I have watched this show a couple of times. The producers seem to have approached the matter of the supernatural with a grain of salt, yet they are reporting certain successes in purportedly psychic phenomena that can not be attributed to experimental error, and that have supposedly been verified by statistical analysis. I do not know if this is true and I have no way of verifying the original studies.
I would like to know if anyone has seen this series and how reliable the material in it may be considered.
I can’t even tell what channel it plays on, as I live in Hong Kong. I think it may be a british show though.
Abe, I do not know the show in question, so I’m taking a stab at this. What they are probably talking about is some studies done by a group - damn, now I can’t think of where or who, but I know I’ve heard about this. !@$#$$$ Anyway, the studies involve computer program evaluations of random chance events looking for psychic influence to the outcomes. They work by generating computer analogs of mechanical devices to do things like flip coins. Instead of flipping a coin over and over, they use computer calculations to simulate thousands of coin tosses simultaneously. The results should be a statistical profile. Then they have someone concentrate on altering the results to more heads or more tails. Similarly with an apparatus that drops ping pong balls into a tray with pegs, the statistical profile should be a bell curve.
The results of these studies seems to show a slight positive correlation of results to the intended effect. We want to shift towards more heads than tails, and poof we get a very small shift towards heads. They claim that while the results are tiny (miniscule), they are statistically valid.
There are problems with the studies. For one thing, their reported results show the effect is not dependent on distance, so the “psychic” can be anywhere on earth thinking about the test and it alters to fit what they want. Second, it appears to be unconstrained by time. They can supposedly effect future test results - concentrate now on a test not being run until 2 days from now.
These lack of limitations on the effects and the very large number of calculations being run suggests there is something that is not quite up to snuff in the computer simulations. Not having looked at the data (and knowing nothing about statistics), I really cannot evaluate it. However, it is my impression that the computer simulations themselves may have subtle variations in results that fall in the margins of noise. It is these small effects that are being recorded as hits. Of course it should be investigated more thoroughly by an independent team.
Spot on, Irishman, your explanation jogged my memory and that is exactly the experiment that the show was describing. Thanks!
What is really curious is that the producers of the show did not talk about computer noise or any other problem–they simply stated that the shifts were not accounted for by statistics and concluded that this may be evidence of the supernatural.
When I was a kid in computing class I attempted to write a program that would produce random lotto numbers (in my simple mind a perfectly random number meant being able to win the perfectly random lotto). Well as it happens the ancient green monochrome display Apple I was working on was not capable of producing random numbers… every time I ran the program I obtained the same sets of numbers! Something similar may be happening here? Why the show refused to investigate such a fantastic journalistic opportunity is beyond me. I guess we’ll have to send in the Amazing Randy.
I imagine that “Supernatural Science” is a waste of time then. I’ve seen another episode since and they really don’t try to explain anything very thoroughy.
Actually, the reason that happened is that computers can’t truly generate random numbers. Computers are algorithm based, so they use algorithms to generate pseudo-random numbers. Each time you ran the program, it ran the same algorithm, so it generated the same set of pseudo-random numbers.
There are many people that hold onto these studies to prove there is something else out there.
And Cabbage is correct - most computers cannot generate truly random numbers. They use algorithms, as he stated. Often they ask for a “seed number” to begin from. As long as you use the same seed number, you will get the same “random” number. Actually, I don’t know if there are any computers that can generate a random number.
Which brings me to wonder, why haven’t any of these tests been performed with simple reactions that are truly random as far as we know? If that psychic could save Schrodinger’s cat just a couple more times than what we’d except, it would certainly be quite amazing…
“Intel® 82802AB/82802AC Firmware Hub (FWH) … part of the Intel® 810 chipset … The Firmware Hub integrates a Random Number Generator (RNG) using thermal noise generated from inherently random quantum mechanical properties of silicon. When not generating new random bits the RNG circuitry will enter a low power state. Intel will provide a binary software driver to give third party software access to our RNG for use as a security feature. At this time, the RNG is only to be used with a system in an OS-present state.” (see Intel Random Number Generator FAQ and Intel® 82802AB/82802AC Firmware Hub (FWH)
I’ll believe it when they collect Randi’s $1 million prize.
I’m surprised that computers actually don’t generate random numbers, I had thought they could. Are there any discernable differences between the numbers generated by pseudo-random generators and truly random generators?
[hijack] When I need a small amount of random numbers, I let my stopwatch run and stop it, using the hundredths of a second numbers as random. Any problems with this method as long as you’re using a relatively small amount of numbers?[/hijack]
