Can quantum superpositions be observed

[Trollfaz]

I heard about something called interaction free measurement which allows us to measure a quantum system without disturbing it. Now I am wondering if this can allow us to observe how an object is like in a superposition because no interaction means no wavefunction collapse.
https://en.m.wikipedia.org/wiki/Interaction-free_measurement
For reference

 

[PeterDonis]

no interaction means no wavefunction collapse

Wave function collapse is an interpretation of QM; it is not part of the basic theory. It's a good idea not to tie your understanding of QM to any particular interpretation, since they all make the same experimental predictions.

As far as interaction free measurements are concerned, calling them "measurements" to begin with is a bit problematic. Basically, they allow you to obtain indirect evidence about the quantum state of an object; they aren't a direct observation of the object the way, for example, looking at photons that bounced off the object is.

Finally, you should be aware that whether a given quantum state is a superposition is basis dependent; it is not an absolute property of the state. So the title question of this thread isn't really well defined.

 

[StevieTNZ]

Can't you see quantum superposition doing the double-slit experiment?

 

[PeterDonis]

Can't you see quantum superposition doing the double-slit experiment?

You see interference in the double slit experiment. Interference is not the same as superposition.

 

[Physics Footnotes]

I heard about something called interaction free measurement which allows us to measure a quantum system without disturbing it. Now I am wondering if this can allow us to observe how an object is like in a superposition because no interaction means no wavefunction collapse.

It is best to think of a quantum state not as something that is observed, but rather as something that is inferred. That is, you observe objective events, and then you infer what kind of state is consistent with those events; there is no concept of 'observing the state'. This remains true even in interaction-free measurements in which inferences about the state can be made on the basis of an observation of a non-event (e.g. a detector not firing).

With this in mind, a state that we might describe as a 'superposition of distinct possibilities' is inferred from the same reasoning whether the measurements are interaction-free or not. Namely, through the existence of interference effects.

 

[StevieTNZ]

You see interference in the double slit experiment. Interference is not the same as superposition.

But the interference is a result of the quantum system going through the left and right slits (not in the classical sense).

 

[PeterDonis]

the interference is a result of the quantum system going through the left and right slits

Yes, but we don't observe the particle in a superposition of going through the left and right slits (and you appear to recognize this, as your parenthetical note indicates). We only observe the interference pattern built up by multiple particle impacts. We infer (as @Physics Footnotes suggests) that the quantum state of each particle is (or better, can be modeled as) a superposition of "going through the left slit" and "going through the right slit", because that model gives us quantum amplitudes that correctly predict the interference pattern.

 

[vanhees71]

In the usual approximate treatment of the double-slit experiment the interference pattern is simply the solution of the corresponding boundary-value problem of the Schrödinger (or rather the corresponding Helmholtz) equation. In odd-dimensional spaces with $d \geq 3$ the Huygens principle holds, and thus it's legitimate to talk about superpositions of the amplitudes for the particles taking the one or the other path to the screen. This view is even more emphasized when using the equivalent Feynman path-integral formulation.

This is pure math and must not be mixed up with the wrong interpretation of the Schrödinger wave function as a classical field. The today accepted and empirically approved interpretation of the wave function is the probabilistic interpretation given by Born in 1926.