PBR no-go theorem: Is ψ epistemic or ontic? An experimental test

[bohm2]

This was a recent experiment done to determine the epistemic versus ontic status of ψ:

If a physicist claims to have produced a system with a particular wave function, does this represent directly a physical wave of some kind, or is the wave function merely a summary of knowledge, or information, about the system? A recent no-go theorem shows that models in which the wave function is not physical, but corresponds only to an experimenter's information about a hypothetical real state of the system, must make different predictions from quantum theory when a certain test is carried out. Here we report on an experimental implementation using trapped ions. Within experimental error, the results confirm quantum theory. We analyse which kinds of theories are ruled out...The fundamental question of the interpretation of the wave function of a quantum system was partially resolved. If systems have real states, regardless of an experimenter or measurements performed, then a natural question is whether the quantum state is epistemic, i.e. corresponding merely to knowledge of these underlying real states. In the presented manuscript we tested for this specific possibility and ruled out the most natural class of such models to a high degree of confidence.

Can different quantum state vectors correspond to the same physical state? An experimental test
http://lanl.arxiv.org/pdf/1211.0942.pdf

 

[Demystifier]

Do the results of this experiment tell as anything that we didn't already know?

Is any SPECIFIC previously used interpretation of QM ruled out?

I think the answer to both questions is - NO.

 

[bohm2]

Do the results of this experiment tell as anything that we didn't already know? Is any SPECIFIC previously used interpretation of QM ruled out?
I think the answer to both questions is - NO.

It arguably rules out the Rob Spekkens/Leifer ψ-epistemic models.

 

[Demystifier]

It arguably rules out the Rob Spekkens/Leifer ψ-epistemic models.

But did anybody (except perhaps Spekkens and Leifer) seriously used these models and honestly believed that they might be correct?

 

[Quantumental]

It arguably rules out the Rob Spekkens/Leifer ψ-epistemic models.

I fail to see that they have been ruled out?

Personally I think that some deeper theory that will unify QM and GR will still be "found" some day. I just don't believe that either Bohm, Everett or any of the other "candidates" are right.

 

[bohm2]

I fail to see that they have been ruled out?

Why? I would imagine that the authors would argue that for all intensive purposes ψ-epistemic models are ruled out even though some unlikely and/or far-fetched loopholes mentioned in the article still exist (like in Bell's). Interestingly, some of the authors (e.g. Rudolph) were ψ-epistemic enthusiasts. Spekkens appears to be a hold-out, though. I'm not sure if this experiment would change his position. Beyond these models suggested by these authors, aren't there other ψ-epistemists, also?

EDIT: Another experiment drawing similar conclusions just published:

Assuming a natural continuity assumption and a separability assumption, we show here that epistemic interpretations of the quantum state are in contradiction with quantum theory. Our argument is different from the recent proof of Pusey, Barrett, and Rudolph and it already yields a non-trivial constraint on epistemic models using a single copy of the system in question. We then present a simple high-precision experiment based on high-dimensional attenuated coherent states of light traveling in an optical fibre which tests the predictions of such epistemic models. Our experimental results are in agreement with the predictions of quantum theory and provide strong constraints on possible epistemic extensions of quantum mechanics.

Experimentally probing the reality of the quantum state
http://lanl.arxiv.org/pdf/1211.1179.pdf

 

[Ilja]

I would like to point to the fact that epistemic interpretations which correspond to the dBB rule of defining effective wave functions for subsystems by $\psi(q,t)=\Psi(q,q_{env}(t))$, where $q_{env}(t)$ is the trajectory of the configuration, cannot be ruled out by similar arguments if they give the $\Psi$ of the whole universe an epistemic interpretation.

The point is that all the properties of the wave function of any accessible system which make it look like a physical entity it borrows in that formula from $q_{env}(t)$.

In particular, the prescription to make the PBR experiment can, from this point of view, never create a state where it is unclear how it has been prepared. The prepared state is always part of greater system, which contains the devices used for the preparation. This environment can, of course, also include some random throwing of dices which preparation procedure has to be used. But, anyway, $q_{env}(t)$ contains also the result of this dice throwing.

So $\psi_1(q)=\Psi(q,q_{prep\uparrow},\uparrow)$ and $\psi_2(q)=\Psi(q,q_{prep\downarrow},\downarrow)$, thus, even if ψ itself is purely epistemic, the λ_i of the two states will be different already because their preparation procedure is different.

 

[bohm2]

So I'm guessing, you are arguing that PBR would have no effect on your model?

The Paleoclassical interpretation of QT
http://lanl.arxiv.org/pdf/1103.3506.pdf

 

[DrChinese]

Considering that the Bohmian view* essentially says that there IS a specific well-defined value for non-commuting observables, although we cannot know it: I would think it is just a matter of time before theorems in the spirit of PBR clearly conflict with the Bohmian hypothesis (non-local hidden variables).

*As I understand it, that being a limited understanding.

 

[Ilja]

So I'm guessing, you are arguing that PBR would have no effect on your model?

The Paleoclassical interpretation of QT
http://lanl.arxiv.org/pdf/1103.3506.pdf

At least at the current moment I think so.

I would not completely exclude that it nonetheless may cause problems. The paleoclassical interpretation is, last but not least, a radical version where the wave function of the whole universe has no own real degrees of freedom. And one has to think about PBR if one constructs a subquantum theory which gives QT in some limit.

But I argue, indeed, that PBR is not an impossibility proof for the paleoclassical interpretation.

Considering that the Bohmian view* essentially says that there IS a specific well-defined value for non-commuting observables, although we cannot know it: I would think it is just a matter of time before theorems in the spirit of PBR clearly conflict with the Bohmian hypothesis (non-local hidden variables).

No, there is only a well-defined value for the configuration q(t). Momentum is not defined. The result of momentum measurements depends on the configuration of the momentum measurement device, except in the case of a momentum eigenstate.

But why you think that an impossibility theorem may be possible? That dBB in quantum equilibrium gives QT predictions is a simple theorem. That quantum equilibrium will be approached is Valentinis subquantum H-theorem - ok, with caveats similar to the original H-theorem, but I cannot see how this may become a problem. And I see no reason to doubt that dBB in itself is consistent.

The wave function as being real is at least a popular variant, if not the standard dBB interpretation.

 

[bohm2]

Considering that the Bohmian view* essentially says that there IS a specific well-defined value for non-commuting observables, although we cannot know it: I would think it is just a matter of time before theorems in the spirit of PBR clearly conflict with the Bohmian hypothesis (non-local hidden variables).

I think most Bohmians treat the wave function as ontic, so they would rejoice in the PBR theorem? In fact, Valentini referred to PBR at "the most important general theorem relating to the foundations of quantum mechanics since Bell's theorem". Having said that, I'm not sure if PBR would have any implications for Bohmians who treat the wave function as "less ontic" (more nomological) like the DGZ group/Maudlin? I doubt it, though, since there is no essential place for any notion of probability in Bohmian mechanics, at least, regarding the basic description of particle motion. But, I haven't seen anyone from this group comment on the PBR theorem even though I've been searching.

 

[DrChinese]

...

1. No, there is only a well-defined value for the configuration q(t). Momentum is not defined. The result of momentum measurements depends on the configuration of the momentum measurement device, except in the case of a momentum eigenstate.

...

2. The wave function as being real is at least a popular variant, if not the standard dBB interpretation.

1. An element of dBB I usually forget about... I guess no well-defined spin either.

2. Bohm2 said essentially the same thing. I guess I just don't understand that. If there are (nonlocal) hidden variables that determine outcomes, how can a probability amplitude/wave function have reality? I thought outcomes could be predicted if all input values were known. Seems to me that HV implies a form of determinism which in turn implies well-defined values for elements of reality.

 

[Ilja]

1.. I guess no well-defined spin either..

The simplest approach is indeed without spin.

There have been different approaches to relativistic QFT for fermion fields, from Bell (beables for quantum field theory) to Colin and Struyve.

But I favour, of course, my own approach. In arXiv:0908.0591 I present a way to obtain two Dirac fermions together with a scalar field from canonical quantization of a scalar field with W-like potential. Once scalar fields in itself present no problem for dBB field theory, this solves the problem even if dBB is not even mentioned there.

2. If there are (nonlocal) hidden variables that determine outcomes, how can a probability amplitude/wave function have reality? I thought outcomes could be predicted if all input values were known. Seems to me that HV implies a form of determinism which in turn implies well-defined values for elements of reality

ψ is in dBB theory not a probability amplitude, but a field with guides the configuration q(t).

There is only a special state, named quantum equilibrium, where we have a connection between ψ and probabilities for q in form of the Born rule. It is because of a "subquantum H-theorem" that arbitrary probability densities become, after some time, almost indistinguishable from this quantum equilibrium. So there is no fundamental connection at all.
Probabilities play the same role as in classical thermodynamics.