Merged New telepathy test: which number did I write ?

[Jack by the hedge]

Where are the fan sites, eagerly discussing whatever Michel's thinking today?

Where are the political bloggers, applauding or decrying Michel's attitude to something or other?

Where are the comedians, joking about annoying it is when we all keep being interrupted by some jackass who keeps thinking numbers into our heads?

Where are they all? Nowhere. They don't exist. And they don't exist simply because nobody really hears anybody else's thoughts. Telepathy is just imaginary.

What is real is that some people get a very strong sensation that other people can hear what they think. But it's just a feeling. It's not a fact. It's not true.

 

[Michel H]

...
However, there is no event where you can call your test valid even with a million answers and no credibility scale whatsoever. Why? Because the choice of numbers is so small that psychological factors come into play. It may be that, asked to choose a number between 1 and 3, most people pick 3. You would at the very least have to run a test to see whether people, in the absence of telepathy, favor one number over others. ...

The problem you mention here is easily solved by running, not just one, but several tests. If I do just one test, for which the target is number 3 (like in this test I have just completed), then, if, for some reason, people like to answer that number (3), I may have a false impression of success (note, however, that in the test I just finished on this forum, actually one person answered 1, another answered 2, and nobody answered 3); perhaps many people have answered 3 just because they like answering 3 (or because they have a tendency to answer 3), and perhaps no extra-sensory communication was involved at all.
However, let us assume that the next target is 1. Then, by answering many 3's again, participants will get disastrous results, so this difficulty of "false positive" is eliminated.

 

[ComfySlippers]

The problem you mention here is easily solved by running, not just one, but several tests. If I do just one test, for which the target is number 3 (like in this test I have just completed), then, if, for some reason, people like to answer that number (3), I may have a false impression of success (note, however, that in the test I just finished on this forum, actually one person answered 1, another answered 2, and nobody answered 3); perhaps many people have answered 3 just because they like answering 3 (or because they have a tendency to answer 3), and perhaps no extra-sensory communication was involved at all.

However, let us assume that the next target is 1. Then, by answering many 3's again, participants will get disastrous results, so the difficulty of "false positive" is eliminated.

Claptrap.

 

[fromdownunder]

However, let us assume that the next target is 1. Then, by answering many 3's again, participants will get disastrous results, so this difficulty of "false positive" is eliminated.

I am trying to understand that if YOU pick one in your next "test" and circle it 38, 312 times, and most "participants" pick 3, how can you possibly come to some sort of conclusion that this is disastrous results for the participants.

That makes no sense whatsoever, even by your standards.

Norm

 

[Michel H]

I am trying to understand that if YOU pick one in your next "test" and circle it 38, 312 times, and most "participants" pick 3, how can you possibly come to some sort of conclusion that this is disastrous results for the participants.

That makes no sense whatsoever, even by your standards.

Norm

I am having some difficulty understanding your strange (first) sentence, Norm. But perhaps I don't understand Australian English good enough...
Two examples:
Test 1:
- target = 3 (that's the number to divine, supposedly using telepathy)
- answers: 1 3 3 2 3 (tendency to answer 3)
- hit rate = 3/5 = 60% (much higher than 33%)
Test 2
- target = 1
- answers = 3 2 3 3 1 (again, same tendency to answer 3)
- hit rate = 1/5 = 20% (smaller than 33%, this time)

So, we see that the propensity to answer 3, which was an advantage in the first test, has become a disadvantage, a liability in the second test. The average hit rate (60 + 20)/2 = 40% has now become fairly close to 33% (the average chance result).

 

[fromdownunder]

So, we see that the propensity to answer 3, which was an advantage in the first test, has become a disadvantage, a liability in the second test. The average hit rate (60 + 20)/2 = 40% has now become fairly close to 33% (the average chance result).

You said that answering "3" in a test where you actually circled "1" would be a disadvantage to the participants. How would this be a disadvantage to the participants? Is that question easy enough?

Norm

 

[Michel H]

You said that answering "3" in a test where you actually circled "1" would be a disadvantage to the participants. How would this be a disadvantage to the participants? Is that question easy enough?

Norm

Because they would answer wrong.

 

[Loss Leader]

However, let us assume that the next target is 1. Then, by answering many 3's again, participants will get disastrous results, so this difficulty of "false positive" is eliminated.

No, it's not because you keep picking and choosing which answers you like enough to keep and discarding the others.

Meanwhile, the only way to say that aggressive, negative, non-credible, uncertain guesses are properly excluded is if you assume that you are telepathic as a function of your test of telepathy.

If you're not telepathic, then the most aggressive answer you get carries as much weight as the one you find credible.

 

[fromdownunder]

Because they would answer wrong.

I think that I have finally discovered all I need to know here. That answer is positively grotesque. But I will go one more time:

How would it be a disadvantage to me, if, as a participant if I answered three instead of 1? Please think about your response before engaging your keyboard.

Norm

 

[Michel H]

No, it's not because you keep picking and choosing which answers you like enough to keep and discarding the others.

Meanwhile, the only way to say that aggressive, negative, non-credible, uncertain guesses are properly excluded is if you assume that you are telepathic as a function of your test of telepathy.

If you're not telepathic, then the most aggressive answer you get carries as much weight as the one you find credible.

If no telepathic transmission is experienced during the test, then I think "I don't know" is the right answer, ideally with some kind, serious and informative comment. Such an answer would not be included in the statistical analysis of the numeric answers, but I would keep it in mind nevertheless, and perhaps mention it in the analysis of the test.

 

[Loss Leader]

If no telepathic transmission is experienced during the test, then I think "I don't know" is the right answer, ideally with some kind, serious and informative comment. Such an answer would not be included in the statistical analysis of the numeric answers, but I would keep it in mind nevertheless, and perhaps mention it in the analysis of the test.

If you have predetermined what the "right" answer is to show you do not have telepathy, then you are continuing to assume as part of the test that you have telepathy.

Is there any actual reason to believe that people will type "I don't know" when they are not receiving a number rather than just guess a number? Have you run any sort of test on how people behave in the absence of telepathy?

You have not. You're just assuming that certain types of answers are valid or invalid without any reason.

That's not entirely true. You have a very good reason: You're trying to make the results look like you are telepathic.

It's not a proper test, it's not a properly administered test, it's not a properly graded test, and it's not a properly analyzed test. You have failed at every conceivable step.

 

[Pixel42]

You won't run a better test, though. You know it will show that you have no special powers.

He did once run a slightly better test - the one where he had to choose a credibility rating before knowing whether the number guessed was correct - and it did indeed show exactly that.

 

[Michel H]

...
It's not a proper test, it's not a properly administered test, it's not a properly graded test, and it's not a properly analyzed test. You have failed at every conceivable step.

Edited by Agatha: 

Edited to remove breach of rule 8.

Loss Leader, I believe you are a lawyer, and I assume you are kinder with your clients than with me.
Not everybody seems to agree that my test is failed and improper.
For example, another member said:

Well this is certainly one of the most robustly controlled experiments I have encountered.

And Ashles never said:

...
All of my responses to any of your tests have been lies.

If I were you, I would discard all my responses as not being credible.
...

I wonder what is left of your credibility after such a post.
You tweeted on February (the month of my birthday) 9:
Your first duty is to God, but since God doesn't exist, just feel free to take the day to yourself.
I am afraid this may reflect (at least partly) your attitude and mentality (and also perhaps attitudes and mentalities of some other members here too) with respect to my tests: "Just do what you like, and everything will be fine!".

I also received recently some words of praise about my test, on a French medical forum.

 

[turingtest]

If no telepathic transmission is experienced during the test, then I think "I don't know" is the right answer, ideally with some kind, serious and informative comment. Such an answer would not be included in the statistical analysis of the numeric answers, but I would keep it in mind nevertheless, and perhaps mention it in the analysis of the test.

This is ridiculous reasoning, Michel. If you were truly testing honestly for telepathic transmission, you wouldn't be discarding the "I don't know" answers from your analysis, because lack of reception could indicate lack of transmission; "I don't know" is as valid an answer to show that as a wrong number. Instead, you've cleanly divided the two things; you're really only testing for a reception of a transmission you're simply taking for granted. A "test" based on such blatant question-begging is no real test at all.

 

[Michel H]

If you have predetermined what the "right" answer is to show you do not have telepathy, then you are continuing to assume as part of the test that you have telepathy.
...

This is ridiculous reasoning, Michel. If you were truly testing honestly for telepathic transmission, you wouldn't be discarding the "I don't know" answers from your analysis, because lack of reception could indicate lack of transmission; "I don't know" is as valid an answer to show that as a wrong number. Instead, you've cleanly divided the two things; you're really only testing for a reception of a transmission you're simply taking for granted. A "test" based on such blatant confirmation bias is no real test at all.

No, I do not assume that my telepathy is real when I do a new test, and all participants have a meaningful opportunity to say "I don't know" (using of course their own words if they wish) in most of my tests. However, I frankly believe that the "I don't know" answers (usually mixed with some aggressivity) are not the most interesting ones; if I really am not telepathic, then the results should be random, but they are not (in my opinion). For example, I find interesting that, on Yahoo (in my latest test), I got a 100% hit rate, a hit rate equal to 40% on Skeptiko and a hit rate equal to 0% on this forum, which has a more skeptic orientation. It seems to me that this is not a random result: members of this forum, I suppose, simply gave wrong answers because they understood that answering right would contradict their skeptical ideology. Nothing complicated here.

 

[Mashuna]

No, I do not assume that my telepathy is real when I do a new test, and all participants have a meaningful opportunity to say "I don't know" (using of course their own words if they wish) in most of my tests. However, I frankly believe that the "I don't know" answers (usually mixed with some aggressivity) are not the most interesting ones; if I really am not telepathic, then the results should be random, but they are not (in my opinion). For example, I find interesting that, on Yahoo (in my latest test), I got a 100% hit rate, a hit rate equal to 40% on Skeptiko and a hit rate equal to 0% on this forum, which has a more skeptic orientation. It seems to me that this is not a random result: members of this forum, I suppose, simply gave wrong answers because they understood that answering right would contradict their skeptical ideology. Nothing complicated here.

If you say you do not assume your telepathy is real when you do a new test, why do you then assume that members of this forum are deliberately giving wrong answers? You could only come to that conclusion if you are assuming that your telepathy is real.

 

[jsfisher]

I also received recently some words of praise about my test, on a French medical forum.

Michel H,
Since you appear to be completely unable to appreciate sarcasm in any form, how do you know the "words of praise" you received weren't mockery?

Your test and your interpretation of the results have been mocked here, but you were unable to tell that until each example was explained in detail to you. Why do you assume the French medical forum was any different?

 

[Loss Leader]

J I assume you are kinder with your clients than with me.

Actually, I am direct and matter of fact with my clients. And when one of them insists on being unrealistic, I sit him/her down and very frankly outline for them what mistakes they are making. My mother used to write in her billing "Attitude Adjustment" but now we write "Conference with client re: goals."

Not everybody seems to agree that my test is failed and improper.
For example, another member said:

Once again, that was sarcasm. The individual meant the opposite.

I wonder what is left of your credibility after such a post.

You've previously argued that I have some special credibility because I'm a moderator. So now will you be disregarding the previous answers I gave that you considered hits? Will you be disregarding the completely random number I generated for use in a different thread that you somehow also considered a hit?

You tweeted on February (the month of my birthday) 9:
Your first duty is to God, but since God doesn't exist, just feel free to take the day to yourself.

Other tweets of mine from February:

I told my wife I had decided to embark on a voyage of self-discovery and she said to take one of the kids.

and

I'm sponsoring a bill in the New York State Assembly to rename "Buttermilk" to "What The Hell Is This?"

Follow me on Twitter: @LossLeader

 

[Loss Leader]

No, I do not assume that my telepathy is real when I do a new test, and all participants have a meaningful opportunity to say "I don't know" (using of course their own words if they wish) in most of my tests. However, I frankly believe that the "I don't know" answers (usually mixed with some aggressivity) are not the most interesting ones; if I really am not telepathic, then the results should be random, but they are not (in my opinion). For example, I find interesting that, on Yahoo (in my latest test), I got a 100% hit rate, a hit rate equal to 40% on Skeptiko and a hit rate equal to 0% on this forum, which has a more skeptic orientation. It seems to me that this is not a random result: members of this forum, I suppose, simply gave wrong answers because they understood that answering right would contradict their skeptical ideology. Nothing complicated here.

The two highlighted passages contradict each other.

Also, follow me on Twitter: @LossLeader

 

[Michel H]

If you say you do not assume your telepathy is real when you do a new test, why do you then assume that members of this forum are deliberately giving wrong answers? You could only come to that conclusion if you are assuming that your telepathy is real.

I do indeed believe that my telepathy is probably real, but I don't need that assumption, or hypothesis when I start a new test. When I start a new test, everything is open, but I seem to often end up with the same conclusion that I am "telepathic". Incidentally, the results in favor of my telepathy hypothesis seem to be usually stronger in long-distance testing done in English, than in shorter-distance testing done in French, as if my telepathy was getting clearer as you move away from me. The reasons for this are not entirely clear; perhaps the alleged phenomenon becomes more extraordinary, and hence more striking.