Asymmetrical Information

What is this Red Nova site?

Interesting, I suppose. But fraught with pitfalls. As mentioned in the article, there's always some major world event that you could peg to a shift in the randomness of the resulting numbers.

But I take greatest issue with the experiment with the cartoons. You're being shown a series of provocative images and the machine is measuring galvanic skin response, changed by your emotions. After a few, aren't you naturally going to anticipate the next image? Aren't you going to have some emotional reaction before you see the image, based on your past experience?

I agree with Bob's statement on the 'galvanic response' example; it is far more likely to be anticipation.

I suppose the real test of the 'eggs' is predictive value; I'm not quite sure how they'd do it, but I think they need real-time scanning of the results.

If there's anything to this, all they have to do is take it to Vegas or Atlantic City. If they're ejected from those places, they can try Wall Street.

On the other hand, this might just be a case of scientists moving goalposts and fooling themselves.

On the gripping hand, they might have been the people behind the mysterious short selling just before the World Trade Center attack.

There's a serious discussion here.

Why did you link to a site with annoying talking flash ads? Don't encourage that kind of Internet pollution.

(That is, Red Nova, not the link from two comments above)

Heck, every Monday and Thursday, almost without exception, I can foretell what will be in Paul Krugman's next column.

I'm more curious about how his "random number generators" are actually generating their numbers. The only really random way I can think of to do that would be to put radioactive isotopes into the things and record the emission data.

If these people are on to anything, there's an easy $1 million waiting for them; all they need to do is click www.randi.org for details. Scientists who want to "verify" paranormal phenomena will discover ways to be successful. Do a search on "Dr. Robert Johnson" (the prime mover of this thing) if you don't think he has a huge vested interest in believing this stuff. That Princeton has lent its name to this quackery for twenty years is a disgrace. The appropriately skeptical and trained observers of these "phenomena" are professional magicians, not scientists. Ask Uri Geller.

My mistake: Should be "Dr. Robert Nelson," above.

I'm sure the SCICOP folks will be perfectly neutral in their evaluation of this.

The random number generators use what they call a 'quantum event' so it does indeed appear to be truely random....If you poke around the site you'll find some more technical discussions.

One random number generator simply uses resistor "shot" noise. Resistors in a circuit generate a little noise by a quantum process. You just amplify this noise until it reaches logic voltage levels, and you've got a string of random bits.

Not all that simple in the real world - you've got to stabilize a circuit that wasn't all that stable to begin with (otherwise the output wouldn't be dominated by self-generated noise, duh), so as to keep getting 50% 1's and 50% 0's, you've got to determine when to sample the analog signal and derive the next bit, and you've got to somehow handle the times when the voltage is exactly halfway-between at sampling time. But it's quite do-able.

But when you start seeing patterns in the random output, you've been looking too hard. Like Schiaparelli looking through a poor (by todan's standards) telescope at Mars and mentally connecting all the little barely-resolvable dots into "canals". Other people saw canals too, but their maps never agreed - that is, they connected the dots differently.

I strongly suspect Princeton's results are coming from confirmation bias - that is, you remember the times when your theory worked and discard the other data. And if there's any subjectivity in deciding what was a significant event...

This project has, of course, been around for a while. Yet, no one's been able to offer an explanation or prove that there's observer bias...

For those interested, I'd also suggest looking into the work of Maj. Ed Dames and his Remote Viewing project. Peeman Enterprises sells books, tapes, and kits if you want to try it at home. And, of course, Gen. Johnson Jameson continues his work at his base in Saskatchewan. And, as John Lear (son of the Lear Jet founder) revealed recently on my blog, those elk deaths near Yellowstone were related to a glitch in the soul catcher apparatus on the moon. Steve Quayle and the Northeast Intelligence Network have more on this on their sites.

I've always been sceptical of reports that "the odds are a million to one against" after the event has occurred. Recalling from long-ago courses in experiment design, one establishes tests and criteria before the experiment is run. By waiting until after the experiment, with the data in hand, one is free to pick and choose. Data can be made to confess to anything. The best illustration of this effect is the "Texas Sharpshooter" fallacy. The sharpshooter goes into a barn with a blindfold, a revolver, and a yellow crayon. He puts on the blindfold and blazes away in all directions. Then he takes the blindfold off and goes around drawing bulls-eyes around the bullet holes, exclaiming, "What a good shot I am!"

Expanding on Eernie's post, if they read their data and said, "A whole lot of will die in Yellowstone" before it happened, and could indicate a timeframe in which the predicted event would happen, that could be impressive. Depending, of course, on how many other predictions didn't come true. But it sounds like they hear about the elk die-off on the news and go back through the recorded data looking for "predictive" squiggles.

Maybe I misunderstand and they are actually logging predictions of "significant events" ahead of time. But would a log from 9/10/2001 that said, "something big's going to happen tomorrow" show that they made a prediction? Not without more specifics - you can find something "big" happening somewhere in the world on nearly every day.

If you keep making random predictions, some of them will happen to come true. E.g., there was Jeanne Dixon, a "psychic" who supposedly predicted President Kennedy's assasination - except that IIRC she didn't actually specify a particular president, a date, or even a decade. Wait long enough and it's bound to come true... Also, Dixon published several "predictions" every day, so along with her half-dozen or so hits there were a huge number of misses.

If you want to do a statistical analysis of attempts to predict non-laborartory events, you've got an impossible job ahead of you. How do you measure the probability of confirming events?

My previous post was an amalgamation of Art Bell guests and topics, together with Phil Hendrie's satire of same. The bit about the elk, etc. has nothing to do with the RCP, other than being a topic of discussion on Art Bell's show.

If you west of the Rockies, I apologize profusely. East of the Rockies, I deeply regret the error.

Missed this when it was first posted. I don't know about this "egg" device but I was a student of Jahn's in the '70's. He got into this as something of side project. He was the Dean of the Engineering school and this was one of those things student engineers do just to try to solve the problem. Overwhelmingly it was a debunking exercise, a way to teach engineers to be careful about their design and conclusions. The real shock was the finding that would not go away. The efort was to understand how to design equipment that was free of error and operator influence, not to prove some paranormal theory.

Lots of people could effect the machine and as I remember they could do it from a distance too. Generations of enginering students have played with this effect at Princeton without getting it solved. Being engineers and not scientists, no one got all worked up about it. "Wow, that's really cool!" was pretty much the reaction.

I always got the feeling that Jahn was a bit embarassed by the whole thing, but he's honest about the results. They were really there, and I don't believe he has ever speculated much on the "how" or "why" part.

PS: The effects are very small. The clincher is not the scope of the change but the repeatablility of the effect.

Comments are Closed.