Difference between measurement and interaction

[Gerinski]

After reading quite many popular science about QM, I still didn't get a real understanding of the difference between a so-called measurement and just an interaction (if there's actually any difference).

My understanding is that a measurement is an interaction "observed" in a way which allows to acquire information about it. The information itself does not need to be acquired by a sentient being, could just be recorded by a data recorder.
Just the fact that the information is there, and the possibility that at any eventual time it could be used, makes it a measurement and would for example destroy interference.

Just an interaction would be an interaction which is not recorded in any way, so that it will never be possible in the future to trace back information about what happened. In this case interference would not be destroyed.

Is this more or less correct? and if it is, does it not certainly and unavoidably put onto the table the famous role of consciousness in QM?

 

[vanesch]

After reading quite many popular science about QM, I still didn't get a real understanding of the difference between a so-called measurement and just an interaction (if there's actually any difference).

It is indeed the whole problem !

My understanding is that a measurement is an interaction "observed" in a way which allows to acquire information about it. The information itself does not need to be acquired by a sentient being, could just be recorded by a data recorder.
Just the fact that the information is there, and the possibility that at any eventual time it could be used, makes it a measurement and would for example destroy interference.

Indeed, that's a way to see things. The whole problem resides in what "happens to the system", and as you point out yourself, a "measurement" is nothing else but an interaction ; and the problem is of course that if you treat it as a "measurement" you need to apply a projection (process 1 according to von Neumann), while if you treat exactly the same operation as an interaction, you have a unitary time evolution (process 2 according to von Neumann).

Just an interaction would be an interaction which is not recorded in any way, so that it will never be possible in the future to trace back information about what happened. In this case interference would not be destroyed.

It is even worse: if you treat whatever "records the information" also as a quantum system, you see that there is no *record* of any particular event, but just an entanglement of both state vectors (the one of the recording device, and the one of the system under study). And if you include YOURSELF in the 'recording system', you have nothing else but the many worlds interpretation of quantum theory !

Is this more or less correct? and if it is, does it not certainly and unavoidably put onto the table the famous role of consciousness in QM?

That's what I also think. von Neumann and Wigner were also of that meaning. Then, a lot of people don't buy this. You're well on your way to get "entangled" into the interpretational problems of quantum theory :-)

cheers,
Patrick.

 

[DaTario]

Possibly, some special kind of interaction deserves the name on measurement interaction. I really don't believe in such artificial projection mechanism. According to the work of G. Rempe, entanglement must be involved in the answer to this problem. As the system shares its information content with some leaky environment, i.e, with subsystems which can only be described termodynamically, the reduced density matrix experiences what we may call as a projection. But the information has not been exactly lost.

Best Regards

DaTario

 

[Juan R.]

After reading quite many popular science about QM, I still didn't get a real understanding of the difference between a so-called measurement and just an interaction (if there's actually any difference).

My understanding is that a measurement is an interaction "observed" in a way which allows to acquire information about it. The information itself does not need to be acquired by a sentient being, could just be recorded by a data recorder.
Just the fact that the information is there, and the possibility that at any eventual time it could be used, makes it a measurement and would for example destroy interference.

Just an interaction would be an interaction which is not recorded in any way, so that it will never be possible in the future to trace back information about what happened. In this case interference would not be destroyed.

Is this more or less correct? and if it is, does it not certainly and unavoidably put onto the table the famous role of consciousness in QM?

I will say is not precise, but can help to you (i suspect you are not an expert).

Any interaction between two quantum systems may follow QM. Schrödinger equation is valid.

A measurement is not explained from QM. The process does not follow Schrödinger equation. There exists several proposals for new equations beyond QM: Ito-Schrödinger, Caldeira-Legget, Penrose gravity, Prigogine theory, etc.

All of those have well-defined problems.