Brent Meeker writes:

I believe this assumes that your thinking is not a QM phenomena, i.e. that you have a soul or mind with 'free-will' independent of physics.

Why do you say that? A similar experiment can be done without the need for any human decisions once it is underway:

A computer generates random binary digits at one second intervals, so that Pr(1) = Pr(0) = 1/2. (The computer has to use a true random number generator, such as one based on an environmental variable, for this experiment to work.) Your role is to sit in front of the screen watching the digits whilst attached to a device that may deliver instant death with probability 1/x the moment a "0" is generated but does nothing if a "1" is generated. After n seconds the experiment stops and you tally up the 1's and 0's. The idea is that at each step, 1 in x of your copies in parallel branches of the multiverse who get a "0" on their computer screen will be eliminated, thereby leaving a slight excess of copies who get a "1". Given enough trials, statistical analysis of the results should be able to show up this anomaly, and thus provide you with evidence that those parallel worlds do in fact exist. Hopefully, with a large enough x, this result will be obtained with minimum risk that 3rd person observers will have to cope with finding your dead body, which is the main practical disadvantage of the Tegmark quantum suicide experiment.

I have done some rough calculations to see what sort of values of x and n are needed to attain statistical significance. The experimental results follow a binomial distribution, which for large n can be approximated by a normal distribution with mean = np and standard deviation = (npp')^(1/2), where p = Pr(1) and p' = 1-p = Pr(0). With the suicide device on, p increases from 1/2 to 1/(2-1/x). Aiming for the mean number of 1s with the suicide device on to be 2 standard deviations higher than the mean without the device, assuming x = 1000, n is calculated as 1.6*10^7. This is terrible: you will almost certainly suffer 3rd person death in any given branch if you could sit through that many trials. Setting n = 1000, x is calculated as 8.9, which is also almost certain to result in 3rd person death.

The above results are a bit disappointing, not really any improvement on the Tegmark experiment as far as saving your from 3rd person death goes. I'll try some other numbers... but I'm getting the feeling that nature means to make us pay before revealing her secrets in this case.

Stathis Papaioannou

Stathis Papaioannou wrote:
Russell Standish writes:

I think I can rephrase Kim's suggestion as follows. Rewards usually
reflect risks, people performing death-defying acts tend to be paid
handsomely, young males performing risky acts earn the admiration of
females (the James Dean stereotype), suicide bombers getting to spend
time with heavenly virgins and so on. Therefore, given QTI gives us
some guarantee that we won't experience death, then doesn't this
encourage QTI followers to do risky things?

The trouble with the notion of QTI suggesting we should all do risky things
is much the same as the argument I give against quantum suicide as a
way of winning the lottery in my book. Most of the avenues of survival
from risky actions are in fact at considerable cost to health, social
standing etc. Only if these costs were outweighed by the benefits
accrued by the risky action is it worth doing. In fact the decision
procedure is not all that different to if QTI were not true - if
anything it make risky actions somewhat less favourable, since QTI
guarantees that you experience negative outcomes from some failed
action rather than having death as a way out.

Here is a variation on the idea of linking gambling with quantum suicide (it may have been discussed before in the earlier years of this list, but I don't recall reading about it). You build an easily concealed device which will allow you to quickly and surely kill yourself when activated; for example, a ring of plastic explosive around your neck which will detonate and decapitate you when you press a button hidden in your pocket. With this device in place, you go to the gambling venue of choice and place a bet, all the while with your finger on the button and the thought in mind that if you lose the bet, you will press it. Now, whatever you think of QTI, you're not crazy, and you are unlikely to go ahead and blow your head off if you lose. But however cautious and squeamish a person you might be, the fact that you are thinking about killing yourself and are able to do so instantly must mean that there is a non-zero probability you will do so if you lose the bet. In the MWI, this means that in some small proportion of the branches in which you lose the bet, you die. The result is that from the first person perspective, since you don't experience those branches where you die, you have managed to skew the probability of winning slightly in your favour. It may not be enough to beat the house (unless you are a gambler down to his last few dollars planning to end it all anyway), but it should be possible by statistical analysis after a sufficiently large number of bets to show that something other than pure chance is at work, and thus to prove (to yourself) the same thing as the QS experiment purports to prove, in a much less messy way.

Another variation on this theme is to watch some natural random event, like the path of a fly walking up a wall. While wearing your suicide device, you stare intensely at the fly and think: "Go to the right! If you don't go to the right, I'll kill myself!" It should seem to you that you can thereby influence which way the fly walks just by thinking about it.

ASUS M5 Ultra-slim lightweight is Now $1999 (was $2,999)

Reply via email to