Why doesn't decoherence apply to unobserved experiments?

[Machine1701]

This is based on the Hugh Everetts interpretation of quantum mechanics, where the waveform never collapses but the observer becomes a part of it, thus experiencing decoherence of the wave form from their point of view.

So, let's say someone is doing the two-slit experiment. The particles going through the two slits are in a state of quantum superposition. But now an observer comes along. In order to observe the particles of the experiment, the observer must interact with them in some way - say photons bounce off those particles and then hit a camera, the particles in the camera interact with other particles along electrical wires and end up producing an image on a screen, photons bounce off the screen and hit the observer's eyes. With each interaction, the particles involved in the interaction become part of the quantum superposition, i.e., part of the experiment's waveform. The observer is also part of the waveform. This is why the observer doesn't see any superposition; they only see each particle go through only one of the two slits. Instead of a single observer seeing one particle in superposition between both slits, different versions of the observer (these different versions are now part of the quantum superposition) each observe a different, single, particle position, and are not aware of any alternatives.

My question is this: if that is how decoherence works, then why does an unobserved two slit experiment result in an interference pattern? Shouldn't observing the interference pattern involve the observer in the waveform too, so that they end up just seeing a normal pattern from each particle going through only one slit? The interference pattern was formed by the experiment and therefore is part of its waveform, and photons interact with the pattern and then hit the observer's eyes, thus involving them in the waveform just as if they had watched the experiment.

There must be something wrong with my understanding of decoherence. Thank you to anyone who can help me clear it up!

Apologies if my explanation was confusing, please feel free to ask me to clarify anything.

 

[PeroK]

My question is this: if that is how decoherence works, then why does an unobserved two slit experiment result in an interference pattern? Shouldn't observing the interference pattern involve the observer in the waveform too, so that they end up just seeing a normal pattern from each particle going through only one slit? The interference pattern was formed by the experiment and therefore is part of its waveform, and photons interact with the pattern and then hit the observer's eyes, thus involving them in the waveform just as if they had watched the experiment.

In terms of the two-slit experiment, it depends on when an observation takes place. If an observation takes place after the particle has hit the screen, then there is a range of probabilities of where the particle hit the screen. In the Copenhagen interpretation, nature has picked out a single location according to these probabilities. And, if you run the experiment for many particles, you get a double-slit interference pattern based on these probabilities.

Note that in this case:

The proposition that the particle went through either slit A or slit B is false.

If, however, an observation takes place behind the slits, then in the Copenhagen interpretation, at that point nature must choose a location for the particle. And, it must choose either A or B.

If the particle goes through slit A, then the particle impacts the screen at a single location according to the probabilities associated with a single slit: you do, in fact, get an interference pattern, but it is single-slit interference.

Likewise if the particle goes through slit B, you get single-slit interference associated with slit B.

Note that: the double slit interference pattern is not the sum of the two single slit interference patterns.

And that is what proves that the proposition above is false.

In this respect, I don't see that decoherence has any relevance, as such, to the interference pattern for the experiments you describe.

In the MWI, although personally I don't know much about that interpretation, nature contrives that in both cases - double-slit interference and single-slit interference - all the possible locations for any particle happen in different "worlds". Instead of nature choosing a single location for an impact: all possible impacts take place in some "world".

Note that in the MWI, as there is no observation at the slits in the first case, there is no splitting into separate worlds at this point.

 

[Machine1701]

Thank you so much PeroK! I finally understand this experiment :D

 

[PeroK]

Thank you so much PeroK! I finally understand this experiment :D

If you have not already watched it, Feyman's lecture on QM is an excellent presentation of this, if you have an hour to spare:

http://www.cornell.edu/video/richard-feynman-messenger-lecture-6-probability-uncertainty-quantum-mechanical-view-nature

 

[Machine1701]

That's sounds great, I'll definitely check it out :)

 

[timmdeeg]

Decoherence applies to unobserved experiments. As Anton Zeilinger has shown " C70 molecules lose their quantum behaviour by thermal emission of radiation."

https://arxiv.org/abs/quant-ph/0402146

 

[PeterDonis]

if that is how decoherence works, then why does an unobserved two slit experiment result in an interference pattern?

Because for decoherence to happen, there has to be interaction with an "environment"--something with a very large number of degrees of freedom. In a two-slit experiment, no such interaction happens until after the quantum object passing through the slits hits the screen. So the process of passing through the slits and hitting the screen is coherent, and shows interference. In other words, the "observation" in this case is the pattern on the screen, and that's it.

In the alternate version of the experiment where there is some way of measuring which slit each particle goes through, the interaction that causes decoherence happens at each slit. That means, according to the MWI, that for each run of the experiment, there will be two versions: one in which the observer sees the particle go through slit A, the other in which the observer sees the particle go through slit B. And the "observation" in this case is the observation of which slit the particle went through, plus where it ended up on the screen. There is no interference in this case because the observation at the slit causes decoherence.

 

[timmdeeg]

Because for decoherence to happen, there has to be interaction with an "environment"--something with a very large number of degrees of freedom. In a two-slit experiment, no such interaction happens until after the quantum object passing through the slits hits the screen. So the process of passing through the slits and hitting the screen is coherent, and shows interference.

Unless the quantum object radiates, like Zeilinger's C70 molecules. If I understood him correctly by this the "which slit information" passing to the environment causes decoherence so that the interference pattern at the screen gets increasingly blurred with increasing temperature of these molecules.

 

[PeterDonis]

Unless the quantum object radiates, like Zeilinger's C70 molecules. If I understood him correctly by this the "which slit information" passing to the environment causes decoherence

The thermal radiation from C70 molecules referred to in Zeilinger's paper is not "which slit information". As he says in the abstract, it's just spreading the coherence over a very large number of degrees of freedom, so as to make it practically unobservable. In other words, in order for an interference pattern to show at the screen, roughly speaking, a small number of degrees of freedom of the C70 molecules would have to stay coherent and isolated from all other degrees of freedom; but thermal radiation entangles them, making the degrees of freedom that could produce interference at the screen less and less coherent with each other.

 

[timmdeeg]

The thermal radiation from C70 molecules referred to in Zeilinger's paper is not "which slit information".

True, I might have read this somewhere else. Please forget it.

Nevertheless, is the "which slit information" somehow and only in principle encoded in the environment?

 

[PeterDonis]

is the "which slit information" somehow and only in principle encoded in the environment?

No. You can measure which silt information, which means it's an observed result. The environment is degrees of freedom that you don't observe.

 

[timmdeeg]

No. You can measure which silt information, which means it's an observed result. The environment is degrees of freedom that you don't observe.

Understood, thanks.