The Fundamental Difference in Interpretations of Quantum Mechanics - Comments

[Stephen Tashi]

You are only aware of one result because your awareness depends on your brain state, and the different branches of the wave function have different states for your brain, just as they have different states for the measured system.

If "you" are only aware of one result, how we avoid the conclusion that there are different physical versions of yourself who are aware of the other results?

The mathematics of modeling ourselves as a superposition of persons who observe different macroscopic results may work out, but it doesn't explain our sensation of being only one of those persons.

 

[meopemuk]

I understand that this is difficult to wrap your mind around.

Exactly. Thank you for your efforts. But, I think that I'll remain unconvinced about the virtues of the MWI. I just don't see how it provides any deeper understanding. Perhaps my brain is entangled the wrong way. So, I'll stick to my old-fashioned idea that a random sequence of measured outcomes is just that -- "a random sequence of measured outcomes".

Eugene.

 

[PeterDonis]

If "you" are only aware of one result, how we avoid the conclusion that there are different physical versions of yourself who are aware of the other results?

If "different physical versions of yourself" just means "different branches of the wave function of your brain that are entangled with different branches of objects you observe", then that is exactly the conclusion, yes. But it's very tempting to read metaphysical implications into that that aren't there, or at least not necessarily there. For example, there is no "copying" of things, as I said before; the time evolution of the wave function is unitary, and unitary operations can't "copy" anything. There is just one wave function.

The mathematics of modeling ourselves as a superposition of persons who observe different macroscopic results may work out, but it doesn't explain our sensation of being only one of those persons.

That's because we can't explain how our conscious experiences arise from the physical happenings in our brains, regardless of which interpretation of QM we adopt. But if we hypothesize that our conscious experiences have something to do with our brain states, then it seems perfectly reasonable that, if there are different branches of the wave function in which our brains are in different states, then we would have different conscious experiences in those different branches.

 

[PeterDonis]

I think that I'll remain unconvinced about the virtues of the MWI.

As are many people. I'm not much of a fan of it myself. But it's still important to be clear about what it actually says, even if you're going to end up not believing it. (Perhaps especially if you're going to end up not believing it.)

 

[bhobba]

Isn't it true that in M(any)W(orlds)I(nterpretation) each measurement creates new Universes -- as many as there are possible measurement outcomes? But you -- the observer -- must jump into one of these Universes to get on with your life. And this jump is random. So, we are back to the fundamental irreducible randomness, this time with all the bells and whistles of multiple worlds.

No.

MW is deterministic - you just have an evolving wave-function - that's it - that's all. But decoherence occurs so that you have an improper mixed state ∑pi |xi><xi>. Here |xi><xi| are the possible outcomes. The interpretation is each of those outcomes is a separate world - that is the interpretive assumption of MW. And, it turns out pi is, as a rational being, the likelihood you will find yourself in a particular world. This of course requires the Bayesian Interpretation of probability - but I prefer in this case calling it the decision theory version - in this version probability is what a rational being would decide as odds if they had to wager on it - a subtle difference - yes - but in this case an important one. Its not the only place that uses it - actuaries do as well. But anyway it turns out, using Gleason's Theorem, you can show the pi are the odds. There is a direct decision theory augment but a lot of people criticize it - Gleason however is water tight. The only out is contextuality - and there is a theorem in many worlds - the non-non contextuality theorem that ensures it is always the case ie its not contextual.

Thanks
Bill

 

[PeterDonis]

I'll stick to my old-fashioned idea that a random sequence of measured outcomes is just that -- "a random sequence of measured outcomes".

One potential issue with the MWI is how, exactly, it predicts that we should all observe a random sequence of measured outcomes if we run a large number of trials of quantum experiments. Or more precisely, how it predicts that each individual branch of the wave function will describe an observer who observes a random sequence of measured outcomes.

If we take the MWI at face value, then, for example, if someone runs 100 trials of a Stern-Gerlach measurement on electrons from a source whose wave function is set up to give a 50-50 chance of each outcome, then there is a branch of the wave function in which the person observes a "spin up" result 100 times in a row. And similarly for spin down. In fact, there will be $2^{100}$ branches, each corresponding to one of the $2^{100}$ possible sequences of measurement results. But we don't ever seem to meet any observers who have observed such things--or, again putting things into MWI language, there don't seem to be any branches of the wave function that we have evidence for, that contain any such observers. (Another way of stating this question is, how does MWI give us the Born rule?)

 

[bhobba]

But still, even within MWI, there should be a point where a choice must be made, which copy of myself --

No - its via decoherence, which happens very very quickly, but not instantaneously. No choice is made - as I said after decoherence the 'state ' (I will not use wave-function here otherwise confusion may result - I will use the exact terminology - state) is now a mixed state. Each outcome is interpreted as its own world and it turns out the pi in the mixed state detailed in my post above are the odds you will find yourself in a particular world if you were there. Of course in this Bayesian type interpretation of probability as the rational belief you have in those as odds, degree of confidence. plausibility, or whatever you want to call it, does not require you to be there - if it did it would make nonsense out of probability theory which is of course not nonsense - you were undoubtedly taught its basics at school - and for good reason - its a very very useful mathematical tool.

What you probably were not taught about is the so called cox axioms:
https://en.wikipedia.org/wiki/Cox's_theorem

That's the version of probability used here.

Thanks
Bill

 

[bhobba]

Another way of stating this question is, how does MWI give us the Born rule?

Gleason - but it is controversial.

All interpretations have strengths and weaknesses - this is the weakness of MW. MW adherents claim it is solved - those not so enamoured by it criticize it on those grounds.

What I can say is the following book is a very mathematical examination of MW by a guy with both a PhD in physics and philosophy (Wallace):
https://www.amazon.com/dp/0198707541/?tag=pfamazon01-20

Here he proves the non-contextuality theorem in MW which puts it on very firm ground.

I have been though the book and can't find any error - someone else may also enjoy trying.

Thanks
Bill

 

[bhobba]

Exactly. Thank you for your efforts. But, I think that I'll remain unconvinced about the virtues of the MWI. I just don't see how it provides any deeper understanding. Perhaps my brain is entangled the wrong way. So, I'll stick to my old-fashioned idea that a random sequence of measured outcomes is just that -- "a random sequence of measured outcomes".

Thats OK - there is no right or wrong in this interpretation stuff. A pity though you don't see the things it says about the formalism:

1. You do not need collapse
2. You can have an entirely deterministic theory where the wave-function is real
3. A more subtle understanding of probability in QM

There are undoubtedly others as well.

It however, while mathematically very beautiful when you go into it, is simply far too weird for most - far too weird for me for example. But weirdness and truth are two different things - its entirely up to you if you believe it - I don't - but that means absolutely nothing - except what I personally find it too weird.

Thanks
Bill

 

[bhobba]

The mathematics of modeling ourselves as a superposition of persons who observe different macroscopic results may work out, but it doesn't explain our sensation of being only one of those persons.

Of course it doesn't. That's part of the assumption each element of the mixed state is a separate world - isn't defining your way out of troubles great? Maybe another reason many are not taken with the interpretation.

Thanks
Bill

 

[lightarrow]

The fact in Copenaghen interp. we can't in general predict a measurement's outcome doesn't seem much surprising to me: even in an interaction between a magnetic field and single electrons (Stern-Gerlach/like experiment) we have a field (magnetic field) interacting with another field (electron field) and fields are not objects always exactly determined in space and time (Fourier).
Can someone help me to understand if my intuition can have some foundation?
Thanks

--
lightarrow

 

[junjunjun233]

There are a lot of development on the ontic view to justify the consistency. A loosely closest intuition of direct approximation on observed values. We detect superposition and provide numerous experiments that closely related to that notion.

In Special relativity/GR. They are always critical on observed values. For instance, if they observed oddity on optics like 4 identical star spread out at a certain distance. Our local intuition will easily fall for "It must be 4 separate stars" until Gravitational Lensing formalism hits them in head and was 1 star projected into 4 all along; computing convergence and deflection even if it appears as if they are in superposition of some sort (at least visually).

 

[bhobba]

fields are not objects always exactly determined in space and time

They are exactly determined by Maxwell's equations so that's not the issue.

It deeper than that - much deeper. Even if its an issue is up grabs - I personally don't think it is, but many do not agree.

Learn a few interpretations, choose one if you like, and you will understand the QM formalism a lot better.

But for heavens sake don't argue the one you choose, say - yes it's the one I use and give a bit of a blurb why you chose it - that's fine - but don't get into long discussions justifying it - they go nowhere, are useless, nobody really learns anything, and will attract the attention of poor overworked mentors like me to clean it up and may even issue an infraction.

Thanks
Bill

 

[stevendaryl]

The fact in Copenaghen interp. we can't in general predict a measurement's outcome doesn't seem much surprising to me: even in an interaction between a magnetic field and single electrons (Stern-Gerlach/like experiment) we have a field (magnetic field) interacting with another field (electron field) and fields are not objects always exactly determined in space and time (Fourier).
Can someone help me to understand if my intuition can have some foundation?

I would say that what's surprising about QM is not actually the nondeterminism, but the certainty. Or the way that nondeterminism and certainty are combined.

In an EPR experiment, you produce an electron/positron pair. When you measure the spin of the electron along the z-axis, you randomly get spin-up or spin-down. It seems that there is no way to get a more accurate prediction. But actually, there is: If somebody has already measured the spin of the positron along the z-axis, you know with certainty that the electron will have the opposite result.

 

[vanhees71]

Hm, you may as well ask, why real euclidean 3D affine space is a useful tool in classical mechanics. The answer is always the same: Physics is an empirical science, and any mathematical tool that describes the empirical findings right, is wellcome to be used to formulate theories. The short answer: Mathematical structure/tool XY is useful, because it works!

 

[bhobba]

Ok - we are now simply discussing viewpoints of various approaches to physics. I think the discussion about the insights article has pretty well run its course. So in order to maintain a well ordered thread so others can read it and learn a bit more about the insights article it will be closed. Thanks to all those that participated - it has been interesting.

Added Later:
Changed paper to the correct name - insights article so as to not confuse anyone.

Thanks
Bill