Another annoying double slit question

[Fiziqs]

I was hoping to get some comments on my interpretation of the double slit experiment. I realize that this is a bit long, and probably not worth the effort to read, much less reply, but if anyone cares to give me some feedback, I'd appreciate it. I'm trying to wrap my brain around the whole concept, and I haven't got much to work with.

Whenever I read about the double slit experiment, the literature always states that the particle's "probability wave" collapses when you measure which slit the particle actually passes through. But after giving it some thought, I began to suspect that it isn't really the particle's probability wave that collapses, but rather the observer's probability wave that collapses. I know this probably sounds a bit stupid, and it most likely is, so I was hoping that someone could point out some of the flaws in my reasoning.

I doubt that I can explain my reasoning clearly, but I'll try. First, I assume, that just like particles, observers have probability waves also. I personally know of no way to test this assumption, so I simply envision the future as being analogous to a probability wave, where all potential outcomes are possible, until conditions and decisions render them impossible. The future exists as a probability wave. Whether this is literally true or not I can't say, but it should be a straightforward mathematical construct. (Of course, I for one could never do it)

So my understanding of the double slit experiment is based upon the observer having a probability wave, in the same way that the particle does. Everything that I read tells me that it's the particle's probability wave that collapses, but it seems to me that the only thing that we can be absolutely certain of, is that the observer's probability wave collapses. Once the particle's path through slit A or B has been measured, the observer can't say for certain whether the particle still goes through both slits or not, all that the observer can say for sure, is that the observer only sees the results of it going through one slit. The particle itself may still be free to go through both, only the observer can't see it.

I know that this sounds ridiculous, but I do have some reasons for preferring this view. For one the "Delayed choice quantum eraser" seems to suggest that if an observer sees only one subset of data (D1,D2), they can see an interference pattern, whereas another observer, (D3,D4), sees no interference pattern, and still a third observer (D0) sees no interference pattern even though one exists. So based upon this information, I conclude that it is possible for an observer's probability wave to collapse, and apparently collapse the particle's probability wave, but in reality not affect the particle's probability wave at all. Just because an observer sees only a specific result, does not mean that the other possibilities cease to exist, only that the observer may no longer see them.

The other reason that I prefer this interpretation is that it would seem to give some explanation for superluminal communication between entangled pairs. The way that I have seen most people explain superluminal communication, is that observing particle A, of an entangled pair, collapses its probability wave, and this information is somehow instantly relayed to particle B. But if my interpretation is correct, then it is the observer's probability wave that collapses, not the particle's. The particles are completely unchanged, no information is exchanged. The particles are still as they were before, only now, the observer sees just one state, where before the observer saw multiple states at once. It's not the particle that changed, but rather the observer. There is no need for superluminal communication, because the effect is confined to the observer.

Ok, that's the feeble logic behind my interpretation of the double slit experiment, any comments, corrections, and additions are encouraged.

 

[DeepSeeded]

The probability wave is a misleading term. It is a real wave that has probabilities. The important point though is that it is a real wave and therefor what was thought to be a particle is actually a wave. This is why a single particle has interference effects and also why a measurement instrument of any sort interrupts the effects.

It is just a way of saying photons are made of waves as well as electrons which solves the entire paradox as long as you can accept the new probabilities involved with thinking of particles as waves.

There is to many probability waves involved in making up an observer so they can not possibly all have an interference effect large enough to consume the entire observer. Thats why people don't just disappear... or at least if they do they reappear so fast no one notices.

 

[JesseM]

I know that this sounds ridiculous, but I do have some reasons for preferring this view. For one the "Delayed choice quantum eraser" seems to suggest that if an observer sees only one subset of data (D1,D2), they can see an interference pattern, whereas another observer, (D3,D4), sees no interference pattern, and still a third observer (D0) sees no interference pattern even though one exists. So based upon this information, I conclude that it is possible for an observer's probability wave to collapse, and apparently collapse the particle's probability wave, but in reality not affect the particle's probability wave at all. Just because an observer sees only a specific result, does not mean that the other possibilities cease to exist, only that the observer may no longer see them.

That description sounds a little confused, you can't see an interference pattern in the D1 and D2 data alone, nor do you see a non-interference pattern in the D3 and D4 data alone. The D1 data would just be a list of times a photon hit D1, with no other spatial information that would allow you to graph different hits at D1 to see a pattern of any kind, and likewise for D2, D3, and D4. On the other hand, the position of D0 is being varied over the course of the experiment--if it makes things simpler you can just imagine D0 as a single wide detector that can can detect signal photons at a variety of spatial locations. Then if you take the total set of all signal photons at D0, and look at the subset of photons at D0 whose time of arrival matches up with the time an idler was detected at D1, it's when you graph this subset that you see an interference pattern. You're not actually graphing idler photons at D1, you're graphing the subset of signal photons at D0 which were entangled with idlers that went to D0. So it's in the D0/D1 coincidence count that you find a spatial interference pattern, not in the D1 data alone. If it helps, imagine taking a graph showing the total pattern of dots made by the signal photons arriving at different spatial positions at D0, then coloring red all the dots whose time of arrival matches that of an idler being detected at D1, and noting that if you only pay attention to the red dots they form an interference pattern even though the total pattern of all dots didn't form an interference pattern.

Similarly, you get another interference pattern if you graph the D0/D2 coincidence count, while you get non-interference patterns if you graph the D0/D3 or D0/D4 coincidence counts. And if you just graph the total pattern of photons at D0 without doing any coincidence counting, you'll also get a non-interference pattern. You don't need multiple observers here, a single observer who has access to all the information from all the detectors can graph anyone of these patterns.

 

[jms5631]

The probability wave is a misleading term. It is a real wave that has probabilities.

This is a little misleading I think, the reality of the wavefunction is one of the most debated points in quantum philosopy. I think it would be more appropriate to say that the ontology of the wavefunction is an unresolved aspect of QM, and right now different interpretations that are all empirically equivalent give you different answers on this issue.

 

[DeepSeeded]

I just want to make sure it is understood that it is a not a wave OF probability like I thought when I first read about it. It stems from the idea that particles are waves rather then point masses.

 

[JesseM]

I just want to make sure it is understood that it is a not a wave OF probability like I thought when I first read about it. It stems from the idea that particles are waves rather then point masses.

But there is nothing in QM that demands you believe particles "are" waves, since after all whenever you measure their position they are found at discrete points. The "wave of probability" is one interpretation of QM, it's not clearly "wrong" if that's what you're suggesting, though it's not clearly "right" either since there are other ways you could interpret it and no experimental tests that would distinguish one interpretation from another.

 

[Fiziqs]

That description sounds a little confused, you can't see an interference pattern in the D1 and D2 data alone, nor do you see a non-interference pattern in the D3 and D4 data alone. The D1 data would just be a list of times a photon hit D1, with no other spatial information that would allow you to graph different hits at D1 to see a pattern of any kind, and likewise for D2, D3, and D4. On the other hand, the position of D0 is being varied over the course of the experiment--if it makes things simpler you can just imagine D0 as a single wide detector that can can detect signal photons at a variety of spatial locations. Then if you take the total set of all signal photons at D0, and look at the subset of photons at D0 whose time of arrival matches up with the time an idler was detected at D1, it's when you graph this subset that you see an interference pattern. You're not actually graphing idler photons at D1, you're graphing the subset of signal photons at D0 which were entangled with idlers that went to D0. So it's in the D0/D1 coincidence count that you find a spatial interference pattern, not in the D1 data alone. If it helps, imagine taking a graph showing the total pattern of dots made by the signal photons arriving at different spatial positions at D0, then coloring red all the dots whose time of arrival matches that of an idler being detected at D1, and noting that if you only pay attention to the red dots they form an interference pattern even though the total pattern of all dots didn't form an interference pattern.

Similarly, you get another interference pattern if you graph the D0/D2 coincidence count, while you get non-interference patterns if you graph the D0/D3 or D0/D4 coincidence counts. And if you just graph the total pattern of photons at D0 without doing any coincidence counting, you'll also get a non-interference pattern. You don't need multiple observers here, a single observer who has access to all the information from all the detectors can graph anyone of these patterns.

I realize that I have to cross reference the data from D0 with the corresponding data in detectors D1,D2,D3, and D4, to get any practical information, but I figured that everyone here would realize that.

I guess what I was trying to get at, was the fact that one observer can see just one subset of data, be it D0/D1, D0/D2 or any of the others, and the corresponding interference pattern, (Or lack thereof) without necessarily collapsing the particle's probability wave.

So when an observer interacts with the particle, it may appear to the observer as if the particle's probability wave has collapsed, when in fact, it has only collapsed from the observer's perspective, from the particle's perspective, or from the perspective of another observer, the probability wave may be unchanged.

Forgetting my astounding lack of clarity, does any of what I said make any sense?

Is it possible that it is the observer's probability wave that collapses, and not the particle's?


P.S. Just to clarify, when I say "observer", I don't mean necessarily a human observer. Anything that interacts with the particle, either directly or indirectly could possibly be construed as an observer.

 

[JesseM]

I guess what I was trying to get at, was the fact that one observer can see just one subset of data, be it D0/D1, D0/D2 or any of the others, and the corresponding interference pattern, (Or lack thereof) without necessarily collapsing the particle's probability wave.

You're talking as though the "collapse of the probability wave" is a real physical event, whereas most interpretations say nothing fundamentally different happens when an observer interacts with a system as opposed to when any other mindless system interacts with it. And even if you don't mean to imply this, it's unclear what you mean by "the particle's probability wave"--the interference pattern is made up of a large sum of particles, and we'd certainly have to assume the observer collapsed the wavefunctions of all the particles that make up the interference pattern in the subset.

So when an observer interacts with the particle, it may appear to the observer as if the particle's probability wave has collapsed, when in fact, it has only collapsed from the observer's perspective, from the particle's perspective, or from the perspective of another observer, the probability wave may be unchanged.

Sure, a second observer can assume the first observer just became entangled with the particle when he observed it, without any collapse being assumed until the second observer measures it. But unless the second observer waits until the information has somehow been completely "erased" from the memory of the first observer (in the sense that you couldn't even in principle determine what the observer had seen by measuring every particle in the first observer's brain, not just in the sense that the first observer can't verbally recall it) and from every system that has interacted with the first observer, then the fact that he assumed the first observer didn't collapse the wavefunction but just became entangled should make absolutely no difference to the second observer's predictions about things he measures, all his predictions about measurements would be exactly the same if he assumed the first observer did collapse the wavefunction. And even in the hypothetical (but impossible in practice) case where the information was completely erased from the state of the first observer before the second observer measured him, the only way he could get any predictions distinct from what he'd have gotten if he just assumed the first observer collapsed the wavefunction would be to make an exhaustive measurement of every particle that had subsequently become entangled with the particle the first observer measured (including every particle in the first observer's body), which is again completely impossible in practice. So in practice, it makes perfect sense to assume that any human observer (or macroscopic measuring-device) collapses the wavefunction of the system it measures.

P.S. Just to clarify, when I say "observer", I don't mean necessarily a human observer. Anything that interacts with the particle, either directly or indirectly could possibly be construed as an observer.

As long as we're talking about a macroscopic measuring-device then what I said above about the practical impossibility of erasing information and subsequently measuring the every particle in the system that became entangled with the photon still applies, so for practical purposes it's fine to assume any such measuring device collapses the wavefunction. If you're extending the definition of "observer" such that the idler itself can be treated as an observer of the signal photon, then obviously in this case it is possible in practice to erase the idler's "memory" of which slit the signal photon went through, so it wouldn't be right to treat the idler-signal photon interaction as a measurement that collapses the wavefunction.