What are the implications of this experiment?

[Fyzix]

Dude, you are obviously borderline psychotic.
Take your medication and just let sane people worry about these issues.
Mentally unstable people shouldn't really do deep philosophy...

Nothing will happen in 2012 by the way, it's all a hoax...
I think you should log off the internet and try to get some therapy and get back into the real world.

Good luck on your road to recovery.

 

[Varon]

Dude, you are obviously borderline psychotic.
Take your medication and just let sane people worry about these issues.
Mentally unstable people shouldn't really do deep philosophy...

Nothing will happen in 2012 by the way, it's all a hoax...
I think you should log off the internet and try to get some therapy and get back into the real world.

Good luck on your road to recovery.

Whatever to make you feel happy.

Anyway. To those who know. So what is the implications of the latest trajectory experiment "Observing the Average Trajectories of Single Photons in a Two-Slit Interferometer?" Nothing? Perhaps those sensational magazine articles about it are just to awe the laymen?

 

[unusualname]

I agree with that. Anytime a theory predicts something found 50 years later in experiment, and it's as close a match as this is to Bohmian Mechanics, I sit up and take notice. The only problem(s) with it that I am aware of is it's not orthodoxy and it requires a pilot wave. I'm not in favor of adding anything unnecessary, but it might be necessary to add the pilot wave to obtain a deterministic model. Essentially, the advantage of a deterministic model is it can be developed, tested by experiment, refined and built upon. Of course, the same can be done to some extent with a statistical model, but if a deterministic model can be developed, it seems like a better way to go for some applications. Remember, we are not talking about something crackpot here. It has rock solid theory, math and experimental verification. Just my opinion.

The point is, could the results be predicted by a more simpler argument, or from alternative interpretations? And does anybody care? :-)

You see, in MWI you even have trouble with justifying the basic born rule, nevermind establishing an ensemble path, and no one really cares about that.

 

[IllyaKuryakin]

The point is, could the results be predicted by a more simpler argument, or from alternative interpretations? And does anybody care? :-)

You see, in MWI you even have trouble with justifying the basic born rule, nevermind establishing an ensemble path, and no one really cares about that.

Right. The question is, is there a simpler model that explains these results, or even better, can predict these results beforehand? I don't know of any simpler deterministic models that have solid theory and math behind them.

Of course, there might be a near infinite number of possibilities, but it's up to the creator of other models to develop the math and theory to answer all questions about their model. If the model is more complex, it should be presented with a method for experimental verification or discrimination to prove it is superior in some way.

I suspect there can be additional experimentation using weak measurement theory or other methods that will aid in developing and building upon the Bohmian Mechanics model. At this point, I believe we have two good methods with solid theory and math to describe quantum wierdness, statistical and deterministic. I really believe theory should focus on experimental methods to futher refine these models and explore their limits to see if they break down at some point. This has been done for decades in orthodox QM, but I believe the deterministic model has not received adaquate experimental attention until now.

 

[unusualname]

Right. The question is, is there a simpler model that explains these results, or even better, can predict these results beforehand? I don't know of any simpler deterministic models that have solid theory and math behind them.

Of course, there might be a near infinite number of possibilities, but it's up to the creator of other models to develop the math and theory to answer all questions about their model. If the model is more complex, it should be presented with a method for experimental verification or discrimination to prove it is superior in some way.

I suspect there can be additional experimentation using weak measurement theory or other methods that will aid in developing and building upon the Bohmian Mechanics model. At this point, I believe we have two good methods with solid theory and math to describe quantum wierdness, statistical and deterministic. I really believe theory should focus on experimental methods to futher refine these models and explore their limits to see if they break down at some point. This has been done for decades in orthodox QM, but I believe the deterministic model has not received adaquate experimental attention until now.

I would rather just ask whether there is ANY explanation of the (ensemble) paths found in this experiment other than the Bohmian analysis?

 

[yoda jedi]

I really believe theory should focus on experimental methods to futher refine these models and explore their limits to see if they break down at some point. This has been done for decades in orthodox QM, but I believe the deterministic model has not received adaquate experimental attention until now.

even for super deterministic models.

Testing super-deterministic hidden variables theories.
http://arxiv.org/PS_cache/arxiv/pdf/1105/1105.4326v1.pdf


.

 

[IllyaKuryakin]

I would rather just ask whether there is ANY explanation of the (ensemble) paths found in this experiment other than the Bohmian analysis?

Good question. I haven't seen anyone else derive this result with any deterministic model other than Bohmian Mechanics. Someone claimed they could derive the same results with orthodox QM, but I haven't seen the math. I'm not sure if that's possible either really, since orthodox QM doesn't contain the equation for the particle positions interpreted as the pilot wave in Bohmian Mechanics. Perhaps someone else here knows the answer?

 

[Varon]

I've re-read this thread for the past 3 hours as well as all the articles except the original paper where I don't have access to.

I'm still a bit confused about something so hope someone can clarity.

Trajectories of ensemble is detected. Yet we are not sure if a single particle has trajectory or not? Or does this confirm there is at least trajectory? If none. How could a single particle doesn't have trajectory yet ensemble of it have??

For years I was exposed to laymen books which say between measurement, the particle turns into a wave.. this means there is no trajectory as waves are everywhere. So does this experiment refutes at least this incorrect laymen book explanations? I think the reason many of you like SpectraCat don't think this experiment has any relevance is because he assumes that particles are always particles (right?) in between measurement, compared to laymen books which say the particle turns into wave without any trajectories in between measurements.

Whatever, this experiment at least gives the idea now that the only valid Copenhagen are those with so called Bohmian trajectories? Or is it still compatible with laymen books which says particle turns into wave between measurement. But I still can't understand how a pure wave has trajectories. Pls. explain. Thanks.

 

[Varon]

I've re-read this thread for the past 3 hours as well as all the articles except the original paper where I don't have access to.

I'm still a bit confused about something so hope someone can clarity.

Trajectories of ensemble is detected. Yet we are not sure if a single particle has trajectory or not? Or does this confirm there is at least trajectory? If none. How could a single particle doesn't have trajectory yet ensemble of it have??

For years I was exposed to laymen books which say between measurement, the particle turns into a wave.. this means there is no trajectory as waves are everywhere. So does this experiment refutes at least this incorrect laymen book explanations? I think the reason many of you like SpectraCat don't think this experiment has any relevance is because he assumes that particles are always particles (right?) in between measurement, compared to laymen books which say the particle turns into wave without any trajectories in between measurements.

Whatever, this experiment at least gives the idea now that the only valid Copenhagen are those with so called Bohmian trajectories? Or is it still compatible with laymen books which says particle turns into wave between measurement. But I still can't understand how a pure wave has trajectories. Pls. explain. Thanks.

http://scienceblogs.com/principles/2011/06/watching_photons_interfere_obs.php

The above is the clearest explanation ever of the original paper. Even if you only do weak measurement of momentum of a particle without collapsing it, the mere fact you can measure its momentum at a particle position means a single particle at least has trajectory. But if you still insist it doesn't prove a single particle has trajectory, pls. explain how so as this is what really confused me all day. In laymen books. They emphased a particle which morphs into a wave has no trajectory in between measurement. Can an expert here confirm it is plain wrong? I just want to know if I'm understanding the whole arguments right.

 

[Varon]

What concerns me is this. The purpose of the wave function in the double slit in Copenhagen is to cause interference. It is only after interference that the particle re-appears. So when the particle have trajectory, then it doesn't make sense that upon collapse, it materializes. Also what pushes it to either the left or the right? Implication of it is that if the particle has existing trajectory (even not well-defined), then it's some variant of Bohmian Mechanics. Unless those who want to retain Copenhagen have to shift positions and state Copenhagen have Bohmian trajectory. But this doesn't make sense. What pushes the particle to the left or right? Hope someone can clarify all this as this is the bottom of the confusion in some of us (or at least me if most of you understood already).

 

[IllyaKuryakin]

even for super deterministic models.

Testing super-deterministic hidden variables theories.
http://arxiv.org/PS_cache/arxiv/pdf/1105/1105.4326v1.pdf


.

Yes, a good example of an experiment design to prove the existence of a deterministic model. I'm not sure if the technology esists currently, but if the experiment is important enough, it seems someone somewhere always finds a way to perform it.

Maybe it also indicates another interpretation of Steinberg's results. Namely, in a one at a time photon source that produces 31,000 photons in 15 seconds, many of those must have been in nearly the same state, and if the results deviate from a regular probability distribution, something other than a statistical probability distribution would seem to be occurring. Not sure about that one, but something interesting to think about.

 

[StevieTNZ]

Good question. I haven't seen anyone else derive this result with any deterministic model other than Bohmian Mechanics. Someone claimed they could derive the same results with orthodox QM, but I haven't seen the math. I'm not sure if that's possible either really, since orthodox QM doesn't contain the equation for the particle positions interpreted as the pilot wave in Bohmian Mechanics. Perhaps someone else here knows the answer?

Here's sentences I read last night in the book "Einstein, Bohr and the Quantum Dilemma" by Andrew Whitaker (publisher: Cambridge University Press):

I have used the term 'hidden-variable theories' rather than 'hidden-variable interpretations', because, although hidden variables may start as an attempt merely to interpret the formalism, their properties must be developed as a genuine addition to it. Thus the word 'theory' seems more appropriate.

 

[IllyaKuryakin]

Will someone please correct me if I'm wrong, but I don't believe orthodox QM could have predicted Steinbergs results? I believe that orthodox QM would have predicted a random probability distribution of photons according to schrodinger's wave equation, yielding NO ensemble trajectories. Have I got that wrong somehow?

 

[IllyaKuryakin]

Here's sentences I read last night in the book "Einstein, Bohr and the Quantum Dilemma" by Andrew Whitaker (publisher: Cambridge University Press):

I have used the term 'hidden-variable theories' rather than 'hidden-variable interpretations', because, although hidden variables may start as an attempt merely to interpret the formalism, their properties must be developed as a genuine addition to it. Thus the word 'theory' seems more appropriate.

Agreed, what follows from Stanford Encyclopedia of Philosophy is more than an interpretation:

"For Bohmian mechanics the state of a system of N particles is described by its wave function ψ = ψ(q1,...,q N) = ψ(q), a complex (or spinor) valued function on the space of possible configurations q of the system, together with its actual configuration Q defined by the actual positions Q1,...,QN of its particles. The theory is then defined by two evolution equations: Schrödinger's equation

iℏ(∂ψ/∂t) = Hψ

for ψ(t), where H is the nonrelativistic (Schrödinger) Hamiltonian, containing the masses of the particles and a potential energy term, and a first-order evolution equation,

The Guiding Equation:
dQk/dt = (ℏ/mk) I am [ψ*∂kψ/ ψ*ψ] (Q1,...,QN)"

So I'd agree that any interpretation must evolve into a theory, or fall apart.

 

[Varon]

I got this disturbing comment that even on purely classical waves, one can get the same result. If true. This means the implications of the experiment is nothing significant. Ken G wrote in the philosophy thread (hope experts here can comment especially when Demystifier returns on monday):

(Ken G wrote:)

"What I'm saying is, I'm not convinced that "weak measurement" is any different from "compiling average trajectories from treating the wave energy flux like a divergenceless scalar field and drawing 2D lines of force for that field." I maintain you could get that exact same picture by measuring the energy flux of a classical wave passing between two slits, and drawing trajectories such that the line density is proportional to the energy flux density. This would be completely consistent with a macroscopic treatment of an energy flux as a photon number flux. Those trajectories don't really mean anything beyond a statistical treatment of where photons go in large aggregations, that they could get the same picture with "weak measurement" of "one photon at a time" doesn't strike me as being at all profound.

Let me put it another way. The key statement that we don't know the trajectory of an individual photon is that we cannot know which slit it went through, and still have that photon participate in an interference pattern. Does this experiment tell us which slit any of those photons went through? No. So what? There are still no trajectories in the physical reality of what happened to those photons, and it's not at all clear that an "average trajectory" is anything different from the usual macro aggregate measurement in the classical limit. To me, all this experiment is is a kind of consistency check that "weak measurement" can recover statistical aggregates, but I see no threat to the CI interpretation that the reality is still only what you measure and not what happens between the measurements. So they can create weak measurements that don't completely collapse the wave function, then recover the aggregate behavior in the same way that complete measurements that do collapse the wavefunction could easily do also. What does that tell us? That weak measurements don't mess up aggregate results? Why should we be surprised-- the weak measurements don't tell us the trajectories of any of those particles."

----

Is it true you can produce the same result using classical waves that don't even have particles?

 

[SpectraCat]

I believe that it is possible to reproduce the results using classical E&M waves, but I haven't finished the math yet. However, I can say that there is nothing in the experiment (aside from the single photon source) that suggests any problem with a classical E&M description. That is, if the quantum dot were replaced by a CW laser, I don't see why the results would be any different. I think the only interpretive issues arise from knowing that single photons travel through the apparatus one by one.

 

[Varon]

I believe that it is possible to reproduce the results using classical E&M waves, but I haven't finished the math yet. However, I can say that there is nothing in the experiment (aside from the single photon source) that suggests any problem with a classical E&M description. That is, if the quantum dot were replaced by a CW laser, I don't see why the results would be any different. I think the only interpretive issues arise from knowing that single photons travel through the apparatus one by one.

I asked Ken:

Varon: They can do weak measurement on a particle before full collapse. This means the particle has trajectory in contrast to pure Copenhagen concept where a particle only pops up upon collapse of the wave function (which stand for wave of possibility of where the particle would be detected).

Ken G: "It doesn't mean that. The second article gives a much more nuanced description than the first. Nothing in that experiment is the trajectory of an individual photon, instead, what they have seems to me is equivalent to what you'd get if you put the detecting screen at various different places and create a field of detection densities, attribute the detection densities to trajectory densities such as could be done with any divergence-free field, and draw the "field lines" and call them average trajectories. I'll wager doing that would generate precisely the same figure. Much ado about nothing.

What they seem to be missing is that the classical picture of waves going through two slits could generate the same figure. What makes the quantum realm so weird is the quantization-- not the averaged behavior. I really don't see what "weak measurement" is adding to the question, it still is not true that you can say which slit any of those electrons went through."

------------------------------------------

SpectraCat. Do "detection densities", trajectory densities", "divergence-free field" got anything to do with the results of the latest experiment? Are these even standard terms? How do you understand them?

 

[unusualname]

I believe that it is possible to reproduce the results using classical E&M waves, but I haven't finished the math yet. However, I can say that there is nothing in the experiment (aside from the single photon source) that suggests any problem with a classical E&M description. That is, if the quantum dot were replaced by a CW laser, I don't see why the results would be any different. I think the only interpretive issues arise from knowing that single photons travel through the apparatus one by one.

This would be a publishable result if you can manage it. The trajectories predicted by Bohmian mechanics seem to be non-intuitive/non-classical.

A bohmian mechanics calculation for a few systems are here (pictures of the trajectories are on the last three pages)
Bohmian Trajectories for photons

 

[IllyaKuryakin]

This would be a publishable result if you can manage it. The trajectories predicted by Bohmian mechanics seem to be non-intuitive/non-classical.

A bohmian mechanics calculation for a few systems are here (pictures of the trajectories are on the last three pages)
Bohmian Trajectories for photons

So, as I read the answers, there are many who claim they can reproduce these results using only orthodox statistical QM, but none have done it. Until I see the math, I remain skeptical since the orthodox QM lacks the guiding equation that provides the basis for the calculated average trajectories in Bohmian Mechanics. If I've missed something here, please correct me, but please be prepared to do the math as I can't see how to get to a calculated average trajectory from a random Schrodinger waveform probability distribution.

I'm even more skeptical that the average trajectories can be calculated in advance using classical e-m waves. I can't even see how classical e-m waves could relate to the average trajectories of an ensemble of photons. I agree, if someone can do this, they should publish the results.

 

[SpectraCat]

@Ilya ... what you seem to be missing is that the trajectories themselves were NOT measured in the experiment. Rather, they were reconstructed mathematically based on the average momenta resulting from the comparison of the slightly different interference patterns measured in the two polarization channels.

What I am saying is that I don't see any reason why the classical wave analysis of the experiment wouldn't give rise to precisely the same interference patterns. Once you have those, the reconstructed average trajectories would be the same.

 

[unusualname]

Fig 1 (on page 9) of http://arxiv.org/abs/quant-ph/0102071

corresponds to the trajectories constructed in this experiment http://www.aip.org.au/Congress2010/Abstracts/Monday%206%20Dec%20-%20Orals/Session_3E/Kocsis_Observing_the_Trajectories.pdf

Now, is there a calculation from standard QM or EM to reproduce the plot?

 

[unusualname]

And I don't mean, can the intensities in each plane be reconstructed. That's not the same thing as constructing trajectories.

Obviously Maxwell's equations predict the correct intensity as standard QM does (otherwise one of them would be wrong), so SpectraCat above is probably thinking of intensity calculations.

Unless there is some simple limit process where the trajectories as plotted above come out in the limit of zero areas for the intensities?

 

[IllyaKuryakin]

Fig 1 (on page 9) of http://arxiv.org/abs/quant-ph/0102071

corresponds to the trajectories constructed in this experiment http://www.aip.org.au/Congress2010/Abstracts/Monday%206%20Dec%20-%20Orals/Session_3E/Kocsis_Observing_the_Trajectories.pdf

Now, is there a calculation from standard QM or EM to reproduce the plot?

Yes, an interesting extension of Bhomian Mechanics from fermions to bosons.

But your question is valid, "Now, is there a calculation from standard QM or EM to reproduce the plot?"

I've looked for such a calculation in published papers, but I can find none. Until I see such a calulation, with all due respect, I have to consider claims that the same average trajectories can be calculated from standard QM suspect, and claims that they can be calculated using classical e-m theory even more suspect. My appoligies if there is such a paper and I've simply missed it.

 

[IllyaKuryakin]

And I don't mean, can the intensities in each plane be reconstructed. That's not the same thing as constructing trajectories.

Obviously Maxwell's equations predict the correct intensity as standard QM does (otherwise one of them would be wrong), so SpectraCat above is probably thinking of intensity calculations.

Unless there is some simple limit process where the trajectories as plotted above come out in the limit of zero areas for the intensities?

Yes, you phrased this better than I did. There is a difference between intensity-EM, probable density-QM and avgTrajectory-deBB.

 

[unusualname]

We must be sure the experiment distinguishes between intensity calculations and trajectories as in the Bohmian Analysis.

If the experimental data is at a resolution such that a simple intensity calculation from classical EM in each plane would explain it then the experiment isn't that great.

 

[SpectraCat]

And I don't mean, can the intensities in each plane be reconstructed. That's not the same thing as constructing trajectories.

Obviously Maxwell's equations predict the correct intensity as standard QM does (otherwise one of them would be wrong), so SpectraCat above is probably thinking of intensity calculations.

Unless there is some simple limit process where the trajectories as plotted above come out in the limit of zero areas for the intensities?

Have you read the paper? Do you understand how the AVERAGE trajectories are reconstructed? They are reconstructed, as I already said, by comparing the interference patterns measured in the two polarization channels, and extracting the AVERAGE transverse component of the momentum. The equation used is (cf. eqn 2 on p.1172)

\frac{<k_x>}{|\vec{k}|}=\frac{1}{\zeta}\frac{I_R~-~I_L}{I_R~+~I_L}

where I_R and I_L are the intensities in the right and left-handed detection channels, at a particular point in the interference pattern, and zeta is an experimentally determined constant.

Note that the authors omitted the <> around k_x in the text of the paper, but it is there in their derivation provided in the supporting information.

Anyway, the point is that the only experimental observables used to create the AVERAGE trajectories are the interference patterns. So, if you can reproduce those, you are done .. using the same reconstruction produce will produce the same AVERAGE trajectories.

 

[unusualname]

Have you read the paper? Do you understand how the AVERAGE trajectories are reconstructed? They are reconstructed, as I already said, by comparing the interference patterns measured in the two polarization channels, and extracting the AVERAGE transverse component of the momentum. The equation used is (cf. eqn 2 on p.1172)

\frac{&lt;k_x&gt;}{|\vec{k}|}=\frac{1}{\zeta}\frac{I_R~-~I_L}{I_R~+~I_L}

where I_R and I_L are the intensities in the right and left-handed detection channels, at a particular point in the interference pattern, and zeta is an experimentally determined constant.

Note that the authors omitted the <> around k_x in the text of the paper, but it is there in their derivation provided in the supporting information.

Anyway, the point is that the only experimental observables used to create the AVERAGE trajectories are the interference patterns. So, if you can reproduce those, you are done .. using the same reconstruction produce will produce the same AVERAGE trajectories.

So go on then, reproduce the average trajectories they found, by instead using standard QM or EM calculations.

I'm feeling that nobody has bothered because nobody thinks it's been worthwhile. But now that we have an actual experiment with actual experimental data and a plot of these trajectories, perhaps someone should show how the trajectories can be explained more simply. And in particular, refute the outrageous claim that this paper is in any way vindicating Bohmian calculations of the trajectories.

 

[SpectraCat]

Fig 1 (on page 9) of http://arxiv.org/abs/quant-ph/0102071

corresponds to the trajectories constructed in this experiment http://www.aip.org.au/Congress2010/Abstracts/Monday%206%20Dec%20-%20Orals/Session_3E/Kocsis_Observing_the_Trajectories.pdf

Now, is there a calculation from standard QM or EM to reproduce the plot?

As far as I can tell, the resemblance is coincidental, but I am far from an expert on BM. Perhaps Demystifier could comment further on this? The plot from the BM paper shows actual single-particle trajectories, where as the experimental data is for average trajectories. My guess is that the dependence of the average momenta measured experimentally on the phase of the interference pattern is similar to the dependence of the Bohmian velocities of the individual photons on the phase of the pilot-wave interference pattern in the BM simulation.

One additional question that seems like the elephant in the room to me is the fact that Bohmian trajectories seem to predict that photons NEVER cross the center line between the slits. Thus, photons going through the left slit contribute ONLY to the left half of the interference pattern, and vice versa. So what I don't understand is how that is compatible with the usual double-slit interpretation, which says that knowing which path the photons went through should destroy the interference pattern. Maybe I should ask that question on another thread ...

 

[unusualname]

Yes probably, but if anyone wants internet fame for a few days and a publication, just calculate the plots using standard QM or classical EM.

 

[Varon]

Yes probably, but if anyone wants internet fame for a few days and a publication, just calculate the plots using standard QM or classical EM.

If there's someone who can do it. It's Ken G. I asked him to participate here as he is not aware this thread exists.

Ken. How hard or long would it take to calculate the plots using standard QM or classical EM? Come on, do it to demystify the latest experiment as it confused the hell out of many quantum enthusiasts.