Wheeler's Delayed Choice Experiment

[byron178]

does this experiment violate any type of causality?

 

[vanhees71]

Since it's explained by quantum electrodynamics this and other delayed-choice experiments do not violate causality since quantum theory is a causal theory.

 

[xts]

Causality? How?

It rather violates our intuitive demand of realism - shows that it really makes no sense to think about photons (and other quantum objects) as existing while they travel or as traveling along some paths. It shows that our attempts to catch the photon on its run are fundamentally void and hopeless.

 

[byron178]

Causality? How?

It rather violates our intuitive demand of realism - shows that it really makes no sense to think about photons (and other quantum objects) as existing while they travel or as traveling along some paths. It shows that our attempts to catch the photon on its run are fundamentally void and hopeless.

I ask this because i was told it violates strong causality.

 

[xts]

I ask this because i was told it violates strong causality.

If I am told something like that, my quick-response-sub-conscious-neurons-specialized-on-criticism immediately respond with two questions:
1. what do you mean by 'strong causality'?
2. could you explain the machanism of violation you claim?

Then, the answer usually is: 'err... I was told so, it was convincing, but I don't remember the reasoning...'

 

[byron178]

If I am told something like that, my quick-response-sub-conscious-neurons-specialized-on-criticism immediately respond with two questions:
1. what do you mean by 'strong causality'?
2. could you explain the machanism of violation you claim?

Then, the answer usually is: 'err... I was told so, it was convincing, but I don't remember the reasoning...'

I ask this because i contacted John G.Cramer and he said that wheeler's delayed choice experiment violates a type of causality.

 

[xts]

I ask this because i contacted John G.Cramer and he said that wheeler's delayed choice experiment violates a type of causality.

So do some follow-up, and ask him for more elaborate answer: how it violates?
I'll be happy to see his answer! Forward it to me!

 

[kannank]

If FTL communication happens on a non-relativistic axis (orthogonal to space-time), does it violate causality on the relativistic dimensions?

 

[xts]

Sure, Faster Than Light communication would violate causality.

FTL means that you may start experiment in one of two modes (called 0 and 1) up to your choice, and your assistant in remote part of the lab will know what was your choice in a time shorter than distance between you and him (divided by c).

But who says that Wheeler's experiment demonstrates FTL communication? No such communication may be constructed using Wheeler's machine.

 

[kannank]

Sure, Faster Than Light communication would violate causality.

FTL means that you may start experiment in one of two modes (called 0 and 1) up to your choice, and your assistant in remote part of the lab will know what was your choice in a time shorter than distance between you and him (divided by c).

Does this violate causality? Like I said the what if the communication happens in a hypothetical non-relativistic axis. Even if it happens I don't think there could be any causality violations. See the below 'experiments'.

Thought Experiment 1 -

If we have a pair of entangled coins. I use mine to bet money and you are helping me cheating the bets. We toss simultaneously on the same inertial frame of reference (to keep that Einstein in you away!). You collapsed your wave function and observed the result as 0, you know my result is going to be 1. You called me up and told me it will be 1. I won. We toss again. This time you got a result as 0, you called me up, again I won.

Now the system we have here is this. Though for others it appears my coin is the one in use in the betting, it's your coin which controls the results. And the results have the same probability functions

f(0)=f(1)=1/2

All we are doing is we have devised a reliable cheating device for me to win. Behavior of both coins are as 'normal' coins. It's just we are predicting the results more accurately.

In a unrigged betting my probability of winning is,

f(w) = 1/2

By rigging it now,

f(w) = 1

Or we can say the probability wave function for all the events in future has also collapsed.

Thought Experiment 2 -

This is like tossing a single coin on to a table and covering it with my hand before betting. Only you and me knows that the table is a one-way glass and you are hiding under the table. You know result before anyone else through the glass table. The device works exactly like the previous entangled device. All the probability wave functions hold good here as well.

I'm going to explain experiment-2 in more sophisticated terms. When the wave function collapses 'below-the-table', it instantaneously collapses above the table. But nothing really travels FTL. And no laws of physics is violated in both cases.


An outstanding problem I see with entangled coins is this -

If I use an entangled pair of coins simultaneously, I will never get the results (0,0) and (1,1). Though both my coins are giving me the expected probabilities, I will never get (0,0) and (1,1).

Maths tell me,

f(0,0) = f(1,1) = 1/4

But my wave functions will never collapse to get these values.

Will this create some violation? I can't think.

But who says that Wheeler's experiment demonstrates FTL communication? No such communication may be constructed using Wheeler's machine.

I did not say this is the case now. I'm just trying to be prepare myself if Dr. Cramer's experiments shows positive results. And I hope he will.

cheers,
Kannan

 

[xts]

There are no 'hypothetical non-relativistic axes' If you insist to think about such, then "communication FTL" loses its meaning and you must consider rather "communication backward in time".

T.E.1
Sorry, but I can't get your point...
Can you reword the same without words like 'collapse', 'wave function', etc? Just describe it from human common perspective? Using just assumption we have 'magical' pair of coins: whenever your one shows head, mine shows tail, and vice versa?

T.E.2
Again, I can't get your point. Try again to describe the experiment without using QM terms. We are tossing ordinary coin this time. As long as you speak about tossing coins, the experiment seems clear and nothing like FTL happens.

When the wave function collapses 'below-the-table', it instantaneously collapses above the table.
I see you've fallen into a pitfall of understanding 'collapse' as something really happening in a nature.
You are sitting above the table. How may you check if the collapse happened or not?

Maths tell me, f(0,0) = f(1,1) = 1/4
Math tells you so under assumption that both experiments are not correlated. But this assumption is not valid: both in case of 'magical' coin and single coin and a spy, we have perfect anticorrelation.

 

[kannank]

There are no 'hypothetical non-relativistic axes' If you insist to think about such, then "communication FTL" loses its meaning and you must consider rather "communication backward in time".

That's why I said 'hypothetical'. Anyways how can you be sure about not having one? With so many unexplained phenomena around us, let's be open.

Try again to describe the experiment without using QM terms. We are tossing ordinary coin this time.

Collapsing the probability wave-function has nothing QM about it. It's plain probability theory. QM used it from probability theory. When you observe the coin and see the result, my probability wave function collapses to 0 or 1. When you observe which slit the electron went through, it's Schrodinger wave function collapses to 0 or 1. It's all same.

When the wave function collapses 'below-the-table', it instantaneously collapses above the table.
I see you've fallen into a pitfall of understanding 'collapse' as something really happening in a nature.

Collapses DO really happen in nature. It happens when the event (observation is a type of event) actually happens and the probability given by your mathematical formula 'collapses' to your 'actual observed probability'.

In a fair-coin toss,

f(0) = 1/2

But now if the event of tossing already happened and you got a value of zero, the retrofitted probability function for that event is

f(0) = 1

f(0) = 1/2, hold good for any future event.

You are sitting above the table. How may you check if the collapse happened or not?

The experiment has you under the table with a cell phone. Let's go step by step.

1) I toss the coin.
a) The probabilities of getting {0,1} for both side of the table is 1/2.
b) The probabilities of you and me observing {0,1} is 1/2.
c) The probability of me winning the bet is 1/2.

2) The coin falls on the table and I cover it with my hand.
a)The even of tossing has happened so the probability of {0,1} for both sides have collapsed complementarity to 0 and 1. Sorry about the usage, I don't have a better way of describing it. f(x) = {0,1}
b) The probabilities of you and me observing {0,1} is still 1/2. When these events happen it will be equal to the probabilities from step-a above. But as long as the observation event doesn't occur, the probability functions does not collapse.
c) The probability of me winning the bet is 1/2.

3) You see the result of the coin on your side as 0 or 1. This observation happens obeying the relativistic rules and this cannot happen FTL.
a)The even of tossing has happened so the probability of {0,1} for both sides have collapsed complementarity to 0 and 1. f(x) = {0,1}
b) The even of observation in your side of the table has happened. The probability collapses and gives a value of 0 or 1. Now since you know the probability will collapse complementarity for both sides, you know the value on my side will be 1 or 0. On your side,
f(x) = {0,1}
your calculated probability for my side will be, f'(x) = f^-1 (x). (I meant, complementary of your result)
But I don't know that yet. So the probability function of my observation still holds good as f(x)=f(0)=f(1)=1/2
c) The probability of me winning the bet is 1/2. No change in this, yet.

4) You make a call to me and say what my result will be when I take my hand off. This communication happens obeying relativistic rules and cannot happen FTL.
a)The even of tossing has happened so the probability of {0,1} for both sides have collapsed complementarity to 0 and 1.
b) The even of observation in your side of the table has happened. The probability collapses and gives a value of 0 or 1. From your phone call my probability has also collapsed
now at my side, f(x) = {0,1}
c) The probability of me winning the bet is 1.

The tossing was rigged because of the probability collapse on my side before the actual even of observation happened. The reasons for that are
1) The knowledge provided by you about the result.
2) The complementarity nature of the coin. Coin has to behave as 0 or 1 in an event, never both. If we used a dice instead of a coin, your knowledge doesn't collapse my probability. It just modifies it.

The 'observation' and 'communication' (the phone call) on both sides happens on relativistic axes and it can never be FTL. But I still defend my case for the probability collapse on both sides of the coin to happen instantaneously (non-relativistically).

Even though the probability collapse on both sides happen instantaneously, the 'observation' & 'communication' (the phone-call) happens relativistically. Causality is not violated since no further event can be triggered on both sides of the table FTL.

To me observation of an entangled pair is quite analogous to this experiment.

My reasoning tells me - The probability wave-forms collapse instantaneously (possibly non-relativistically) on both path-A and path-B for an entangled pair. But since the observation happens relativistically, entanglement shall never communicate FTL.

I know this has to be substantiated. I'm going to do some further thinking on this.

Sorry about this long explanation. I hope I made myself clear this time.

Appreciate your thoughts.

cheers!
KANNAN

 

[xts]

Anyways how can you be sure about not having one? With so many unexplained phenomena around us, let's be open.
Relativity is not a phenomenon, but a deductive theory, one of those constituting physics. You may as well be open to the idea that 2+2=5. Feel free to use such 'openess' to build your vision of the world, but don't expect me to criticize it further.

Collapsing the probability wave-function has nothing QM about it. It's plain probability theory.
Really? Would you give me any reference to any (peer-reviewed = non-freak) article, using the term 'probability wave-function' in context other than QM (or its children, like QED, QCD etc.)?

Collapses DO really happen in nature.
You may believe so - such belief is harmless. The results you obtain from calculations will be the same. If you really want to build your model of world using collapses and backward-in-time casuality controlling them - your choice.

The experiment has you under the table with a cell phone. Let's go step by step.
Let me rewrite your scenario in simple words, filtering out all the unnecessary gibber, leaving just empirical facts:
1. You toss the coin and you don't know the result.
2. Coin falls on table table and you still don't know the result
3. I see the result
4. I am telling you the result
5. You know what the result is, so (if you are allowed to put your bet as late as now), you are a millionaire

The tossing was rigged because of...
Not tossing got rigged (you tossed the coin fairly). You just put your bet later than you possesed knowledge about result, so you won the game betting for known result.

But I still defend my case for the probability collapse
Of course, you may add as much as you want words like 'probability' and 'collapse' to the simple and full description of events that happened.
But you don't need them - they just violate Occam's principle.

You haven't explained what the 'collapse' is, nor what do you need it for.
The only meaning I may find in your story is that collapse occurs at the moment when coin falls on the table. I call it simpler: event happens. Yes, I agree, events DO happen!

To me observation of an entangled pair is quite analogous to this experiment.
Finally I agree! But we still don't need any 'collapses' to describe what happens.
There are several differences however. One is important in such case - both sides of entangled experiment table are separated by quite a long distance, they may be in relative motion. So if you want to assign the meaning to word 'collapse' and you insist that collapse on both sides of the table occurs simultaneously, you (living in relativisting world) must decide: simultaneously from whos perspective?
Go on with your approach, until you drown in paradoxes of non-simultaneous simultaneity...

 

[virtualzx]

it is helpful to keep in mind that the interpretation of any of these experiments suffers from the highly abstract and sketchy separation of the equipment and the observer from the system under investigation.

it has long been pointed out that if one put the observer into the system and separate them from the rest of the universe, the difficulty of interpretation goes away.

when he or she tries to observe the entangled state A(+)B, a cat state of the observer is created:
A(+)B + the observer ready for the measurement with his equipment
----->
(the observer seeing result from state A, wondering when the collapse into a pure state A from an entangled state happened, and pondering upon the philosophy implications of this experiment, thinking about maybe publishing a paper or post on forum about this + the equipment turned off + that system that is in its post measurement state that no one really care about anymore now that the experiment is already done)
(+)
(the observer seeing result from state B, wondering when the collapse into a pure state B from an entangled state happened, and pondering upon the philosophy implications of this experiment, thinking about maybe publishing a paper or post on forum about this + the equipment turned off + that system that is in its post measurement state that no one really care about anymore now that the experiment is already done)

However, this analysis has to be done by some other scientist that never interact with this observer at all. Simply knowing the outcome of the experiment, by chatting on-line or by reading his paper, you too will become entangled. quickly this entanglement spread and creates a cat state of the whole universe!

Apparently the observer or anyone else will not be able to calculate the wave function of the whole universe, nor does anyone want to do so, nor does anyone really care about any of the other states that is entangled with himself that he or she or anyone he or she knows will not likely ever have anything to do with.

To make life understandable, one has no choice but to separate the equipment and himself from the system, from one level or the other, in a cold hard but practical way.

and there comes the collapse of the wavefunction. there is nothing metaphysical about the mind or knowledge of observer that magically collapse the wave function. instead, the measurement drags the observer into the system and disables him or her to understand the system in a logically consistent manner.

the collapse is not created by the measurement, but is instead created by the interpretation of the measurement.

 

[byron178]

So do some follow-up, and ask him for more elaborate answer: how it violates?
I'll be happy to see his answer! Forward it to me!

Xts, Here is what he said: I would say that STRONG causality, which asserts that a cause must always precede its effect in any reference frame, is violated in
many quantum optics experiments, in particular Wheeler's Delayed Choice Experiment and the Quantum Eraser Experiment (which
you can read about at http://www.npl.washington.edu/AV/altvw90.html ). In such experiments, the "cause" is a measurement performed
in the present, and its "effect" is the path of a photon in the past, and whether it passed, for example, through two slits or through only one.

`These are examples in which the present influences the past, but not in a way that allows one to send messages to the past.

 

[xts]

In such experiments, the "cause" is a measurement performed
in the present, and its "effect" is the path of a photon in the past

Here you have an implicit assumption, that there exists such physical reality, as "path of photon". Could you propose any empirical test of its existence? Or maybe I should ask for a bit more: are you able to track or map the photon's path? No, you can't. Quantum Mechanics prohibits to track photons. But you believe photons must travel along some path! All rabbits, mammooths, men and even stupid fizzy women always travels along paths. Photons must do that as well. You can't imagine how it may happen, that photon started at some point A, got detected at other point B, but its path make no sense. Thus - the path must exist, even if it is fundamentally impossible for photon to leave single footprint on its way. Bees don't leave footprints, but they fly over defined paths, so maybe photons are more like bees than like rabbits?

What, if the intuition of 'path of photon' just makes no sense? What is then affected by Wheeler's measurement?

present influences the past, but not in a way that allows one to send messages to the past.

That's the point!
Dragon in my garrage! (I hope you know this metaphore, if not: http://www.godlessgeeks.com/LINKS/Dragon.htm)
He is there, he was there yesterday and my todays action affect his past. But, of course, he is invisible and there is no way to check that empirically. But I have perfect intelectual arguments, that my present affects dragon's yesterday path...

 

[byron178]

Here you have an implicit assumption, that there exists such physical reality, as "path of photon". Could you propose any empirical test of its existence? Or maybe I should ask for a bit more: are you able to track or map the photon's path? No, you can't. Quantum Mechanics prohibits to track photons. But you believe photons must travel along some path! All rabbits, mammooths, men and even stupid fizzy women always travels along paths. Photons must do that as well. You can't imagine how it may happen, that photon started at some point A, got detected at other point B, but its path make no sense. Thus - the path must exist, even if it is fundamentally impossible that photon left single footprint on its way. Bees don't leave footprints, but they fly over defined paths, so maybe photons are more like bees than like rabbits?

What, if the intuition of 'path of photon' just makes no sense? What is then affected by Wheeler's measurement?


That's the point!
Dragon in my garrage! (I hope you know this metaphore, if not: http://www.godlessgeeks.com/LINKS/Dragon.htm)
He is there, he was there yesterday and my todays action affect his past. But, of course, he is invisible and there is no way to check that empirically. But I have perfect intelectual arguments, that my present affects dragon's yesterday path...

so the present can influence the past?

 

[xts]

so the present can influence the past?

OK. I'll be less ironic and more explicit...

It depends on your definition of the word 'influence'.

If the 'influence' means 'produces empirically detectable effects' - physicist's anwer is: 'no'

If the 'influence' means 'there are effects, but they are restricted to undetectable beings (daemons, photon paths, invisible dragons, ghost of your grandma...)' - physicist's answer is: 'go with this question to a priest or shaman'

BTW: good teological question: makes it sense to pray for changes in the past?

 

[byron178]

OK. I'll be less ironic and more explicit...

It depends on your definition of the word 'influence'.

If the 'influence' means 'may produce any empirically detectable effect' - physicist's anwer is: 'no'

If the 'influence' means 'there are effects, but they are restricted to undetectable beings (daemons, photon paths, invisible dragons, ghost of your grandma...)' - physicist's answer is: 'go with this question to a priest or shaman'

by what i mean if the present can influence the past i mean like time travel backwards.

 

[xts]

After time travel your presence would be detectable there, thus you could bring some information backward in time, thus it is impossible.
But, of course, your may penetrate past as an invisible and undetectable ghost. Just don't clang chains! That's not allowed.

 

[byron178]

After time travel your presence would be detectable there, thus you could bring some information backward in time, thus it is impossible.
But, of course, your may penetrate past as an invisible and undetectable ghost. Just don't clang chains! That's not allowed.

so its safe to say they wheelers delayed choice allows influence to the past? i ask this because you said that he was right in saying that the present can influence the past.

 

[xts]

so its safe to say they wheelers delayed choice allows influence to the past?

We (I hope you too) live in countries respecting freedom of thought and speech. It is safe.

Once again: you may define the word 'influence' such or such. Up to your philosophical taste.
If you are a mystic and you like to speak about 'undetectable influence' - say so.
If you are empiricist and you restrict meaning of 'influence' to 'detectable influence' - you can't say so.

If you want to be honest - make clear to the audience which meaning you use.

 

[byron178]

We (I hope you too) live in countries respecting freedom of thought and speech. It is safe.

Once again: you may define the word 'influence' such or such. Up to your philosophical taste.
If you are a mystic and you like to speak about 'undetectable influence' - say so.
If you are empiricist and you restrict meaning of 'influence' to 'detectable influence' - you can't say so.

If you want to be honest - make clear to the audience which meaning you use.

ok by what i mean is in wheelers delayed choice allow influence the past in physical reality,im confused because you say that the present influences the past but not in a way that allows you to send information to the past,thats what confuses me.that is kind of contradiciting.

 

[xts]

So now you just changed your philosophic-lexical question from 'what influence means' to 'what physical reality means'.

If your ontological taste restricts 'reality' to something detectable, answer is again 'no'

If you accept that photon paths, invisible dragons, or your grandma's ghost are physically real - feel happy with answer: 'wheelers delayed choice allow influence the past in physical reality'

What I am trying to say is that 'sending information' is equivalent to 'influencing detectably'.

 

[byron178]

So now you just changed your philosophic-lexical question from 'what influence means' to 'what physical reality means'.

If your ontological taste restricts 'reality' to something detectable, answer is again 'no'

If you accept that photon paths, invisible dragons, or your grandma's ghost are physically real - feel happy with answer: 'wheelers delayed choice allow influence the past in physical reality'

ok, to me a violation of causality is an influence that can affect the past in reality,i hope you see the confussion here.also is there a view in which the present does not affect the past?

 

[xts]

ok, to me a violation of causality is an influence that can affect the past in reality,i hope you see the confussion here.

I see, and I am happy you've finally accepted common-sense defnition of reality as something detectable.
I also hope you see that Wheeler experiment does not influence past physical reality (in a meaning you've just accepted) and thus doesn't violate causality.

also is there a view in which the present does not affect the past?

Oooch? Are there people who do not believe in ghosts, invisible dragons and photon paths? Yes. They are. Not many, but there are some.

 

[byron178]

I see, and I am happy you've finally accepted common-sense defnition of reality as something detectable.
I also hope you see that Wheeler experiment does not influence past physical reality (in a meaning you've just accepted) and thus doesn't violate causality.


Oooch? Are there people who do not believe in ghosts, invisible dragons and photon paths? Yes. They are. Not many, but there are some.

im confused because in post #16 you said that the present can influence the past.

 

[xts]

im confused because in post #16 you said that the present can influence the past.

#16? I said so?
I was sure it is me who need some sleep (2AM) to think clearly... Goodnight then!

 

[byron178]

#16? I said so?
I was sure it is me who need some sleep (2AM) to think clearly... Goodnight then!

im sorry i was wrong,i meant that you agreed with cramer that the present can influnce the past.

 

[mathal]

so its safe to say they wheelers delayed choice allows influence to the past? i ask this because you said that he was right in saying that the present can influence the past.

Wheeler proposed delayed time experiments to demonstrate that actions in a 'present' could alter the outcome of an experiment where photons (or any other quantum object) began their path before the alteration occured. Experiments have demonstrated over and over again that he was correct.
To make this clear.
0.Wheeler did accept that actions in the present can alter the past. He went to the extreme of referring to measurements in the present altering a photons history to the 'present' from it's inception billions of years ago, to make that point.
the experiments - precis
1. a photon begins it's path.
2. the path of one of the possibilities of the photon is altered.
3. a measurement of each photon's path is noted.

The statistical results of such experiments (and that is necessarily all you'll ever get-statistical non specific data) is in agreement with the conditions if the experiment had been set up ahead of time ,as if the entire experiment was like when the photon's path was altered.

There is only one model, the De Broglie-Bohm theory, that can offer an explanation that is not either acausal or time-reversed.

perhaps there is another model? (i.e.do you know of one or is De Broglie-Bohm what you aver to?)

Personally I prefer a block universe acausal model.

mathal

 

[byron178]

Wheeler proposed delayed time experiments to demonstrate that actions in a 'present' could alter the outcome of an experiment where photons (or any other quantum object) began their path before the alteration occured. Experiments have demonstrated over and over again that he was correct.
To make this clear.
0.Wheeler did accept that actions in the present can alter the past. He went to the extreme of referring to measurements in the present altering a photons history to the 'present' from it's inception billions of years ago, to make that point.
the experiments - precis
1. a photon begins it's path.
2. the path of one of the possibilities of the photon is altered.
3. a measurement of each photon's path is noted.

The statistical results of such experiments (and that is necessarily all you'll ever get-statistical non specific data) is in agreement with the conditions if the experiment had been set up ahead of time ,as if the entire experiment was like when the photon's path was altered.

There is only one model, the De Broglie-Bohm theory, that can offer an explanation that is not either acausal or time-reversed.

perhaps there is another model? (i.e.do you know of one or is De Broglie-Bohm what you aver to?)

Personally I prefer a block universe acausal model.

mathal

So it does allow us to interact with the past? does it happen in reality? so that would mean it violates a type of causality?

 

[xts]

There is only one model, the De Broglie-Bohm theory, that can offer an explanation that is not either acausal or time-reversed.

You missed one word. You should say There is only one realistic model...
'Realistic' :== 'assuming objective existence of intuitive, but nondetectable entities like "photon paths" '
Or maybe you meant that using the word 'acausal'...?

Anyway - since EPR/Bell I definitely prefer and advocate non-realistic models than shy time-reversed causalities , Bohmian pilots, and similar constructs.

 

[byron178]

You missed one word. You should say There is only one realistic model...
'Realistic' :== 'assuming objective existence of intuitive, but nondetectable entities like "photon paths" '
Or maybe you meant that using the word 'acausal'...?

Anyway - since EPR/Bell I definitely prefer and advocate non-realistic models than shy time-reversed causalities , Bohmian pilots, and similar constructs.

has it been verified that wheelers expierament allows it to influence the past?

 

[xts]

Come on!

It is fully verified and non-controversial what is the outcome of the Wheeler's experiment.

The answer to your question depends only on definition you put under the word: 'influence', which depends in turn on your ontological and lexical taste.

The dispute is not about anything related to the world. It is related to the word. And, unfortunately, this word has no precisely defined and commonly accepted meaning - different people (philosophers) use it differnetly.

 

[Chronos]

Simultaneity is the issue here. In GR, there is no absolute simultaneity. Extending quantum theory to macroscopic reality is as difficut as extending GR to the quantum realm.