Simulating "Presentiment" Experiments

[Robin]

Well I have probably spent more time on this than it really merits. So my last contribution will be to suggest a way of detecting whether an anticipation artifact is causing an apparent "presentiment" effect. It turns out to be a little like unscrambling an egg.

First start with two similar results, the first caused by an anticipation artifact with the random number generated at 0. The second is a simulation of what presentiment might really look like, with the assumption that about 3/4 of people will have varying degrees of presentiment from tiny to slight and variable targets:

​

The underlying data looks pretty much the same in each case. Any analysis done by Dean Radin and others make invalid assumptions about this effect and so their tests are ineffective. The trends are lost in other trends and noise:

​

But the anticipation effect depends upon the effect of local maxima and minima, not of absolute value, but of gradient (otherwise the lines would be parallel, not diverging).

So my strategy is to produce a series of aggregates. The first aggregate leaves out emotive trials following runs of more than 5 calm, and calm trials following runs of more than 5 emotive trials. The next aggregate does the same, but with runs of 4 and so on.

This should progressively collapse the gap between the two trends and in my simulation it does - even reversing it:

Whereas in the simulated "real" presentiment data the gap remains more or less the same:

And graphing the average gap should bear this out.

Naturally in real life this distinction may not be so clear cut, I could certainly have underestimated the amount of noise to add. On the other hand noise increases the probability that the two trends will diverge with no other artifact present.

 

[Zeuzzz]

Have you contacted Radin?

I think it is likely he will comment on this, or try to redo the experiments by eliminating this potential factor.

How would one go about eliminating this from the experiments and really testing for potential presentiment effects?

Its just that I have seen him replicate this effect in many documentaries recently, on BBC horizon (which tends to avoid woo, but is not perfect by any means) and was beggining to come to the conclusion that it was an easily reproducable effect that maybe did have some credibility, as he always seemed to get the effect from what I saw. But appearances can be decieving.

 

[Zeuzzz]

Thanks. I am definitely considering preparing an article for publishing.

You should submit it to The Journal of Scientific Exploration.

Leon

This would be the logical thing to do. I'm sure they would appreciate valid critisism of their work and may publish it. Although they are highly unorthodox, they are not completely impervious to reality.

 

[Paul C. Anagnostopoulos]

Have people seen this paper? Read the Conclusions section carefully. I believe the authors are saying that the experimenters are intuitively starting trial runs in such a way as to bias the results toward appearing as if the subjects have presentiment, presumably by knowing what the random number will generate.

http://www.liebertonline.com/doi/abs/10.1089/acm.2005.11.695?cookieSet=1&journalCode=acm

~~ Paul

 

[Kuko 4000]

Good work Robin, thanks for the info and the time you put into it. I hope Radin will try to stick to his guns and prove you wrong.

Its just that I have seen him replicate this effect in many documentaries recently, on BBC horizon (which tends to avoid woo, but is not perfect by any means) and was beggining to come to the conclusion that it was an easily reproducable effect that maybe did have some credibility, as he always seemed to get the effect from what I saw. But appearances can be decieving.

I've seen a couple of the new Horizon's and they have been extremely bad from a scientific point of view. And probably even worse for the intelligent lay person who is interested in, but not aware of, how science works. The way things were presented in the Horizon where Radin appeared (the one w/ the math guy giving dating advice, the fighter pilots, and coffee priming) was abysmal and I would think many people were thinking: "Ahh, is this how they do science? No wonder it's wrong all the time". Very unprofessional.

 

[barefootgoddess]

Very interesting read and yet somehow as I was reading it-it made me think of an explanation of subliminal and the learning use of it in tv/movies through advertising. If someone can predict through a set of pictures in the manner you have suggested through reading subjects reactions to those pictures what makes you assume someone hasn't used that data to subliminally manipulate.