx that is posted in the thread? If so, that would be a problem with the blinding. Edit: I see you've specified adding some more unguessable characters to the answer prior to hashing it, so that should be fine.
This is a very good remark you are making, Agatha. This seems indeed to be rather serious weakness in my protocol (and even an error, actually). I am glad that I said "if some of you think this protocol is not valid, or not good, I am of course willing to consider reasonable changes". It is time now for me to consider a change ;) . But the change is, like you said, just to add many (almost) unguessable characters to the full answer, before computing the MD5 hash (see initial post [I say this, not for you, Agatha, but for the other posters]). It is unfortunate that, on this forum, it no longer possible to make additions to the opening post after some time, otherwise I would post a correction there.

So, to all posters and participants, I say this:

Please, together with your "masked" answer (with "xx"), post a MD5 hash of your properly "complexified" answer (which you will have to reveal at the end of the test), and also say to whom you sent your answer (Agatha, Femke or nobody).
 
The first number that came to my attention is xx, so that is my choice for this test.
Very good, Hokulele, thank you. This sounds like a very good start. Now, could you also please post a MD5 hash of your properly "complexified" answer (like, for exemple: "The first number that came to my attention is 5, so that is my choice for this test. uhzuhvf oihevbio hervoije roihjervoir voiervhjeroih jveihsfdkyg hzerqresmlln")? Avoid too long character strings with no blank space, because those could get modified by the forum software.

Also, please say to whom you sent your full answer (if you sent it to anybody).
 
The OP repeats one of the choices (4) at least eleven times in describing the test. This alone is sufficient to introduce enough potential bias to invalidate the test (if it wasn't bad enough already).
 
Status
Not open for further replies.

Back
Top Bottom