Double Slit Experiment: Does Observing Affect Interference?

[jtbell]

demonstrate how QM predicts that this "single" photon passes through both doorways ?

QM is agnostic about what a photon "really" does while it is en route from the source to the detector. It predicts different probability distributions for the arrival of the photon at various locations, depending on whether one slit is open or two slits. It makes no statement about whether the photon "really" goes through only one slit, or somehow through both slits at once. Such a statement is an inference by us, in the absence of a direct observation. Many physicists would say that we should avoid making such inferences altogether.

 

[DaveC426913]

Yes, it's not the photon itself that makes a pattern, it's the distribution of photons (or electrons, or atoms - it works for all of them).

- A photon is released and creates a single hit on the screen at point X. No pattern (with only one sample, how could there be?)
- Five seconds pass.
- A second photon is released and creates a single dot on the screen. You've got two dots separated by some arbitrary distance.
- etc. etc.
- After a few dozen photons hit the screen, you notice that the hits are not completely unifromaly distributed on the screen - they are slightly more concentrated in vertical bars, with gaps of no hits between each.
- As you let more and more photons through, one at a time, the clarity of this pattern gets more pronounced.

 

[Pickels]

Question: If the detector is capable of breaking the wave why isn't the screen? After all it too records which slit the electron passes (even if it's all of them). I imagine this is because the electron at that point has already gone through superposition and at that stage of the experiment we are dealing with the electron that has interfered with itself, in the case of no detector(?).

 

[masudr]

The slits act as position detectors, and the resulting wavefunctions are delta functions at the positions of the slit. The screen acts as a momentum detector in the direction perpendicular to motion.

See the Marcella paper that ZapperZ has referred to countless times.

 

[Rade]

I think you missed my point. A single particle being detected (photon, electron) cannot produce any pattern--classical, quantum, whatever. A single "dot" does not a pattern make. It is only after multiple runs, after many particles get detected, that a pattern emerges. And the pattern that emerges for the double slit experiment clearly demonstrates QM interference--even if there was never more than a single particle in the apparatus at any time.

Thank you--just so I understand--does this then mean that if I took 100,000 individual humans (and had them move up and down to simulate a wave) and asked them to randomly walk toward two doors, move through one randomly, and make a dot on a screen on the other side of the doors, that the dot pattern thus produced would be identical to the dot pattern observed for 100,000 photons or electrons in a double slit experiment ?

 

[DaveC426913]

...does this then mean that if I took 100,000 individual humans ... the dot pattern thus produced would be identical to the dot pattern observed for 100,000 photons or electrons in a double slit experiment ?

No. Humans cannot pass through both slits and then interfere with themselves.

You can only get the interference when the objects are given the opportunity to pass through both slits (that does not mean either/or, it means *both*).

 

[Doc Al]

Thank you--just so I understand--does this then mean that if I took 100,000 individual humans (and had them move up and down to simulate a wave) and asked them to randomly walk toward two doors, move through one randomly, and make a dot on a screen on the other side of the doors, that the dot pattern thus produced would be identical to the dot pattern observed for 100,000 photons or electrons in a double slit experiment ?

I'm at a loss as to why you would think this. What makes you think that a human (a huge macroscopic, incoherent collection of particles) would behave like a single microscopic particle such as an electron or photon? The humans would behave classically, just as one would expect. For one thing, their behavior is hardly random! And they clearly can be seen going through one door or the other, not both.

And why do you think that "moving up and down" has anything to do with the de Broglie wavelength that describes the wavelike behavior of particles? No reason to think that electrons are "moving up and down" as they travel to the slits.

Also, if you were to treat the humans as coherent particles, the associated de Broglie wavelength would be incredibly small (due to the high momentum of walking humans): way, way smaller than any doorway that you could walk through. (The slits and their separation must be on the order of the particle's wavelength in order to clearly display interference.) So I see no feasible way to even imagine such an experiment that could demonstrate interference of human-sized "particles".

 

[Michael_E]

QM is agnostic about what a photon "really" does while it is en route from the source to the detector. It predicts different probability distributions for the arrival of the photon at various locations, depending on whether one slit is open or two slits. It makes no statement about whether the photon "really" goes through only one slit, or somehow through both slits at once. Such a statement is an inference by us, in the absence of a direct observation. Many physicists would say that we should avoid making such inferences altogether.

This statement seems to parallel the attitude that Feyman takes in his lectures and popular book about QED. He just shrugs his shoulders and says, " We don't know why and can't come up with a scenario to figure it out". Then goes on to discuss how they came up with a calculations that work to determine the probability of where the effect of a photon will be detected.

BUT this attitude seems to be kind of in opposition to the way that science works, that being to make a supposition or inference that can be tested, test it and then change your suppositions based on that testing and do it all over again. If you decide not to make any suppositions that can be tested, you would get high centered and be unable to proceed in your understanding wouldn't you ?

In the case of these interference experiments and the idea of superposition and the interference of probablities, (whatever that means), what is the thinking in regard to where that superposition begins and how it ends in a single effect ? Does the superposition of different possible locations for an effect start at the moment that the photon is generated and then end once the photon is absorbed with little ability to tell what is happening in between ? Can you even say anything about what a photon "actually" is ? If any of you have a good source for reading on this I'd appreciate it, as most of the stuff on the web seems to be bogus, too simplistic or inconclusive.

Michael E.

 

[jtbell]

BUT this attitude seems to be kind of in opposition to the way that science works, that being to make a supposition or inference that can be tested, test it and then change your suppositions based on that testing and do it all over again.

The key phrase here is "can be tested," of course. According to QM there is no way to test whether the particle "actually" goes through slit A, slit B or both, without destroying the two-slit interference pattern. As far as I understand, experiments along this line bear this out so far. Someone who comes up with a way to do this and maintain the two-slit pattern would probably be high on the list for a Nobel Prize.

 

[masudr]

And why do you think that "moving up and down" has anything to do with the de Broglie wavelength that describes the wavelike behavior of particles? No reason to think that electrons are "moving up and down" as they travel to the slits.

Just to re-emphasise this point:

Firstly, the "wave" associated with any coherent object is NOT an undulating wave of the stuff the matter is made up of. This is a wave that exists not in the space that we inhabit but instead the state space of the system.

Secondly, these waves are not necessarily all perfectly sinusoidal by any means, in fact the best way to understand them are kets (or vectors) living in some complex vector space. While the "wave/particle" discussion is a good pedagogical tool sometimes, the real physics is contained in the full mathematical description. If you are familiar with that, then you willl realize why humans bobbing up and down have nothing to do with photons through 2 slits.

And finally, "observation" itself appears to be merely interaction with a large macroscopic object. And since humans are just that, there's no reason to expect human sized objects to be in a superposition of being at both slits and hence no reason to expect an interference pattern.

 

[Rade]

No. Humans cannot pass through both slits and then interfere with themselves. You can only get the interference when the objects are given the opportunity to pass through both slits (that does not mean either/or, it means *both*).

OK, thanks. So are you are saying that a "single" electron with mass can move though "both" slits at the same time (not either, or) because it is a wave function, and not a single particle ? But, if so, then why is it that this electron does not produce two dots on the screen, one associated with each slit that it moves through--since from above statement by ZZapper it was made clear that QM does not predict that a single electron moving through slits will produce two dots ?

 

[mizpah12]

follow the energy!

There is no change in how we describe the electron, other than to say it has transferred some of its energy.

Alfredblase
I think You nailed it. So obvious. I should have seen it myself!
Follow the energy! Like a detective follows the money.
Is the following awkward statement true?
“Directed” energy causes the particle to become a wave. When that wave strikes a “solid” barrier (no holes or slits) it releases the energy at the barrier and becomes a particle.
Mizpah