Expansions of Schrodinger's cat thought experiment

[Duda Jarek]

This experiment requires a 'perfectly separated box' - that there is no interactions with its outside. It can be realized by spatial separation - using that interactions can travel with at most speed of light.

So imagine we've found a planet let say a light year from us - that we've analyzed it, can simulate it as perfectly as possible.

We can send some 'reports' between us and this planet with a speed of light - they are received a year after sending.

 

Let there be a cat which life depends on a spontaneous emission - we will get a report a year later, so before that we will be out of causal cone of this episode - from our perspective - this cat is in entangled state of life and death...

More sophisticated example - imagine we've settled a colony there, such that some of their essential decisions are made using 'a real random generator', like measuring spin of photon 45deg from its polarization ('blind justice').

Each of such choices (like choosing the leader) starts a different timeline - before we get the report, from our perspective this colony is in entanglement of many timelines.

 

So from our perspective - receiving the report is the measurement act - it 'chooses' one of timelines for the colony.

If this report was 'good enough' - we could precisely track their real timeline - it's like getting corpuscular behavior in double slit experiment in which we determine the slit - we get classical behavior.

But if there were some information which couldn't be determined - like in double slit experiment without determining the slit - shouldn't we observe kind of interference between timelines?

 

Understanding this thought experiment should give good understanding of quantum entanglement.

There is large discussion about QM interpretation:

http://www.advancedphysics.org/forum/showthread.php?p=52905

 

What do You think about it?

[GDG]

***

Let there be a cat which life depends on a spontaneous emission - we will get a report a year later, so before that we will be out of causal cone of this episode - from our perspective - this cat is in entangled state of life and death...

 

Schrodinger's cat is a fine thought experiment, but we can't take it too literally. Quantum states decohere rapidly if exposed to the environment. Certainly, the colony is not going to be in a superposition of states because (a) it interacts constantly with its environment, and (b) there are way too many observers. And I'm not sure that the original Schrodinger's cat setup would remain in a superposition of states for a substantial length of time.

[swansont]

This sounds like a which-path quantum eraser experiment. Once you know which path, the interference goes away. In this case, even if you had interference between the states, the cat can only end up in one state or the other.

[Duda Jarek]

Schrodinger's cat is a fine thought experiment, but we can't take it too literally. Quantum states decohere rapidly if exposed to the environment. (...)

Perfect isolation denotes that inside and outside of this box are causally separated - there are two separate environments (e.g. by a light year).

The cat/the interior of the box are already environments - thanks of detector+poison we amplify the quantum measurement. But this measurement doesn't inflict the outside of the box (or inflict a year later) - from this point of view we have entanglement of life and death cat.

 

Ok - imagine that in the middle we would create two entangled photons, send them in opposite directions and in the colony measure its spin and kill the cat if and only if the spin is up.

Wouldn't the photon we received be entangled with life/death of the cat?

 

I wanted this picture to show that entangled means - connected causally in the past.

And that saying about entanglement we have to specify the point of view.

 

This sounds like a which-path quantum eraser experiment.(...)

Yes, but in quantum erasure we usually have a coincidence counter.

To make this 'coincidence measurement', we have to gather information from at least two sources - in presented experiment it takes at least half a year...

Edited April 7, 2009 by Duda Jarek

[swansont]

A coincidence counter requires two things to be measured. You only have one cat.

[Duda Jarek]

You've started with quantum eraser, but I see that this analogy was too far

 

But ... let us take experimental setup like in the middle of:

http://en.wikipedia.org/wiki/Delayed_choice_quantum_eraser

A photon goes through both slits. Now using 'spontaneous parametric down conversion' (a photon is converted into pair of entangled photons having twice smaller energy) - we get 'entangled two pairs of photons'.

 

Now as on the picture - one pair is sent to a detector in the colony. But this detector is placed in some (the first) minimum of eventual interference ... and causes death of poor car if it catches photon.

The second pair is sent to us, but we don't allow beam splitters to choose if we can determine the slit or not, but as in Wheeler's experiment - just measure these photons with selected tool:

- if we would use a telescope which can distinguish both slits - photons observed in the colony should behave corpuscularly - thee is no interference - the probability of killing the cat is large,

- if we would use telescope with smaller resolution than required to distinguish the slits - we should get wavelike behavior - the probability of killing the cat would be smaller.

 

What is wrong with this picture?

[swansont]

Perhaps I am just misunderstanding your point, but for interference to manifest itself, you need an ensemble of particles — even though a photon interferes with itself, you can't tell that it has done so by detecting a single photon.

 

But you have only one cat. How does interference matter at all?

[Duda Jarek]

Look at:

http://en.wikipedia.org/wiki/File:Kim_EtAl_Quantum_Eraser.svg

Now place the upper detector in minimum of interference and connect to a poison and so the cat.

On the lower part instead of these mirrors just choose to be able to distinguish paths or not - choosing probability of cat's death...

[swansont]

Look at:

http://en.wikipedia.org/wiki/File:Kim_EtAl_Quantum_Eraser.svg

Now place the upper detector in minimum of interference and connect to a poison and so the cat.

On the lower part instead of these mirrors just choose to be able to distinguish paths or not - choosing probability of cat's death...

 

And?

 

If you know the path, you kill the cat. If you don't know the path, the cat survives.

[Duda Jarek]

If you know the path, you kill the cat. If you don't know the path, the cat survives.

Exactly!

Choosing observation tool we can choose if probability of that is large or small - doesn't we have instant communication (a light year) here?

And we get even more interesting effects if optic length to the cat is much smaller than to us...

... especially if we are near the cat ... (photons could be e.g. reflected back)

 

If You are afraid that it's not physical, please look here:

http://www.scienceforums.net/forum/showthread.php?t=39296

Edited April 8, 2009 by Duda Jarek

[swansont]

Exactly!

Choosing observation tool we can choose if probability of that is large or small - doesn't we have instant communication (a light year) here?

 

No. These experiments do not violate casuality, nor do they claim to. You have to establish that they are set up to test FTL phenomenon (rather than some other interesting physics) before you could possibly draw that conclusion.

[Duda Jarek]

So please show me what is wrong with it?

 

So what is causality? Do You believe in CPT conservation?

How You understand Wheeler's experiment:

http://en.wikipedia.org/wiki/Wheeler%27s_delayed_choice_experiment

that by choosing telescope, we choose behavior of photon a million years ago...?

Edited April 8, 2009 by Duda Jarek

[swansont]

Quantum eraser experiments generally do not test for FTL phenomena. You can't simplistically assume that if you scaled them up (i.e. moved the detectors far apart), they would exhibit this behavior. If you can find an experiment that explicitly tests for this, then please cite it. Otherwise it's a thought experiment that assumes the answer you're looking for.

 

The Wheeler experiment page doesn't claim anything happens FTL

 

http://en.wikipedia.org/wiki/Delayed_choice_quantum_eraser

 

Pay attention to the discussion section, where it is explained why this does not send information FTL or violate causality.

[Duda Jarek]

But in this article there is only the same argument - because it would violate the causality...

Ok - there is reference to a paper in which is 'proven' that it's impossible.

If it can be generally proven - shouldn't showing a problem with with given experiment be a piece of cake?

So please tell me - why exactly this specific experiment won't work?

 

Unfortunately 'proofs' in physics are extremely subtle things.

To prove something we have to some complete information about the situations - assumptions.

In physics we still don't have it - so we use some strange assumptions, approximations and proudly say that something is proven...

No it isn't!

For example - in phenomenological thermodynamics we 'prove' 2nd law of thermodynamics. But this 'proof' uses continuous functions of density, pressure, etc ... silently ignoring microscopic structure...

All serious theories we use conserve CPT symmetry, so let us assume that there is a mathematical proof that in given model there is statistical property which always increase with time (entropy). Now use the same proof after CPT symmetry - the same entropy has to work - we get contradiction.

 

Returning to the experiment - in this discussion on wikipedia is written:

However, the interference pattern can only be seen retroactively once the idler photons have already been detected

But what does it mean when photons goes in opposite directions?

For example gets to us and the colony 'in the same moment'...

What does it even mean for two points of spacetime 'not connected causally'?

Special relativity says that we could change a frame of reference - make a boost: in one we get photons earlier, in a second - the colony.

So from one frame of reference the experiment would work, but from the second not?

 

Ok simpler question - still: what do You think about Wheeler's delayed choice experiment?

Edited April 9, 2009 by Duda Jarek

[swansont]

But in this article there is only the same argument - because it would violate the causality...

Ok - there is reference to a paper in which is 'proven' that it's impossible.

If it can be generally proven - shouldn't showing a problem with with given experiment be a piece of cake?

So please tell me - why exactly this specific experiment won't work?

 

No, it's often a pain the ass to find the error when you make a problem complex enough, and doubly so when it is described in vague terms. But the result — FTL communication — tells me that there is an error, because it violates relativity, which is self-consistent. Thought problems that have a contradiction contain an inconsistency, either in the setup or in the application of the physics. The only thing that can disprove a theory is an actual experimental result. Do you have one?

 

You should be aware that argument from ignorance (i.e. "it's true until proven false") is a logical fallacy.

[Duda Jarek]

Why are You so sure that it's 'violating reality' and its 'self-consistency'?

I will cite myself, from the link I've already attached, in which I'm trying to show that it doesn't have to lead to inconsistencies like time paradoxes:

 

"But let's think about a possibility of a channel in which we could send information back in time on which we cannot fully rely - sometimes it's easier for physics to make it lie.

Observe that quantum mechanics gives physics large freedom in choosing effects below our recognition to stabilize such causal time-loops to avoid time paradoxes: like choosing what spin will be measured, or make some small influences on statistically looking behaviors ... or eventually make that this backtime channel would lie.

 

Assume that there are allowed causal time-loops, but they have to be stable.

So if there would be such 'device' to transmit information back in time and I for example would send to me a picture of the girl who will be my wife ... and just before sending it back in time I would change my mind - this timeline would never appeared - there would be completely different situation (without this knowledge).

So if I'd really get this picture from myself - I will have no choice but to really send it.

Like in (a good...) SF movie - they go back in time to repair something ... and finally it occurs that the situation is exactly as it was ...

 

How to use such 'device'?

Imagine we take some real random number generator - for example measuring spin of photon 45 degrees from its polarization.

Now the procedure is:

(1) make a choice according to this generator,

(2) if from future there is a message that it was a wrong choice - take a different one

(3) wait for results of this choice

(4) finally if it was wrong choice - send this information back in time to (2)

 

So if there was a satisfying choice - it has created stable time loop - so in fact the possibility of using this device made that our random number generator (quantum mechanics) has already chosen properly before the loop.

It can be imagined that each of choice starts a different timeline - in our language they are all entangled and physics will choose one of them in what we call future, such that there for example are no time paradoxes (like that on a string is allowed some discrete set of wavelengths).

 

What if there wasn't any satisfying choice? Would it create time paradox?

Not necessary - most probably the physics would destroy the weakest link of such loop - make that this 'device' had lied.

 

Observe that even without this real random number generator, such 'device' could work without actually being used: if for example there had to be a successful terrorist attack, there would be sent information to prevent it ... and finally in a stable timeline this attack would never happen (because of e.g. proper quantum choices below our recognition)."

 

What is wrong with such point of view, especially that the channel we are talking about (with cat and telescope) works probabilistically?

[swansont]

Why are You so sure that it's 'violating reality' and its 'self-consistency'?

I will cite myself, from the link I've already attached, in which I'm trying to show that it doesn't have to lead to inconsistencies like time paradoxes:

 

You appeared to claim that it would lead to instant communication in post #10.

[Duda Jarek]

Yes I claim

But instant communication seems to be much less against our intuition than retrocausality ... which is clearly seen in Wheeler's experiment ... CPT conservation suggests that causality should be able to go in both causality cones ...

If we accept it, instant communication is a piece of cake - send information back and forward or oppositely...

 

This doesn't mean that physics is 'nonlocal' as Bell's inequalities' enthusiasts claim - if we think about physics as in any field theory (QED, standard model, general relativity): fourdimesionally - it's absolutely local.

---

Merged post follows:

Consecutive posts merged

If something is interested, there are two more interpretations of QM in which we try to understand QM fourdimensionally as in CPT conserving field theories - 'transactional interpretation' and 'theory of elementary waves'.

I believe here starts new large discussion about it:

http://groups.google.com/group/sci.physics.electromag/browse_thread/thread/749d5a06be67485f/eac28a1f73a81aab?lnk=raot#eac28a1f73a81aab

Edited April 11, 2009 by Duda Jarek
Consecutive posts merged.