Sheldrake tests telephone telepathy

[Paul C. Anagnostopoulos]

Here is a recent paper by Sheldrake and Smart about telephone telepathy:

http://www.sheldrake.org/papers/Telepathy/experiment_tests.html

Any statisticians care to take a look?

~~ Paul

 

[john_v_h]

They have to guess who the caller is before the caller says anything.

There's the problem. The caller may not have said anything but the microphone is on. Ambient sounds picked up by the mic, as well as the characteristic noise generated by the particular phone and line connection could signal the identity of the caller.

Additionally, this study suffers from problems of self-selection and self-reporting. The experimenters recruited persons who believed they had this telepathic ability. There was little attempt to verify the truthfulness of the subject's reports.

 

[Paul C. Anagnostopoulos]

Yup, my jaw dropped all the way to the floor when I read the second paragraph under "Test procedures." You get to pick up the phone before you have to record your guess. Astonishing.

~~ Paul

 

[davidsmith73]

Yup, my jaw dropped all the way to the floor when I read the second paragraph under "Test procedures." You get to pick up the phone before you have to record your guess. Astonishing.

~~ Paul

I have to agree. This is awful experimental design. I just don't understand how Sheldrake could not care about this leakage route. I might e-mail him and ask.

 

[CFLarsen]

Rupert Sheldrake is one of the participants in the experiment?

Guess who Pam Smart is? She's the one who owned the dog, JayTee, who Sheldrake claimed had psychic powers.

It's the same crappy, incredibly incompetent experimental design we also have seen Schwartz delve into. He also participated in the Arizona Experiment he conducted.

Incredible. Simply incredible. Sheldrake tests himself and his friends.

 

[asthmatic camel]

Telepathy or telephony? That is the question.

Regards,

AC.

 

[juninho]

Here is a recent paper by Sheldrake and Smart about telephone telepathy:

http://www.sheldrake.org/papers/Telepathy/experiment_tests.html

Any statisticians care to take a look?

~~ Paul

Ahhhhh! Why do we have so many nutters in my country. Its enough to make you want to emigrate, oh hang-on, I am

 

[Interesting Ian]

Here is a recent paper by Sheldrake and Smart about telephone telepathy:

http://www.sheldrake.org/papers/Telepathy/experiment_tests.html

Any statisticians care to take a look?

~~ Paul

I've just bought this book by Sheldrake called "The sense of being stared at". I'm only on page 15,but there's a chapter on telephone telepathy starting from page 95 (to 110).

 

[MRC_Hans]

The statistics look sound enough, but as already mentioned, there are faults in the procedure. The report does address those error sources, but the dismissals are not always credible.

In the discussion, it is obvious that the experimenters have a number of preconcieved ideas of how telepathy works, and their dismissal of other psi effects as possible reasons for the results betray their belief system.

I would rate this report as interesting, but it is not a sound scientific experiment. Since the experimenters, by their own admission, are biased for a positive result, the experiement should have been double-blinded.

As is often seen, they try to make several experiments at the same time, varying several paramters: Familiar/unfamiliar caller, distance, and also there are changes in protocol.

A more sound protocol would be:

One telephone connection, isolated from the normal system, to rule out unrelated calls.

ALL callers use the same phone. The sequence of callers is determined a priori.

Receiver does not answer call, but simply writes down his/her guess of the caller when the phone rings.

At each test, a random sequence of callers make calls at regular time intervals, e.g. 5 minutes.

Only after completion of the test series is the recorded guesses compared to the sequence of callers.

This protocol would eliminate the fault sources, be much more effective, and cheaper.

Hans

 

[Interesting Ian]

There's the problem. The caller may not have said anything but the microphone is on. Ambient sounds picked up by the mic,

They have to say immediately who the caller is.

as well as the characteristic noise generated by the particular phone and line connection could signal the identity of the caller.

Yes that seems at least remotely possible. I would have thought the 4 possible targets really must use the same phone.

Additionally, this study suffers from problems of self-selection and self-reporting. The experimenters recruited persons who believed they had this telepathic ability.

How is this a problem? One should only pick people who do not believe they have any telepathic ability??

There was little attempt to verify the truthfulness of the subject's reports.

What do you mean? Either they "guess" correctly or don't "guess" correctly. Presumably this is verified. Explain please.

Anyway, I admit I have only read about the first 20% of the article, so better get down to further reading.

 

[Interesting Ian]

Yup, my jaw dropped all the way to the floor when I read the second paragraph under "Test procedures." You get to pick up the phone before you have to record your guess. Astonishing.

~~ Paul

Yes that seems bad.

 

[Interesting Ian]

I have to agree. This is awful experimental design. I just don't understand how Sheldrake could not care about this leakage route. I might e-mail him and ask.

Ah! You think so as well. I was thinking the exact same thing myself! (just read your response now) I was wondering if other non-skeptics would agree with me.

 

[Jeff Corey]

It strikes me that there have been too many cases where researchers in parapsychology studies are also their own subjects. It's even worse when the subjects are friends, colleagues or relatives of the primary researcher. I believe that Jahn at Princeton let one of his lab assistants take the binary generator home and collect her own data, unsupervised. (I might be mistaken and will try to track it down after two more cups of coffee).
Could this be a case of the Mendel's Gardiner Effect?

 

[Interesting Ian]

Actually I might dip into that book now and read the relevant chapter. Just to see if he says anything of any relevance their which isn't mentioned on the web page.

 

[Paul C. Anagnostopoulos]

Ian said:

What do you mean? Either they "guess" correctly or don't "guess" correctly. Presumably this is verified. Explain please.

Sounds like I pick up the phone, state who is calling, and then the caller tells me who they are. When do I write down my guess? And who verifies what I write down?

Toward the end of the paper:

One final possibility for a leakage of information remains. In all trials described in the present paper, the participants picked up the telephone before making their guesses. It is therefore possible that they heard characteristic background noises, electronic hisses or other sounds that enabled them to identify the caller. But in our filmed experiments this possibility was eliminated because the participants made their guesses before they picked up the telephone. If background noises and hisses could explain the results in the unfilmed trials, the positive effect we observed should have disappeared in the filmed trials, but it did not.

In other words: Yes, these experiments are a crock, but we did them another way later. Notice also that the subjects were better at guessing long-distance calls. Gee, I wonder if you can tell a call is long distance by the sound of the line?

Let's check out those later experiments:

http://www.sheldrake.org/papers/Telepathy/calls_video.html

~~ Paul

 

[Interesting Ian]

Hmmmm, just started reading up on the actual experimental tests and there is a huge discrepancy already! It says here they "had to guess who the caller was before they picked up the phone". (emphasis added).

 

[Interesting Ian]

Toward the end of the paper:
In other words: Yes, these experiments are a crock, but we did them another way later. [/B]

Ah right. Sorry, didn't read your post before my post above. Yes I agree with you about them being a "crock".

 

[Interesting Ian]

The book just seems to contradict what that web site said about whether the subjects guess before or after they pick up the phone. Yet it seems to be the same set of experiments. In appendix B it states: (using voice recognition utility for this! LOL)

my research associate Pam smart and I recruited subjects through advertisements in local newspapers or through Internet recruitment agencies. We paid subjects for taking part (usually at a rate of £10 per trial). They received the same payment whether or not their guesses were right. We asked people to take part in an initial series of 10 trials, and some subjects were then asked to take part in further (sic) series of 10 trials.

At a prearranged time, the subject received a call from one of these four people. Before answering the phone (emphasis added) he or she had to guess who was calling. In trials that were not videotaped, the caller answered the phone by saying "hello, [name]" before the other person had spoken. The caller reported immediately afterwards what this guess had been. In videotaped trials, the subject sat in a chair in the field of view of the camera and was filmed continuously on time-coded videotape for 15 minutes before the time prearranged for the call. The subject spoke to the camera what his or her guess was before picking up the telephone.

. . . In a total of 571 unfilmed trials completed by September 2002, 231 guesses were correct (40%) and 340 wrong.

. . . By the same date, we had completed a total of 283 videotaped trials, with 127 correct guesses (45%).

Anyway, there is nothing in the chapter or appendix B which states that with some of the trials the subject guessed the name of the caller after they had picked up the phone

So why is this?

 

[Interesting Ian]

Toward the end of the paper:
quote:
--------------------------------------------------------------------------------
One final possibility for a leakage of information remains. In all trials described in the present paper, the participants picked up the telephone before making their guesses. It is therefore possible that they heard characteristic background noises, electronic hisses or other sounds that enabled them to identify the caller. But in our filmed experiments this possibility was eliminated because the participants made their guesses before they picked up the telephone. If background noises and hisses could explain the results in the unfilmed trials, the positive effect we observed should have disappeared in the filmed trials, but it did not.
--------------------------------------------------------------------------------


In other words: Yes, these experiments are a crock, but we did them another way later. Notice also that the subjects were better at guessing long-distance calls. Gee, I wonder if you can tell a call is long distance by the sound of the line?

Paul,

The book states they had completed both filmed and unfilmed tests by the same date (Sep 2002). But why are some filmed and some unfilmed?? And although the web page and the book are in agreement with the filmed trials, why is there this contradiction with the unfilmed trials? (over the issue of when the subjects made the guesses).

And wouldn't the filmed tests be more reliable in any case?? (And the results are more statistically significant!).

 

[juninho]

. Gee, I wonder if you can tell a call is long distance by the sound of the line?

Entirely true. When my fiancee calls me from Australia, I know that its her immediately without even having to speak a word. The line quality, delay, etc. immediately gives it away. That and I've got caller ID (only joking).

Actually, thinking about it, were time zones factored out of the equation in these experiments. I mean, its fairly easy to assume that when its say 3am in Nigeria (or whereever) its unlikely that they are going to be the ones ringing. Must admit I couldn't be bothered to read through the whole paper (I do have a job) so I may be talking rubbish.