An SDMB test of astral projection

[umop_ap_sdn]

Originally posted by glee *
I don’t think any of my objects are hard to identify. However you raise a valid point. If I dabble in some obscure field and have a ‘quantum recursive heuristic holographic analyser’ widget, then many people would be baffled by it.

It’s not so much the quantum recursive analyzer thing (or even a 1920s style death ray for that matter :D) but perhaps you have something that I don’t know the word for. It wasn’t that long ago I didn’t know what a comforter was, for instance, and yet I consider a soldering iron a household instrument. Go figure.

Very probably there will not be any confusing items that I will be unable to identify. I trust you to choose things that the average person is familiar with. My concern was that if there should happen to be one that’s not, then a description would have to suffice.

[glee]

*Originally posted by Paco *

1. Sunglasses
2. calculator
3. mug
4. mittens
5. pencil sharpener
6. physics book
7. toothbrush
8. notebook
9. hat
10. old phone
11. cd case
12. cow

I claim no psychic ability, but I did sleep at a Holiday Inn Express last night.

Thank you!
(P.S. I’m going to assume you mean a model cow. My desk is not that big - or sturdy.)

I have an strong urge to say how well these control predictions are going. Fortunately, for the experiment’s integrity I can resist!

Paco, your prediction number is 2.*

*this has no meaning, but I wanted to make you feel welcome!

[glee]

*Originally posted by cityboy916 *
It’s not so much the quantum recursive analyzer thing…

Ah, you know about that!

*Originally posted by cityboy916 *
… (or even a 1920s style death ray for that matter… :D)

I don’t know if the death ray is covered by UK gun legislation. But I ain’t taking no chances. If I have one, it’s not in plain view!!

*Originally posted by cityboy916 *
…but perhaps you have something that I don’t know the word for. It wasn’t that long ago I didn’t know what a comforter was, for instance, and yet I consider a soldering iron a household instrument. Go figure.

Very probably there will not be any confusing items that I will be unable to identify. I trust you to choose things that the average person is familiar with. My concern was that if there should happen to be one that’s not, then a description would have to suffice.

Well I think all my objects are ‘things that the average person is familiar with’.
But this is just a start. You describe them, and we’re getting somewhere!

[glee]

*Originally posted by IWLN *
That was Spock??? I thought it was Buzz. Did the dog chew on it?

How on earth do you know (DOWN, SPOT!!) I have a dog? :eek:

[glee]

David B,

can you confirm you have received my e-mail with the list of 12 objects, plus the chosen one?

Ta.

[Shade]

*Originally posted by glee *
I have an strong urge to say how well these control predictions are going.

Ooh, then I deduce that they’re going exceptionally well. Or exceptionally badly.

[IWLN]

*Originally posted by glee *
**How on earth do you know (DOWN, SPOT!!) I have a dog? :eek: **

I sent you one of those new breed of IPU dogs. Guess what color MisSpot is? That’s not a remote camera hanging around her neck. Really, it’s a flea tag.

[Shalmanese]

I light of that, I propose the following modified test:

Split the objects into different categories, for example:

Famous people, Soft drink brands, Colours, shapes, animals etc.

From each category, choose 10 or so different items which are:

1. easily recognisable
2. easily distinguishable
3. easily obtainable

Post ALL choices onto the board. Categories would have to be approved by cityboy and glee. If possible, post an example of all objects used so there is no confusion.

AFTER this step is done, devise some random method of choosing one object from each step. I would prefer it to be software based and not merely heuristically random like the book flipping example. Tossing a coin or rolling a die would also be acceptable if you describe the algorithm beforehand. Once that is done, place the items chosen onto the table and await results. Some enterprising statistician could then post the confidence Interval for cityboy being a psychic.

As for cityboy, I am trying to make this as easy for you as possible while still setting up a decent amount of odds. It’s far easier to pick an object out of a group than to remeber the object afterwards. You can also say that you were uncertain about an object or say it could be X or Y but keep in mind that this will reduce the confidence interval.

As a suggested guide:

Categories:

Shapes & Colours (combined), stylised cartoon picture of an animal, Famous shot of a celebrity, religious symbol, sporting equipment, cartoon picture of musical instrument, Letter of the alphabet.

shapes:

1. Square
2. Circle
3. Triangle
4. Kite
5. Trapezium
6. Arrow

Colours:

1. Red
2. Green
3. Blue
4. Yellow
5. Black
6. Purple

Animal:

1. Dog
2. Cat
3. Fish
4. Cow
5. Chicken
6. Monkey

you get the point…

If you choose 8 categories and 6 options from each category, its an 1 in 1.6million chance of getting them all right from guessing yet I am thinking that it would be trivial if you were shown a photograph of them.

[Musicat]

I think Shalmanese has a very, very good point. A pre-published list of possible objects removes the ambiguity from a subjective evaluation. Either the object was seen or it wasn’t. The tester can agree in advance that he will be able to recognize each proffered object, and if he has any doubt, just replace the object with another before any testing begins.

I would even go so far as to say take a picture of each object from several angles and post it on a web site along with the object’s list number. This way, the tester can see what he is expected to be looking for and cannot say, “that doesn’t really look like a coffee cup to me,” or “I only saw the back side and didn’t recognize it.”

Using a randomizer (coin toss, whatever), one object should be chosen from the list that all parties have seen and agreed to. That object should be placed in full view on the desk for an agreed-upon time. At the end of the time, CityBoy916 will email the number of his choice to the moderator (or none, if this is allowed). As soon as it is received, a new object is randomly chosen from the same, complete list and the process repeated.

When a certain number of tests have been completed (the number must be agreed upon in advance), the mod will compare the object numbers for each test and publish the results, along with the mathematical odds of chance alone.

If anything goes wrong, or conditions are requested to be changed at any time during the test, the test shall be null and void, since it was not completed as originally agreed upon, and must be started over.

[Musicat]

*Originally posted by Musicat *
When a certain number of tests have been completed (the number must be agreed upon in advance), the mod will compare the object numbers for each test and publish the results, along with the mathematical odds of chance alone.

Furthermore, before the test begins, the odds of chance alone must be computed, and the minimal value of what constitutes success be mutually agreed upon.

Example: If a list of four objects is used, and four tests are done, by chance alone, one test would be expected to be correct. If three tests or more are correct, that might be sufficient to establish the claim’s veracity.

A statistician would have to tell us how many objects, how many tests and how many correct results would be reasonable to attain a 95% confidence level or better.

Let the games begin!

[umop_ap_sdn]

*Originally posted by Musicat *
**I think Shalmanese has a very, very good point. A pre-published list of possible objects removes the ambiguity from a subjective evaluation. Either the object was seen or it wasn’t.

I would even go so far as to say take a picture of each object from several angles and post it on a web site along with the object’s list number. This way, the tester can see what he is expected to be looking for and cannot say, “that doesn’t really look like a coffee cup to me,” or “I only saw the back side and didn’t recognize it.”**

Well it’s up to glee to decide if this is a good idea or not. If there should be pictures available for me to compare, then I would only use them after the projection has completed, and then only if I have any doubts as to the identity of an object.

Using a randomizer (coin toss, whatever), one object should be chosen from the list that all parties have seen and agreed to. That object should be placed in full view on the desk for an agreed-upon time.

Such an experiment would be a lot easier had I already developed the ability to project at will. That’s why I did not want a time limit. If a time limit were imposed, and I did not succeed in having a projection in that span, then it would be necessary to guess which would contaminate the results with non-astrally obtained data.

If anything goes wrong, or conditions are requested to be changed at any time during the test, the test shall be null and void, since it was not completed as originally agreed upon, and must be started over.

I agree with the test as it currently stands. As long as glee also agrees to the terms and conditions he or she has proposed, I see no reason to change the experiment. Who decides when to commence? Do we proceed as soon as glee and I are both ready, or do we elect an impartial party to supervise?

[IWLN]

*Originally posted by cityboy916 *
I agree with the test as it currently stands. As long as glee also agrees to the terms and conditions he or she has proposed, I see no reason to change the experiment. Who decides when to commence? Do we proceed as soon as glee and I are both ready, or do we elect an impartial party to supervise? **

Darn it cityboy916, you haven’t left your body yet??!! Get crackin’

[glee]

*Originally posted by Paco *

1. Sunglasses
2. calculator
3. mug
4. mittens
5. pencil sharpener
6. physics book
7. toothbrush
8. notebook
9. hat
10. old phone
11. cd case
12. cow

I claim no psychic ability, but I did sleep at a Holiday Inn Express last night.

Oh, Paco, please could you say which of the above objects has the post-it on it?

[glee]

*Originally posted by Shade *
Ooh, then I deduce that they’re going exceptionally well. Or exceptionally badly.

Or are hitting expected chance levels, or have successfully identified some objects I didn’t expect, or have an interestingview of my character, or…

[glee]

*Originally posted by Shalmanese *
I light of that, I propose the following modified test:

Split the objects into different categories, for example:

Famous people, Soft drink brands, Colours, shapes, animals etc.

From each category, choose 10 or so different items which are:

1. easily recognisable
2. easily distinguishable
3. easily obtainable

Post ALL choices onto the board. Categories would have to be approved by cityboy and glee. If possible, post an example of all objects used so there is no confusion.

AFTER this step is done, devise some random method of choosing one object from each step. I would prefer it to be software based and not merely heuristically random like the book flipping example. Tossing a coin or rolling a die would also be acceptable if you describe the algorithm beforehand. Once that is done, place the items chosen onto the table and await results. Some enterprising statistician could then post the confidence Interval for cityboy being a psychic.

As for cityboy, I am trying to make this as easy for you as possible while still setting up a decent amount of odds. It’s far easier to pick an object out of a group than to remember the object afterwards. You can also say that you were uncertain about an object or say it could be X or Y but keep in mind that this will reduce the confidence interval.

As a suggested guide:

Categories:

Shapes & Colours (combined), stylised cartoon picture of an animal, Famous shot of a celebrity, religious symbol, sporting equipment, cartoon picture of musical instrument, Letter of the alphabet.

shapes:

1. Square
2. Circle
3. Triangle
4. Kite
5. Trapezium
6. Arrow

Colours:

1. Red
2. Green
3. Blue
4. Yellow
5. Black
6. Purple

Animal:

1. Dog
2. Cat
3. Fish
4. Cow
5. Chicken
6. Monkey

you get the point…

If you choose 8 categories and 6 options from each category, its an 1 in 1.6million chance of getting them all right from guessing yet I am thinking that it would be trivial if you were shown a photograph of them.

I think this is good scientific stuff.
However, although it is possible to calculate exactly the chances of achieving the above by chance, there is (sadly) a big weakness elsewhere - you can’t have that much confidence in people you know only as glee, cityboy916 (and even David B!).
As it happens, I am pretty sure all of us are honestly doing our best.
But I think your test is ‘too good’ for these anonymous conditions.
I am sure that if / when this stuff reaches Randi, he will use a quality test like the one you propose.

[glee]

*Originally posted by Musicat *
I think Shalmanese has a very, very good point. A pre-published list of possible objects removes the ambiguity from a subjective evaluation. Either the object was seen or it wasn’t. The tester can agree in advance that he will be able to recognize each proffered object, and if he has any doubt, just replace the object with another before any testing begins.

Using a randomizer (coin toss, whatever), one object should be chosen from the list that all parties have seen and agreed to. That object should be placed in full view on the desk for an agreed-upon time. At the end of the time, CityBoy916 will email the number of his choice to the moderator (or none, if this is allowed). As soon as it is received, a new object is randomly chosen from the same, complete list and the process repeated.

When a certain number of tests have been completed (the number must be agreed upon in advance), the mod will compare the object numbers for each test and publish the results, along with the mathematical odds of chance alone.

If anything goes wrong, or conditions are requested to be changed at any time during the test, the test shall be null and void, since it was not completed as originally agreed upon, and must be started over.

Some comments on your worthy analysis:

• I will using my trusty roleplaying d12 as the ‘randomiser’
• the second test will indeed be a list of objects (and I will welcome your help in ensuring they are distinguishable)
• the number of tests we do depends on cityboy916 (he’s doing all the work, I’m just leaving stuff on my desk!)
• as previously, we can calculate the precise odds of guessing, but not that somebody anonymous is cheating

[glee]

*Originally posted by cityboy916 *
I agree with the test as it currently stands. As long as glee also agrees to the terms and conditions he or she has proposed, I see no reason to change the experiment. Who decides when to commence? Do we proceed as soon as glee and I are both ready, or do we elect an impartial party to supervise?

Technically we should wait until David B confirms he has received my e-mail (sent Dec 4th).

But I don’t see any harm if you start viewing straightaway, but don’t reveal anything you see until David B confirms he has the answers.

Oh, and if another Mod wants to referee, I’ll send them the e-mail too.

[Musicat]

Our overriding goal should be to separate chance from (paranormal) skill.

*Originally posted by cityboy916 *
Well it’s up to glee to decide if this is a good idea or not. If there should be pictures available for me to compare, then I would only use them after the projection has completed, and then only if I have any doubts as to the identity of an object.

Whether you would use such pictures before, after or not at all is up to you. But in either case, if pictures have been supplied to you in advance of the test, and you fail, you cannot say you didn’t know what the object looked like.

The purpose of a picture and/or unambiguous description in advance is to assist you and increase the accuracy of your viewing. If you are actually astrally projecting, this will help. If you are a fraud, it will not hinder. If you are not able to perform, it will not be because there was a problem identifying an object, but because your claim is likely false. Nothing will be disguised, nothing hidden. Surely if you can view remotely as you claim, such a simple test will be a piece of cake (hey, there’s another possible object for the list ).

Such an experiment would be a lot easier had I already developed the ability to project at will. That’s why I did not want a time limit. If a time limit were imposed, and I did not succeed in having a projection in that span, then it would be necessary to guess which would contaminate the results with non-astrally obtained data.

I wasn’t necessarily suggesting a short time limit. But surely we need some kind of time limit or this experiment could go on forever. Would a week for each round be OK?

How about the option of not entering an object number at all for any test, like a “pass”. Of course, that would alter the odds, so we would have to agree on how many skipped rounds would be acceptable. A statistician would be useful here.

I agree with the test as it currently stands. As long as glee also agrees to the terms and conditions he or she has proposed, I see no reason to change the experiment.

Well, this is Glee’s thread and test, so he is in charge, not me. You’re welcome to ignore my advice, at your peril.

[IWLN]

*Originally posted by glee *
**Some comments on your worthy analysis:

• I will using my trusty roleplaying d12 as the ‘randomiser’
• the second test will indeed be a list of objects (and I will welcome your help in ensuring they are distinguishable)
• the number of tests we do depends on cityboy916 (he’s doing all the work, I’m just leaving stuff on my desk!)
• as previously, we can calculate the precise odds of guessing, but not that somebody anonymous is cheating **

Since we’re talking about cityboy916 actually viewing the desk as opposed to some sort of psychic type vague “sense” of what they’re seeing, it seems like this is fine. How’s late Friday night? Does that work for you.

[umop_ap_sdn]

*Originally posted by Musicat *
**Our overriding goal should be to separate chance from (paranormal) skill. **

Agreed. In other words, increase the signal to noise ratio (signal = evidence of actual astral projection and noise = chance lucky guesses).

The purpose of a picture and/or unambiguous description in advance is to assist you and increase the accuracy of your viewing. If you are actually astrally projecting, this will help. If you are a fraud, it will not hinder.

If I am actually projecting, it will not hinder; if I am a fraud, it will help.

I wasn’t necessarily suggesting a short time limit. But surely we need some kind of time limit or this experiment could go on forever. Would a week for each round be OK?

If the time limit is to be one week, then I’ll get back to ya when I can project at will. I know it sounds ridiculous to say this, but I have yet to learn how to control this. Was it silly of me to agree so quickly? Maybe. But I’m not going to chicken out of it now. Actually there is a time limit, since I will not live forever, but for all I know that could take 60 years and even I don’t propose waiting that long.

How about the option of not entering an object number at all for any test, like a “pass”. Of course, that would alter the odds, so we would have to agree on how many skipped rounds would be acceptable.

Usually I’d say keep it simple, and “pass” rounds would only complicate things, but if the rounds are one week apiece there probably will be skipped rounds at first. The same difficulties would apply if I were challenged to run a mile in 10 minutes or stay underwater for an unusually long time.