Quantum Suicide: Why is only experimenter convinced?

[greypilgrim]

Hi,

Tegmark argues that after $n$ repetitions of the Quantum Suicide experiment, the experimenter can rule out collapse interpretations of QM with certainty $1-0.5^n$ where, if I understand him correctly, $0.5^n$ is the probability that a collapse interpretation is correct but he was just on a streak of luck as the gun didn't fire in every of the $n$ repetitions.

He argues that this could only convince (with $n$ large enough) the experimentalist, but not the watching assistant or anybody else. Why is that? If the experimentalist is alive after $n$ repetitions, essentially he and his assistant are both in a lab and just heard $n$ clicks, but no gunshot. Why does it make such a big difference that the gun is pointed at the experimentalist and not the assistant?

Another thing I don't quite understand: Tegmark requires the trigger mechanism to be such that "the timescale between the quantum bit generation and the actual firing is much shorter than that characteristic of human perception, say $10^{-2}$ seconds".
Why do we need that?

Also: Of how much relevance is human consciousness in this experiment? A major problem of Copenhagen is the question of what systems can cause a collapse, and some authors think that only a conscious being can. MWI in general doesn't need such anthropocentric assumptions, but what about this particular experiment?

 

[Nugatory]

I'm sorry, but a thread that starts with Tegmark's somewhat idiosyncratic gloss on the MWI isn't likely to go anywhere good.

Anyone who believes that they have a good answer to GreyPilgrim's question, PM me and we'll consider reopening the thread.

 

[Nugatory]

Open again after some offline discussion

 

[craigi]

This thought experiment does allow us, in principle, to determine between the MWI and other interpretations and where the discussion would lead, if it were allowed, is to other ways that the MWI might be proven in the not too distant future (in the next century), such as work by Deutsch.

Hi,

Tegmark argues that after n repetitions of the Quantum Suicide experiment, the experimenter can rule out collapse interpretations of QM with certainty 1−0.5n where, if I understand him correctly, 0.5n is the probability that a collapse interpretation is correct but he was just on a streak of luck as the gun didn't fire in every of the n repetitions.

In the world where he survives every trigger pull, he was lucky compared to the other versions of experimenters who died, in the other worlds. But after n iterations, he could only be in the world where he survives, because he, as a consciouss entity, only exists in one world. To contemplate whether he survived requires that he is still alive. This is just classical probability.

What is the probability that he will survive n binary life or death events? 0.5^n

What is the probability that he has survived n binary life or death events, given that he's around to ask the question? 1

He argues that this could only convince (with n large enough) the experimentalist, but not the watching assistant or anybody else. Why is that? If the experimentalist is alive after n repetitions, essentially he and his assistant are both in a lab and just heard n clicks, but no gunshot. Why does it make such a big difference that the gun is pointed at the experimentalist and not the assistant?

As n approaches infinity, the experimenter would be forced to conclude that both life and death outcomes are realized, and in turn conclude that the MWI is correct and that other contrary interpretations are incorrect. In the world where the experiment survives, he would be with an assistant who would draw the same conclusion as himself. However, there would be an overwhelming number of versions of assistants in other worlds, who wouldn't be able to draw this conclusion. So an assistant observing this experiment, wouldn't be able to draw any conclusion about the validity of the MWI.

Another thing I don't quite understand: Tegmark requires the trigger mechanism to be such that "the timescale between the quantum bit generation and the actual firing is much shorter than that characteristic of human perception, say 10−2 seconds".
Why do we need that?

The only reason for this, is that if the experimenter becomes aware that he is about to die, then will not be able to draw the conclusion that MWI is true. However, there would still be a version of the experimenter who would conclude that the MWI is true, even though an overwhelming number of versions experience their own deaths.

Also: Of how much relevance is human consciousness in this experiment? A major problem of Copenhagen is the question of what systems can cause a collapse, and some authors think that only a conscious being can. MWI in general doesn't need such anthropocentric assumptions, but what about this particular experiment?

The relevance of consciousness is, nothing more than a requirement to understand whether the MWI is correct, after performing the experiment.I think it goes without saying, don't try this at home!It's worth noting, in case you missed it. That under the other interpretations, with the exception of MWI derivatives (such as the Relational and the Cosmological interpretations), as n approaches infinity, the experimenter has a probaility tending towards 1 of simply ceasing to exist.

 

[bhobba]

Also: Of how much relevance is human consciousness in this experiment? A major problem of Copenhagen is the question of what systems can cause a collapse, and some authors think that only a conscious being can. MWI in general doesn't need such anthropocentric assumptions, but what about this particular experiment?

The quantum suicide experiment is, to be blunt, rubbish, as easily seen by the fact its argument is symmetric. It says you never experience death because in some universe you are alive. Of course you can just as well argue you never experience life because in some universe you are dead. What it relies on is something philosophers know well - namely a certain intuitive feeling we have about death and the status of experiencing it. You will find a detailed discussion about it in David Wallaces book:
https://www.amazon.com/dp/0198707541/?tag=pfamazon01-20

That isn't an issue at all for modern Copenhagen because these days an observation is considered to have occurred once decoherence has happened which is an unambiguously defined process. Our modern understanding of decoherence has morphed the measurement problem somewhat - its now not what causes collapse - its why we get any outcomes at all, which at a technical level is how a an improper mixed state becomes a proper one:
https://www.amazon.com/dp/0198707541/?tag=pfamazon01-20

Conciousness has nothing at all to do with QM, except in some very fringe interpretations. Despite Tegmarks attempt to show otherwise it has nothing to do with Quantum Suicide either.

Thanks
Bill

 

[stevendaryl]

The quantum suicide experiment is, to be blunt, rubbish, as easily seen by the fact its argument is symmetric. It says you never experience death because in some universe you are alive. Of course you can just as well argue you never experience life because in some universe you are dead.

Well, no, because you can't "experience" death. That's the asymmetry. (Of course, you can experience the intense pain immediately prior to death, but presumably you can arrange things so that death happens so quickly that there's no time to experience anything.)

The stuff about suicide is only partly about quantum mechanics. It's also about the nature of identity, which is an interesting subject, but it's more philosophy than physics. However, I don't think you can firmly come down on one side or the other of a philosophical argument without engaging in philosophy. The conclusion "it's rubbish" is as much a philosophical conclusion as "it sounds convincing to me".

 

[ShayanJ]

The conclusion "it's rubbish" is as much a philosophical conclusion as "it sounds convincing to me".

Well, such a thing is rubbish, as far as physics is concerned. In physics, we work with statements that can be decided whether from first principles, accepted theories or experimentally verified facts, not statements that are philosophical. So I'm with bhobba, its rubbish!

 

[stevendaryl]

Well, such a thing is rubbish, as far as physics is concerned. In physics, we work with statements that can be decided whether from first principles, accepted theories or experimentally verified facts, not statements that are philosophical. So I'm with bhobba, its rubbish!

But "It's rubbish" is a philosophical conclusion, not a scientific one.

 

[ShayanJ]

But "It's rubbish" is a philosophical conclusion, not a scientific one.

OK, its not rubbish. Its something we don't like in physics so we put it out for garbage collectors to take. Note I'm not telling its wrong, I'm just telling its useless, gives no conclusion. We do thought experiments because with some reasoning, we can draw conclusions from them. But what is the conclusion of this experiment? It just says if you do this experiment, the probability of MWI being true increases with the amount of time the experimenter stays alive after starting. You can do no reasoning, you can draw no conclusion. So why is it called a thought experiment? Why is it in physics? I prefer to put it outside for others to get it, whether they are garbage collectors or philosophers!

 

[stevendaryl]

OK, its not rubbish. Its something we don't like in physics so we put it out for garbage collectors to take. Note I'm not telling its wrong, I'm just telling its useless, gives no conclusion. We do thought experiments because with some reasoning, we can draw conclusions from them. But what is the conclusion of this experiment? It just says if you do this experiment, the probability of MWI being through increases with the amount of time the experimenter stays alive after starting. You can do no reasoning, you can draw no conclusion. So why is it called a thought experiment? Why is it in physics? I prefer to put it aside for others to get it, whether they are garbage collectors or philosophers!

I thought the point was that MWI is experimentally verifiable, if you're willing to risk your life.

 

[ShayanJ]

I thought the point was that MWI is experimentally verifiable, if you're willing to risk your life.

That's the point. Even if you risk your life, you can't verify it. For that, you need an ethnic cleansing of physicists!
Because the reasoning is, as we iterate more and the experimenter doesn't die, the probablity of MWI being true increases. So if the experimenter does the experiment for all of his lifetime and doesn't die, its very probable that MWI is true. But we know that its not going to be like that. There is a really low, near zero, probability for that to happen. So You get the first physicist, he dies after a week. The second one, after two days. The third one, after 10 days. The fourth one, after one month. And so on and so fourth. You will never get a conclusion. The experiment is inconclusive even in telling us how probable it is for MWI to be correct, let alone telling us its correct or not!

 

[ShayanJ]

its good to continue further. Imagine we get to the luckiest physicist of all time and s\he performs the experiment all of his\her lifetime. But it only proves to the experimenter himself\herself that MWI is probably right. Others won't understand this. So this lucky experimenter understands MWI is probably correct but nobody believes him\her because there is no evidence. So s\he dies and the remaining physicsits will continue the experiment until they run out of physicists. Then its up to the last physicist to be smart enough and forget about MWI. But if s\he's not, s\he'll do the experiment too and dies. And physics goes to extinction along with all of the physicists. R.I.P. Physics!

 

[craigi]

its good to continue further. Imagine we get to the luckiest physicist of all time and s\he performs the experiment all of his\her lifetime. But it only proves to the experimenter himself\herself that MWI is probably right. Others won't understand this. So this lucky experimenter understands MWI is probably correct but nobody believes him\her because there is no evidence. So s\he dies and the remaining physicsits will continue the experiment until they run out of physicists. Then its up to the last physicist to be smart enough and forget about MWI. But if s\he's not, s\he'll do the experiment too and dies. And physics goes to extinction along with all of the physicists. R.I.P. Physics!

Of course, the experiment couldn't be performed to demonstrate to the scientific community that the MWI is correct. It was never intended as such. To dismiss it on those grounds is to misunderstand its purpose.

It's a stark illustration, or the differing implications between collapse and relative state intepretations.

Contrary to common belief this is not purely philosophical and doesn't represent a dead-end for physics. I think people are generally getting this so let me offer a next step in developing this.

Suppose I were to tell you that I were about to roll a 6-sided die, once only, and if it doesn't land on 6, I will instantaneously erase your memory. Presuming of course you believe that I have the technology, what are your expectations for observing a 6 given each of the collapse and MWI interpretations?

This isn't the physics part yet. You'll have to wait for that, but try not to get thrown off by dismissing it as another useless thought experiment.

 

[bhobba]

Well, no, because you can't "experience" death.

That's the philosophical issue where it gets traction.

Replace it by a machine to see the silliness of the argument.

Thanks
Bill

 

[bhobba]

But "It's rubbish" is a philosophical conclusion, not a scientific one.

Like I said its based on a feeling we have about death that introduces an asymmetry. That feeling is on shaky philosophical ground as easily seen by replacing it with say a computer that passes the Turing test.

Its not really something we discuss on this forum but the philosophy of the situation is detailed in Wallace.

IMHO is a total load of BS.

Thanks
Bill

 

[bhobba]

But it only proves to the experimenter himself\herself that MWI is probably right.

It doesn't prove that.

Any interpretation will explain that thought experiment.

Thanks
Bill

 

[Quantumental]

It doesn't prove that.

Any interpretation will explain that thought experiment.

To make matters even worse, consider the fact that the metaphysics of MWI isn't worked out. MWI can be interpreted metaphysically as being about worlds splitting (fission picture) or diverging. In the divergent picture all the worlds exist from the dawn of time, but diverge as changes occur in the branches. So world A and B is identical until X happens in A and in world B Y happens. If you are on branch B, you always were on branch B, you just didn't know yet which you were on.

David Wallace as well as Simon Saunders propose the diverging view in the "Many Worlds?" book. Alastair Wilson has written several papers about this which can be found here: http://alastairwilson.org David Deutsch's 'fungible' model is also resembling this rather than the traditional "splitting" which we are often presented with.

So how does this relate to Quantum Suicide/Immortality? If you never split, only diverge, you are isolated on your branch and so if you die there, you die there just as you would in a classical one world model, you were never your doppelganger.

 

[Nick666]

Any interpretation will explain that thought experiment.

Thanks
Bill

What kind of interpretation ?

In the world where the scientist doesn't die even after one million clicks, its obvious not only for himself but for other people in that world that are witnessing the experiment, that the chance of what has happened is ultra-low. If it would be our world, what would you conclude about such an experiment ?

 

[bhobba]

What kind of interpretation ?

Did you see the word any?

In the world where the scientist doesn't die even after one million clicks,

That's the first issue isn't it - the proof there are multiple worlds.

If many worlds was true the conclusion I would draw is I happen to be in a world where he was alive and that world has low, but non zero, probability. So?

Thanks
Bill

 

[atyy]

Tegmark is wrong. To prove MWI, he must know that he died in the other world. Otherwise, his observations are consistent with a failure of quantum mechanics (or at least that he is using the wrong Hamiltonian).

 

[Nugatory]

A thread that starts with Tegmark's somewhat idiosyncratic gloss on the MWI isn't likely to go anywhere good.

Just sayin...

 

[atyy]

The shorter branch is better - it was a great thread up to then, since I got to learn high class English like "idiosyncratic gloss".

 

[atyy]

I guess the longer branch is also good, because I get to learn low class English like "Just sayin..."

 

[craigi]

Tegmark is wrong. To prove MWI, he must know that he died in the other world. Otherwise, his observations are consistent with a failure of quantum mechanics (or at least that he is using the wrong Hamiltonian).

Differentiating between interpretations, of course presumes that QM is correct within its domain. As soon as we suppose it isn't, all interpretations are invalid.

Just sayin...

Ok. Let's try to get a bit more serious.

Suppose we could create a reversible observer, which could perform the role of testing whether collapse actually occurs or if the observed state is relative to the state of the observer.

Is anyone in this this thread interested in this?

 

[atyy]

Differentiating between interpretations, of course presumes that QM is correct within its domain. As soon as we suppose it isn't, all interpretations are invalid.

But some interpretations require that QM is not correct in all domains, eg. Bohmian Mechanics, http://arxiv.org/abs/quant-ph/0506115.

Suppose we could create a reversible observer, which could perform the role of testing whether collapse actually occurs or if the observed state is relative to the state of the observer.

Real collapse usually also produces deviations from QM. There are ongoing efforts to try and test these deviations, eg. http://arxiv.org/abs/1402.5421, http://arxiv.org/abs/1410.0270.

 

[craigi]

But some interpretations require that QM is not correct in all domains, eg. Bohmian Mechanics.

Sure, but we know that QM isn't universally correct. Which is why we need QFT and QG.

The purpose of the quantum suicide thought exeriment is to illustrate the different implications between collapse and relative state interpretations.

Real collapse usually also produces deviations from QM. There are ongoing efforts to try and test these deviations.

True, but I'm not talking about objective collapse, just plain old, CI style, agnostic collapse, as opposed to relative state observation.

 

[atyy]

True, but I'm not talking about objective collapse, just plain old, CI style, agnostic collapse, as opposed to relative state observation.

It simply cannot, because the epistemology behind the CI interpretation given repeatedly surviving the Tegmark's test is that he is at least using the wrong Hamiltonian.

 

[craigi]

It simply cannot, because the epistemology behind the CI interpretation given repeatedly surviving the Tegmark's test is that he is at least using the wrong Hamiltonian.

I'm trying to move the thread on.

I'm not talking about quantum suicide (which has been wrongly attributed to Tegmark, though Tegmark has published on the subject).

I'm talking about an observation made by a thermodynamically reversible observer.

If the relative state formulation (MWI) is correct, then the reversal of the state of the observer should mean a reversal of the state of the observed object.

In contrast, the CI has the extra axiom that observation causes wavefunction collapse.

 

[kith]

Regarding the quantum suicide experiment, I see a big practical problem: if you try to shoot yourself, the probability that you survive the experiment in a severely injured state is much higher than the probability that the bullet tunnels through your head. So even if the experiment works like proponents claim, I don't think that you can repeat it.

 

[craigi]

Regarding the quantum suicide experiment, I see a big practical problem: if you try to shoot yourself, the probability that you survive the experiment in a severely injured state is much higher than the probability that the bullet tunnels through your head. So even if the experiment works like proponents claim, I don't think that you can repeat it.

You're right. It's definitely not something to try at home.

Everett believed that his formulation made him immortal. Tegmark pointed out that death is rarely a binary event. It was someone else who devised the quantum suicide thought experiment. I don't recall who.

The truth is, for every world where we cheat death, there is an overwhelming number where we barely cling to life, in pain, barely conscious. Immortality with the ultimte price.

If you really want to go down this route, you need to define the minimal state of consciousness to apply anthropic selection bias to. This is a fascinating problem in itself but, it seems unsolvable.

For a thorough treatment you try Nick Bostrom's book.

 

[zonde]

I think quantum suicide experiment is very poor starting point for any scientific discussion as this experiment is not scientific.
I see at least three problems. First, it does not allow falsification of MWI - if someone sees "experimenter" dead he shouldn't consider MWI false. Second, it relays on biased sample (even if MWI is considered confirmed) - at the end we consider one world where "experimenter" has survived while we say nothing about any other world. Third, scientific community does not gain any knowledge - idea of experiment is that MWI is confirmed only for "experimenter".

 

[Demystifier]

For my own critique of the quantum suicide experiment see
https://www.physicsforums.com/threads/critique-of-the-quantum-suicide-experiment.526213/

In short, even the experimentalist should not be convinced by his own survival, provided that he is rational.

Let me present one additional argument. Assume that the experimenter believes that MWI is wrong, but that the invisible pink elephant-angel protects him. Each time he survives (for whatever reason), he will be more and more convinced that his belief is correct. Does it make his belief justified?

 

[Quantumental]

I think my issue of divergence vs splitting should be given more consideration within the MWI community as it directly falsifies all forms of Quantum Immortality instantly and is also the metaphysical view argued for by the leaders of modern Everett reading.

 

[Demystifier]

Quantumental, the splitting interpretation is motivated by decoherence. What is the motivation for the divergence interpretation?

 

[atyy]

Regarding the quantum suicide experiment, I see a big practical problem: if you try to shoot yourself, the probability that you survive the experiment in a severely injured state is much higher than the probability that the bullet tunnels through your head. So even if the experiment works like proponents claim, I don't think that you can repeat it.

Tegmark pointed out that death is rarely a binary event. It was someone else who devised the quantum suicide thought experiment. I don't recall who.

The truth is, for every world where we cheat death, there is an overwhelming number where we barely cling to life, in pain, barely conscious. Immortality with the ultimte price.

If there is a state dead that is orthogonal to the state alive, and given the initial setup states are orthogonal, could one use that unitarity preserves orthogonality to argue that the other possibilities are never seen? The main problem, of course, is preparing yourself in the same state every time. But should be doable in principle.