Sheldrake tests telephone telepathy

[Paul C. Anagnostopoulos]

Hey, I've got an idea! Who wants to try to replicate these experiments? Six or seven of us could conduct one or two trials a day over the next four months and present a paper at TAM2.

~~ Paul

 

[Darat]

Darat asked:
Sheldrake always seems to make everything complicated and variable.


Hold on there! What do you mean? Are you saying the rings on her telephone are different from normal rings?

...snip...
~~ Paul

It appears from what she says that when her phone rings from a call on a non-mobile phone the pattern of the rings starts at a different point in the standard pattern:

e.g.

landline

ring, ring, ring.......... ring, ring, ring.........

my mobile

ring,ring.......... ring,ring,ring..........

So it gives her a pretty big clue (along with my semi-regular phoning times) that it is me calling.

Not saying of course this happens in these experiments but it would be very simple to remove any chance of "phone cues", suprised he didn't.

 

[Darat]

Hey, I've got an idea! Who wants to try to replicate these experiments? Six or seven of us could conduct one or two trials a day over the next four months and present a paper at TAM2.

~~ Paul

Happy to join in.

 

[CFLarsen]

I'm in.

The paper will, of course, also be published in SkepticReport.

 

[MRC_Hans]

*Snip*
Not saying of course this happens in these experiments but it would be very simple to remove any chance of "phone cues", suprised he didn't.

I'm afraid it is not surprising at all. They evidently did not try very hard to eliminate errors; just note the complete lack of blinding.

I think we are forced to seek the explanation in this fact: The experimenters believe in telepathy and set out to prove it. They specifically recruited testers that believed they had telephatic abilities.

Seing that the protocol leaves open several possibilities of cheating, deliberate or not, it is not very surprising that those people found what they were looking for.

Great idea, Paul. Being overseas, I cannot participate, but it could be interesting. Be sure to design a better protocol, though

Hans

 

[CFLarsen]

Being overseas, I cannot participate

Why not? I'm in Denmark. Sheldrake also used overseas callers.

(Oh, and, Hans...it's Paulie the Greek who is "overseas".... not you! )

 

[Interesting Ian]

Forgive my ignorance of statistics, but isn't the arrangement pseudorandom and not truly random at all? It's not blinded, either.

What arrangement is pseudorandom? Who should be blind to what? Could you outline specifically the faults of the filmed experiments and tell us how they could be improved?

Thanks.

 

[Interesting Ian]

I've read about half of the second paper describing the videotaped experiments. The protocol seems much better here; I haven't found anything obviously silly yet. This paragraph keep nagging at me:
quote:
--------------------------------------------------------------------------------
There was a striking difference between Sue’s performance with familiar and unfamiliar callers. With the two familiar callers, she was right 25 times out of 35 (71%; p = .00000001). With the unfamiliar callers, she was right only 5 times out of 35 (14%), not significantly different from the chance level (see Figure 1). The difference between success rates with familiar and unfamiliar callers was very significant statistically ( p = .000001).
--------------------------------------------------------------------------------

~~ Paul [/B]

I haven't read the 2nd paper yet. But what exactly is wrong with that paragraph? I mean apart from her scores with familar callers being excessively high.

 

[Interesting Ian]

I'm in.

The paper will, of course, also be published in SkepticReport.

Unless it's a positive result of course

 

[Interesting Ian]

I'm afraid it is not surprising at all. They evidently did not try very hard to eliminate errors; just note the complete lack of blinding.

Who should be blind to what?

I think we are forced to seek the explanation in this fact: The experimenters believe in telepathy and set out to prove it. They specifically recruited testers that believed they had telephatic abilities.

The subjects specifically said "I believe I have telepathic powers/abilities"?

Seing that the protocol leaves open several possibilities of cheating, deliberate or not, it is not very surprising that those people found what they were looking for.

Could you name these possibilities for the filmed experiments?

 

[Interesting Ian]

On a serious note, I've related this idea to Schwartz twice now. Either directly or through SG, that it would be advantagous to involve "skeptical consultants" to review experimental designs before one is implemented. If nothing else, it would allow the experimenters to consider and compensate for potential skeptical (so-called) "nit-picking" in advance.

Why should skeptics know more about experimental design than non-skeptics? Don't get me wrong though, I think it's an excellent idea when we consider that if a experiment produces overall statistically significant results, skeptics on the whole are more likely to pick faults with the experimental design afterwards.

But I heard that parapsychologists have been doing this over the last few decades anyway? (ie consulting skeptics in their experimental design)

Of course neither Sheldrake or Schwartz are parapsychologists anyway

 

[MRC_Hans]

Why not? I'm in Denmark. Sheldrake also used overseas callers.

(Oh, and, Hans...it's Paulie the Greek who is "overseas".... not you! )

Let me rephrase more honestly, then: The timezone issue would make it inconvinient for me to participate

Hans

 

[MRC_Hans]

Who should be blind to what?

Test subjects should not know their score till after the experiment. Neither should experimenters. This eliminates a lot of possibilities for cheating.

The subjects specifically said "I believe I have telepathic powers/abilities"?

According to the paper, they advertized for people who thought they had telepathic abilities. So, yes.

Could you name these possibilities for the filmed experiments?

Well, lack of blinding for sure. And biased test subjects.

--- and from another post:

Why should skeptics know more about experimental design than non-skeptics?

Very interesting question. Ideally they should not, but it seems they do. Could it be because correctly designed experiments invariably produce negative results for paranormal phenomenon? Thus those that understand experimental design unavoidably BECOME skeptics.

Don't get me wrong though, I think it's an excellent idea when we consider that if a experiment produces overall statistically significant results, skeptics on the whole are more likely to pick faults with the experimental design afterwards.

I'm sure you are right here. Most people are less likely to critizise an experiment when the result confirm their thinking. However, that is only one more reason to insist on good experimental design.

Hans

 

[CFLarsen]

Unless it's a positive result of course

On the contrary: Then it will get top billing.

Are you implying that I do not allow articles that show positive evidence of a paranormal phenomenon?

 

[Pyrrho]

What arrangement is pseudorandom? Who should be blind to what? Could you outline specifically the faults of the filmed experiments and tell us how they could be improved?

Thanks.

Simple. They are pure BS and should never see the light of day.

 

[Interesting Ian]

On the contrary: Then it will get top billing.

Are you implying that I do not allow articles that show positive evidence of a paranormal phenomenon?

I really have no idea. How on earth should I know? But anyway, do the majority of articles in skepticReport dwell upon positive evidence for paranomal phenomena? Or is it skewed towards reporting negative results?

 

[Interesting Ian]

Originally posted by Interesting Ian
What arrangement is pseudorandom? Who should be blind to what? Could you outline specifically the faults of the filmed experiments and tell us how they could be improved?

Thanks.
--------------------------------------------------------------------------------

Pyrrho
Simple. They are pure BS and should never see the light of day.

You need to give details. If these tests are flawed in numerous ways then I really want to know in which ways. I'd be the first person to acknowledge it if they are. So could you please justify what you have said?

BTW I am not at all suggesting that you're not right. I haven't read the article yet (on the filmed experiments). Just started reading it now.

 

[JamesM]

Forgive my ignorance of statistics, but isn't the arrangement pseudorandom and not truly random at all? It's not blinded, either.

Just to explain to anyone who hasn't read the paper: the callers were chosen by rolling a die.

Pyrrho, would you (or anyone else) mind expanding on these comments? What would be a truly random method? I thought pseudorandom number generation was the best we could do. Also, how does the fact that the set of callers is limited to 4 adversely affect the statistics, compared to say, limiting it to 100 callers? Wouldn't this just change the expected hit-rate from 1/4 to 1/100? Would 50 trials where you had to choose between 100 callers produce more rigorous results than 50 trials where you had to choose between 4?

Love the TAM2 idea, by the way. Although unless I win the lottery, I won't be attending.

 

[MRC_Hans]

Ian, if Paul's experiment shows a positive result, the problem will not be who publishes it, but who gets the million$

BTW, I wonder why Sheldrake et al have not filed for the JREF challenge?

I have so far found two, perhaps minor, faults in paper #2:

- They are comparing the scores for different callers in the same experiment. This is not stistically sound, since the figures for callers are interconnected. Obviously, if the callee guesses wrong, she will place her guess on another potential caller, thus influencing the statistics of that caller. This will make differences appear larger (on average doubling them).

- They claim that the timestamp is "burnt into the film". Since the redording is on videotape, this is obviously wrong. They are implying that the timestamps could not be faked, while obviously that is possible.

Hans

 

[Interesting Ian]

This is amazing! It actually says about the unfilmed experiments:

Nevertheless, in the experiments we have just described we were relying on the honesty of the participants and their callers.

So why were these experiments carried out in the first place if it is possible for the subjects to cheat!??

I still think they might have done the unfilmed experiments for the reason I mentioned yesterday.