Quantum Mechanics is weirder than I thought

[cdave]

I recently read "The Fabric of the Cosmos" by Brian Greene. It contained a rather brilliant description of the Delayed Choice Double Slit experiment, in which effect precedes cause. Rather damaging to the notion of free will I thought.

This led to me think about an Asimov story where such an experiment takes place. They wait until the experiment says that water was/will be poured on this in 24 hours, and then seal the container. They find that the universe conspires to ensure that there always was/will be someone to do just that. Be it a Janitor spilling something, or a junior lab technician not understanding, or something more.

I was wondering what would happen if the same thing was done with this Delayed Choice test. So I started writing it up as popular science essay, that I was aiming at being understood by anyone without a science background.

I was comparing this to the Bell's inequality experiment, when I finally realised what I'd missed. In the Bell's inequality tests, non-local quantum entanglement cannot be used to send information faster than light because it involves random processes, and you need to combine information from detectors on both sides of the experiment before you can see the non-local connection.

Similarly, there is no way to send information back in time using the Delayed Choice test, as the only way to show that effect has preceded cause is to combine information from both detectors at the cause and effect time of the experiment before you can see the non-temporal connection.

{ETA} This makes the weirdness of quantum mechanics really clear to me. I knew that entanglement meant that entangled particles have some faster-than-light spacial connection, but hadn't made the obvious connection that they therefore must have some faster-than-light chronological connection.

I think this implies that all events must be predetermined.

Comments

[cdave]

Hmm. That is another way of looking at it. ... ... Makes my head hurt, but I keep coming back to the future not being in flux.

[shpalman.livejournal.com]

There's nothing random in the equations of quantum mechanics.

[cdave]

I thought that the general consensus with quantum mechanics is that things don't have definite values until you make the measurement, and there is no way to determine which value it will be beyond a statistical probablity.

That is what I meant by random in other comments.

Here's a clearer version of what I meant by "future not being in flux".

Previously my understanding of the universe was that even quantum mechanics turns out to be deterministic, there's no way to make completely accurate predictions of the future of the whole universe, as there's no way to make all the calcuations needed fast enough to keep up with the rest of the universe. In other words even if the future attributes of every particle ever is already determined those attributes are not "real" as the have not happened yet.

Except that if they can spread influence backwards, they must have happened.

[shpalman.livejournal.com]

I'm not sure - I think that idea may be going out of fashion. If measurement itself is a quantum interaction then it has to be itself described by the time-dependent Schrödinger equation.

It depends also on what you mean by attributes - the wavefunction exists (as a complex, nonobservable entity) even if some of the more intuitive "attributes" don't.

[cdave]

But AFAIK (and I could be wrong) there is no known equation that describes how a wavefunction is transformed by measurements, and that's precisely transition is what the Delayed Choice Double Slit is really weird with.

[shpalman.livejournal.com]

That's because “measurements” is too general - but it doesn't mean that the interactions which are going on during any particular measurement are beyond the scope of quantum mechanics.

[cdave]

Okay, let's take the interactions in a standard double split.

Wavicle is emitted.
It is not detected at first screen so must have passed through slits.
The interaction of the waveform from both slits can then be used to make predictions about the likelyhood of finding a particle at any point.
Once the second screen registers an impact the probability of finding it anywhere else is zero.
This is not time invariant. At one instant there is a non zero probability space filling waveform, the next there is not. Reversing this doesn't isn't possible.
Of course neither is measuring this probability wave. I guess the only question is if you saw a particle leap off the rear screen, and saw it arrive at the `source` could you tell this was time reversed, and the answer would be no.

I'm not trolling here, I genuinely understood that the collapsing waveform is not time reversable.