Beth Clarkson's application
- Thread starter roger
- Start date
-
- Tags
- application
roger
Penultimate Amazing
- Joined
- May 22, 2002
- Messages
- 11,466
Jack, that's great, exactly what we need. However, I can think of some stuff that needs to be controlled.
Convection currents: As the candle burns, it generates heat. this will cause convection currents, which can result in fairly complex, and perhaps chaotic (I'm using the precise mathematical term) behavior.
Height of the candle: As the candle burns, it will pool hot wax, and it will get shorter. The height will affect the measurements done by the camera, and while I have no idea of the exact effect of the hot wax, I bet it can influence the flame behavior.
quality of the candle: I can hypothesize that if the wick is not exactly centered in the candle, it could influence the behavior of the flame.
vibration: obviously the apparatus will need to be shielded from any vibration.
02 levels: if the candle is enclosed in a box, it may consume part of the oxygen, which could affect flame height, and thus movement. Ventilation sufficient to maintain normal O2 levels near the flame while not causing airflow could be tricky.
I could conceive of any of these to cause a false positive or negative in the tests, if she is only getting very subtle movements in the flame.
The first thing I would consider is replacing the candle with a more consistant and controllable flame source. Once instrumented with Jack's apparatus, you could determine if you have controlled for all of the variables with some lengthy runs once everything has reached thermal equilibrium.
Convection currents: As the candle burns, it generates heat. this will cause convection currents, which can result in fairly complex, and perhaps chaotic (I'm using the precise mathematical term) behavior.
Height of the candle: As the candle burns, it will pool hot wax, and it will get shorter. The height will affect the measurements done by the camera, and while I have no idea of the exact effect of the hot wax, I bet it can influence the flame behavior.
quality of the candle: I can hypothesize that if the wick is not exactly centered in the candle, it could influence the behavior of the flame.
vibration: obviously the apparatus will need to be shielded from any vibration.
02 levels: if the candle is enclosed in a box, it may consume part of the oxygen, which could affect flame height, and thus movement. Ventilation sufficient to maintain normal O2 levels near the flame while not causing airflow could be tricky.
I could conceive of any of these to cause a false positive or negative in the tests, if she is only getting very subtle movements in the flame.
The first thing I would consider is replacing the candle with a more consistant and controllable flame source. Once instrumented with Jack's apparatus, you could determine if you have controlled for all of the variables with some lengthy runs once everything has reached thermal equilibrium.
Hm, for a more controllable source, what about something as simple as a cigarette lighter? Their flames seem to be more stable.
I wonder if this woman randomizes her attempts - if the candle tends to blow left and the target object is on the left...it seems (and I think this was suggested before in a more clever way) that randomly defining the target (e.g. left then left then right then left...) even if the target is binary, would be required.
I wonder if this woman randomizes her attempts - if the candle tends to blow left and the target object is on the left...it seems (and I think this was suggested before in a more clever way) that randomly defining the target (e.g. left then left then right then left...) even if the target is binary, would be required.
I'm going to suggest that a candle flame is too eratic to measure. And as has been pointed out already, it is really just a measure of minute and chaotic air currents anyway, which the flame both mixes with and affects in turn.
Instead, I would suggest that if she can understand statictics, she will be able to understand the concept of air currents as well. AIn which case she should also be able to understand that her ability should easily work on other easily moved objects that might be far more easily measured and produce unequivocal results.
For example, the density of tiny particles suspended in air (think dust or talcum powder or steam) can be measured using such devices as lasers (scattering, scatterometers??). Her ability should be able to change an even density space such that it becomes more/less dense at one place. Multiple measuring locations should reveal this consistently.
The simplest system I can think of is a container with laser "density measuring" devices around the sides (not top to bottom). The container is then filled evenly with suspended particles (I suggest steam - water particles), and baseline measurements taken over a period of time. For the applicant run, they have to "try to push the particles towards one side only". Repeat as often as statistically necessary. Compare with the baseline. Result is obvious, one way or the other.
Instead, I would suggest that if she can understand statictics, she will be able to understand the concept of air currents as well. AIn which case she should also be able to understand that her ability should easily work on other easily moved objects that might be far more easily measured and produce unequivocal results.
For example, the density of tiny particles suspended in air (think dust or talcum powder or steam) can be measured using such devices as lasers (scattering, scatterometers??). Her ability should be able to change an even density space such that it becomes more/less dense at one place. Multiple measuring locations should reveal this consistently.
The simplest system I can think of is a container with laser "density measuring" devices around the sides (not top to bottom). The container is then filled evenly with suspended particles (I suggest steam - water particles), and baseline measurements taken over a period of time. For the applicant run, they have to "try to push the particles towards one side only". Repeat as often as statistically necessary. Compare with the baseline. Result is obvious, one way or the other.
Many replies here suggest protocols that include isolating the target. If I tried to reach out and move a target that was behind a clear barrier... I would fail. I fear that the claimant will be wary of the same thing. That the barrier will block "whatever" it is that she is projecting.
I think this is a likely objection... so might as well discuss it before we get ahead of ourselves.
I think this is a likely objection... so might as well discuss it before we get ahead of ourselves.
Regarding Jackmott's test setup, I think that only the software solution will be acceptable. The challenge rules state that no judging will be allowed.
If the software solution cannot be done for one or the other reason, you could consider putting four weights suspended in fine threads around the candle so close that the heat will eventually cause the thread to melt, if the candle is moved consistently in that direction.
If the software solution cannot be done for one or the other reason, you could consider putting four weights suspended in fine threads around the candle so close that the heat will eventually cause the thread to melt, if the candle is moved consistently in that direction.
Well if she has a real problem about this then she will have to wear a cover over her mouth and nose.
I can't see the challenge going ahead if she is allowed close unprotected access to the flame.
Maybe something like this, but with cling-film over the holes:
As a simple, preliminary test you might try having three people judge which of three or four candles she was concentrating on. Simply put up a screen between her and them so that they can see the candles but not her, one candle is randomly picked for her to influence, the three observers are then asked to pick which of the candles is being influenced. She could do this herself in private with the folks she's comfortable with who already know what to look for.
No "judging"...
There can be no "judging".
The test protocol must be of such a design that there is no room for interpretation or "judging". No other protocol is acceptable.
The results must be entirely self-evident. Period.
There can be no "judging".
The test protocol must be of such a design that there is no room for interpretation or "judging". No other protocol is acceptable.
The results must be entirely self-evident. Period.
IXP
Graduate Poster
- Joined
- Oct 19, 2004
- Messages
- 1,395
The claimant's setup of a ring of wax suspended over the candle is superior to any other methods proposed here. If the candle rests on a paper target divided into pie-sliced sections, and each trial uses a new target, you have a permanent record in the form of a paper target with one drop of wax for each trial. For drops that are on the line, you simply throw out that trial. The results are unambiguous and require no judgement, either the drop is in the section the clamant was trying to push it or it is not.
I don't think it is necesary, but you could also account for air currents by doing both an influenced and non-influenced trial for each randomly selected target, then doing the statistics based solely on how much more ofen the influenced trials land in the target regioin vs. the un-influenced ones.
I don't think it is necesary, but you could also account for air currents by doing both an influenced and non-influenced trial for each randomly selected target, then doing the statistics based solely on how much more ofen the influenced trials land in the target regioin vs. the un-influenced ones.
Re: No "judging"...
But the judges would not be judging the results. The results would be based on whether they picked the right candle or not. Either they were able to discerne which candle was being influenced or they were not. This is a test that Beth could do herself with the people that believe she can it and would save the embarrassment of a public test.
KRAMER said:
But the judges would not be judging the results. The results would be based on whether they picked the right candle or not. Either they were able to discerne which candle was being influenced or they were not. This is a test that Beth could do herself with the people that believe she can it and would save the embarrassment of a public test.
IXP
Graduate Poster
- Joined
- Oct 19, 2004
- Messages
- 1,395
Re: New Protocol
I am beginning to see how difficult it is to set up a protocol with a claimant.
This woman is clearly intelligent and articulate, and yet her lengthy description of the setup she has used provides more questions than answers.
How is the target determined? Is it always the same or does it vary with each trial? These clearly need to be randomized to eliminate a systematic error.
What are the meanings of the two lists of 4 numbers? If I had to guess, they are angles measured from the target angle with and without influence. But the ones without influence are decidedly non-random, they are already too close to the target (some should be > 180 degrees certainly).
Since the apparatus is changing as the candle burns, the order in which the control trials and influenced trials are done is important and should be randomized as well, are they?
Are the wax disks used more than once? They should not be since they are altered by the experiment and no longer symmetric.
KRAMER said:
I am beginning to see how difficult it is to set up a protocol with a claimant.
This woman is clearly intelligent and articulate, and yet her lengthy description of the setup she has used provides more questions than answers.
How is the target determined? Is it always the same or does it vary with each trial? These clearly need to be randomized to eliminate a systematic error.
What are the meanings of the two lists of 4 numbers? If I had to guess, they are angles measured from the target angle with and without influence. But the ones without influence are decidedly non-random, they are already too close to the target (some should be > 180 degrees certainly).
Since the apparatus is changing as the candle burns, the order in which the control trials and influenced trials are done is important and should be randomized as well, are they?
Are the wax disks used more than once? They should not be since they are altered by the experiment and no longer symmetric.
Re: Re: New Protocol
degrees is the maximum that the angle can possible be, since
if it were more than 180 degrees around to the left, it would be
less than 180 degrees around to the right. (Think about the angle between 12 o'clock and the hour hand. At 3pm, the angle is 90 degrees, but at 9, it's also 90 degrees.)
If you do the average of her four "control" conditions, it's "about" ninety degrees, which is where it should be if the wax is really randomly melting. On the other hand, her average for the experimental setup is a little over 20, which is not at all in line with the randomness assumption.
Similarly, she stated that she did not reuse the wax disks : "I use 4 rings for the experiment and 4 for the control." She also states that she randomized the trial order : "The experimental trials were randomized with the controls, so there was no time effect."
According to her protocol, she always sets the target at 0 degrees. This is open to interpretation; this could mean that she picks different target directions, and redefines 0 on each trial, or it could mean that she always uses the same target/direction, which I agree would introduce the possibilty of systematic error or even outright fraud. Similarly, you need to have a trusted (blinded) third party set up the equipment and take the angle measurements, but that's obvious. (There's some interesting cut-and-paste effects in her writing, sometimes she refers to herself as "I", sometimes as "the subject," and sometimes she refers to the person setting the experiment -- who is not the subject -- as "I." As I said, interesting.)
I'd be more concerned about the physical setup. Although she says that "the subject is not deliberately attempting to alter the heat of the air from the flame with their breath," I think we all know what that particular statement is worth. She's trying to control for air currents/conditions already, so I definitely think that strengthening those controls would be a good idea. Can you put the entire setup in a hurricane lamp chimney or something?
Personally, I like and am happy with the idea of running a t-test (although I can consult with some "real" stats geeks about any other tests that would be appropriate to run), but that also violates the spirit, if not the letter, of the JREF's "the test will be self-evident and no judging will be required." One possibility would be to put a piece of paper underneath the wax, with a 90 degree "target" colored on it (45 degrees to either side of "target"). By random chance, a drop of wax would fall on the colored bit one experiment out of four. If she could hit the colored bit five times in a row, that would be the necessary 1:1000 shot. Ten times in a row would give 1:1000000.
I mention this because, in her experimental setup at home, she's never missed by as much as 45 degrees. So the experiment she ran, she was four-for-four.
Um, no. She's measuring the angle between the target and "the place on the ring where the wax first begins to melt." 180IXP said:
degrees is the maximum that the angle can possible be, since
if it were more than 180 degrees around to the left, it would be
less than 180 degrees around to the right. (Think about the angle between 12 o'clock and the hour hand. At 3pm, the angle is 90 degrees, but at 9, it's also 90 degrees.)
If you do the average of her four "control" conditions, it's "about" ninety degrees, which is where it should be if the wax is really randomly melting. On the other hand, her average for the experimental setup is a little over 20, which is not at all in line with the randomness assumption.
Similarly, she stated that she did not reuse the wax disks : "I use 4 rings for the experiment and 4 for the control." She also states that she randomized the trial order : "The experimental trials were randomized with the controls, so there was no time effect."
According to her protocol, she always sets the target at 0 degrees. This is open to interpretation; this could mean that she picks different target directions, and redefines 0 on each trial, or it could mean that she always uses the same target/direction, which I agree would introduce the possibilty of systematic error or even outright fraud. Similarly, you need to have a trusted (blinded) third party set up the equipment and take the angle measurements, but that's obvious. (There's some interesting cut-and-paste effects in her writing, sometimes she refers to herself as "I", sometimes as "the subject," and sometimes she refers to the person setting the experiment -- who is not the subject -- as "I." As I said, interesting.)
I'd be more concerned about the physical setup. Although she says that "the subject is not deliberately attempting to alter the heat of the air from the flame with their breath," I think we all know what that particular statement is worth. She's trying to control for air currents/conditions already, so I definitely think that strengthening those controls would be a good idea. Can you put the entire setup in a hurricane lamp chimney or something?
Personally, I like and am happy with the idea of running a t-test (although I can consult with some "real" stats geeks about any other tests that would be appropriate to run), but that also violates the spirit, if not the letter, of the JREF's "the test will be self-evident and no judging will be required." One possibility would be to put a piece of paper underneath the wax, with a 90 degree "target" colored on it (45 degrees to either side of "target"). By random chance, a drop of wax would fall on the colored bit one experiment out of four. If she could hit the colored bit five times in a row, that would be the necessary 1:1000 shot. Ten times in a row would give 1:1000000.
I mention this because, in her experimental setup at home, she's never missed by as much as 45 degrees. So the experiment she ran, she was four-for-four.
IXP
Graduate Poster
- Joined
- Oct 19, 2004
- Messages
- 1,395
New Drkitten,
Thank you for correcting the gross errors in my post. Of course the max angle is 180. Honestly, I was thinking of measuring the angle on an absolute scale of 0 to 360, but the difference from the target makes much more sense and is likely what she is doing. However, the control numbers are still suspicious. Take a look at them:
Control: 77, 67, 83, 105
Test: 10, 25, 19, 38,
They should vary from 0 to 180 uniformly if there is no preferred direction, but they are way too tightly clustered around the mean (which should be 90). I know you cannot draw any conclusion from so few trials, but I would have looked for a large bias given these number for the controls.
She even states that the control numbers follow a normal distirbution which would not be expected unless there were a preferred direction (it should be uniform distribution) So the direction of flame is biased in the first place, not a good starting point, but her numbers do still show a shift in bias.
If indeed these results are true, I wouild guess the breath of the subject is the cause of the shift in bias. Randomized target directions would certainly improve this experiment.
Thank you for correcting the gross errors in my post. Of course the max angle is 180. Honestly, I was thinking of measuring the angle on an absolute scale of 0 to 360, but the difference from the target makes much more sense and is likely what she is doing. However, the control numbers are still suspicious. Take a look at them:
Control: 77, 67, 83, 105
Test: 10, 25, 19, 38,
They should vary from 0 to 180 uniformly if there is no preferred direction, but they are way too tightly clustered around the mean (which should be 90). I know you cannot draw any conclusion from so few trials, but I would have looked for a large bias given these number for the controls.
She even states that the control numbers follow a normal distirbution which would not be expected unless there were a preferred direction (it should be uniform distribution) So the direction of flame is biased in the first place, not a good starting point, but her numbers do still show a shift in bias.
If indeed these results are true, I wouild guess the breath of the subject is the cause of the shift in bias. Randomized target directions would certainly improve this experiment.
roger
Penultimate Amazing
- Joined
- May 22, 2002
- Messages
- 11,466
I see many potential issues with this setup:
Is the tumbler of uniform thickness? To what degree of accuracy?
Is the tumbler round, square, hexagonal, etc?
Is it oriented exactly the same relative to the 0 degree target each time?
To what degree of accuracy is the candle "centered"?
How uniform is each candle in circumference?
How uniform is each candle in height?
How uniform is the thickness of the wick for each candle?
Is the wick centered in the candle?
Does it bend?
How is the candle lit?
Is it done from the same side/angle each time?
Is it lit before or after the wax ring is placed on the tumbler?
Was the tumbler either cleaned or replaced between each trial?
Does the person How do you ensure that the candle is truly vertical in the tumbler (a 1deg tilt is not visible to the eye).
Of each of these, have you measured the results they have on the apparatus?
How is this wax ring manufactured?
What are your QC controls for the ring?
How uniform is the thickness?
How uniform is the hole in the center?
Please define "centered nicely" in terms of measurements and statistical deviations.
Have you run tests that show if any of these questions influence where the disk melts?
The first drip?
The first 'soft spot?
How is 'judging' elimanated from this step (a requirement of the challenge)?
Who does this, the subject?
Is the person who does this aware of what the 0 point is?
Is the person who does this aware of whether this was a control or real trial?
How is the subject isolated thermally and atmospherically from the candle?
How is the subject isolated vibrationally from the candle? (an unintentional foot tap on the floor can be enough to vibrate a candle flame on a table top)
The wording of this is vague - is the glass moved after the candle is lit, or is this merely descriptive?
The subject sets up the apparatus - this is bad!
Was the tumbler and other experimental apparatus (including the table the tumbler sat on) allowed to return to thermal equilibrium between each trial?
Is the experiment otherwise run identically when it is a control vs real trial?
Apparently the author does not recognize/understand the existence of convection currents caused by the heat of the candle, and which could be unconsciously manipulated by body position alone of the subject, let alone by their breath.
This is not clear. Is it the same position each time, or is it randomized?
Is it chosen before or after the candle is lit, and before or after the apparatus is set up (candle in tumbler, disk on top)?
Does the person who sets up the apparatus know where the target position will be?
Is this target position in any way consistant when compared to the orientation of the room?
It's nice that deliberate fraud is not taking place, but as we know, many of these effects are quite unconscious. (dowsing - ideomotor effect, for example).
Also, this statement implies that the subject could influence the candle, by not ruling it out, and, we can conclude, probably does influence it.
No experimental evidence provided for this assertion.
I'll tell you what, give me this setup and 3 weeks to practice, and I will win the $1M under these conditions. I am pretty confident that I could cheat my way through this set up, so long as Randi wasn't there to ask the same kinds of questions that I posed above.
There are just too many independent, uncontrolled variables to conclude that the movement is not due to normal physical processes.
However, with all that, this design has a serious flaw - it does not measure flame movement. It measures heat convection - which is not necessarily the same thing at all. First you you would have to prove that this apparatus has a high correlation between flame movement and heat convection.
edited to add: When I say I could cheat my way through it, I am in no way suggesting that Beth Clarkson is cheating. I suspect faulty experimental design combined with unconscious factors.
Is the tumbler of uniform thickness? To what degree of accuracy?
Is the tumbler round, square, hexagonal, etc?
Is it oriented exactly the same relative to the 0 degree target each time?
To what degree of accuracy is the candle "centered"?
How uniform is each candle in circumference?
How uniform is each candle in height?
How uniform is the thickness of the wick for each candle?
Is the wick centered in the candle?
Does it bend?
How is the candle lit?
Is it done from the same side/angle each time?
Is it lit before or after the wax ring is placed on the tumbler?
Was the tumbler either cleaned or replaced between each trial?
Does the person How do you ensure that the candle is truly vertical in the tumbler (a 1deg tilt is not visible to the eye).
Of each of these, have you measured the results they have on the apparatus?
How is this wax ring manufactured?
What are your QC controls for the ring?
How uniform is the thickness?
How uniform is the hole in the center?
Please define "centered nicely" in terms of measurements and statistical deviations.
Have you run tests that show if any of these questions influence where the disk melts?
How?
The first drip?
The first 'soft spot?
How is 'judging' elimanated from this step (a requirement of the challenge)?
Who does this, the subject?
Is the person who does this aware of what the 0 point is?
Is the person who does this aware of whether this was a control or real trial?
How is the subject isolated thermally and atmospherically from the candle?
How is the subject isolated vibrationally from the candle? (an unintentional foot tap on the floor can be enough to vibrate a candle flame on a table top)
The wording of this is vague - is the glass moved after the candle is lit, or is this merely descriptive?
The subject sets up the apparatus - this is bad!
Was the tumbler and other experimental apparatus (including the table the tumbler sat on) allowed to return to thermal equilibrium between each trial?
Is the experiment otherwise run identically when it is a control vs real trial?
Apparently the author does not recognize/understand the existence of convection currents caused by the heat of the candle, and which could be unconsciously manipulated by body position alone of the subject, let alone by their breath.
This is not clear. Is it the same position each time, or is it randomized?
Is it chosen before or after the candle is lit, and before or after the apparatus is set up (candle in tumbler, disk on top)?
Does the person who sets up the apparatus know where the target position will be?
Is this target position in any way consistant when compared to the orientation of the room?
It's nice that deliberate fraud is not taking place, but as we know, many of these effects are quite unconscious. (dowsing - ideomotor effect, for example).
Also, this statement implies that the subject could influence the candle, by not ruling it out, and, we can conclude, probably does influence it.
No experimental evidence provided for this assertion.
I'll tell you what, give me this setup and 3 weeks to practice, and I will win the $1M under these conditions. I am pretty confident that I could cheat my way through this set up, so long as Randi wasn't there to ask the same kinds of questions that I posed above.
There are just too many independent, uncontrolled variables to conclude that the movement is not due to normal physical processes.
However, with all that, this design has a serious flaw - it does not measure flame movement. It measures heat convection - which is not necessarily the same thing at all. First you you would have to prove that this apparatus has a high correlation between flame movement and heat convection.
edited to add: When I say I could cheat my way through it, I am in no way suggesting that Beth Clarkson is cheating. I suspect faulty experimental design combined with unconscious factors.
IXP
Graduate Poster
- Joined
- Oct 19, 2004
- Messages
- 1,395
roger said:
I agree, if all of the questions you raise are left unanswered. That is why the protocols are unlikely to be accepted as proposed by the claimant.
I think this setup would be a reasonble preliminary test if the claimant is isolated from the device by being in a different room and the targets are randomly chosen after the entire apparatus is set up for each trial. I seriously doubt that she could perform under those circumstances.
As far as the position of the flame vs. heat convection: influencing either by telekenesis would be equally valid as a paranormal claim. The claim should be changed from "I can influence the position of a flame," to "I can influence the position of the first drip of wax from a ring of wax postioned over a candle flame."
I think this is the most interesting claim we have seen in a while, and I hope it proceeds to testing. I would really like to know just why this person is getting, or thinks she is getting positive results. If I could work with her, I would first let her demonstrate this under the conditions that she is using. If she succeeds, then I would eliminate, one at a time, each possible means by which the the results might be achieved until the effect melts away.