Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Observation vs measurement

  1. Jan 11, 2012 #1
    what is the difference between observation and measurement in reference to quantum mechanics?
     
  2. jcsd
  3. Jan 11, 2012 #2

    Demystifier

    User Avatar
    Science Advisor

    Observation requires a conscious observer, measurement doesn't.
     
  4. Jan 11, 2012 #3
    Demystifier's reply is clever and a convenient intutive reminder...

    but Wikipedia sees no real distinction:


    Also check this out for information on "observable" :

    http://en.wikipedia.org/wiki/Measurement_in_quantum_mechanics#Quantitative_details

    If it suits your purposes, also think about "preparation".

    lots of additional links at the bottom of the article.
     
  5. Jan 11, 2012 #4

    Ken G

    User Avatar
    Gold Member

    I'd have to agree that there is no established scientific difference between an observation and a measurement, other than the implication that a measurement is quantitative and an observation can be more informal (I "observe" that the Moon is full tonight, etc.). One might argue that one has to do an observation following every measurement to see how the measurement came out, but again the distinctions seem scientifically unimportant. Whether or not either of those needs to be carried out by a conscious individual is a question of subjective philosophical priorities. For example, we can maintain that only conscious beings do observations, or we can maintain that an observation occurs whenever we could imagine a conscious being on the scene without affecting the outcome in any scientifically relevant way (but still need the concept of conscious beings), or we can maintain that conscious beings have nothing to do with measurements, and in a universe devoid of any life, one rock can still be twice the size of another one. I can't really see the latter as a coherent perspective personally (rocks have no idea what "size" even means), but I see nothing in science that adjudicates these possibilities, and indeed that is a good thing, because it allows us all to agree on scientific results even though we don't agree on these philosophical priorities.
     
  6. Jan 11, 2012 #5
    In regards to a macroscopic object, an observable in QM would be every bit of the object we can see (not just position, speed, etc)?
     
  7. Jan 11, 2012 #6

    Ken G

    User Avatar
    Gold Member

    I think one has a pretty hard time putting the meaning of an "observable" of a macro object onto a sound quantum mechanical basis. What is the quantum mechanical meaning of the position of a macro object? If we say it is its center of mass, then we must admit that the center of mass is contributed to by the locations of all its parts, but then the location of the COM is not itself a measurement, but rather some average of measurements on the locations of the parts. The same holds if we measure the gravitational field of the object to locate its COM-- the field is contributed to by all the parts, it is not something the object itself generates in some holistic way (at least, no one seems to think it is). The terms become inescapably vague, I believe that is the entire justification behind the "correspondence principle" and the Copenhagen "cut" between what is macro and micro.
     
  8. Jan 12, 2012 #7
     
  9. Jan 12, 2012 #8
    so who tells a quantum system to collapse?
     
  10. Jan 12, 2012 #9

    Ken G

    User Avatar
    Gold Member

    If we adopt the lexicon that a measurement is a quantification and an observation is a perception (which is a kind of qualia but can refer to a quantity, like the qualia of a quantity), then which one "collapses" the wave function depends on our interpretation of QM. In the "consciousness collapse" picture, which essentially places the collapse at the "last possible moment" but seems too mystical to many physicists (I have no idea why, I think they may have lost track of their own mysticisms), then we would hold that the observation causes the collapse, not the measurement. Or for those who favor interpretations in which collapse is a physical dynamical phenomenon like any other (so independent of our consciousness), the measurement causes the collapse (in the sense that a measurement involves coupling with macroscopic systems, regardless of whether or not any of them include a brain). Then of course we have orthogonal views like Bohm and MWI-- in Bohm, the system starts out collapsed, so what we perceive as collapse is simply when we get the message, and in MWI, it starts out uncollapsed and never collapses, our perceptions are a kind of pervasive illusion. (To which I would argue that when an illusion is pervasive enough, we don't call it an illusion any more.)

    Which of these is the "real truth"? Well, shouldn't we know by now that none of them are, but they each give us a way of thinking about what is happening? It might seem the most natural to use either Bohm or consciousness collapse, because those place the collapse either as soon as possible, or as late as possible, and putting it anywhere else is rather arbitrary. But we should probably expect arbitrariness in interpretations, so it's not really a problem that they all place the collapse somewhere different, and so the real message is probably that either our current theory has no way to say when the collapse happens, or perhaps nature itself leaves the issue among its many indeterminacies. In any event, the theory of QM does not appear to answer the question, so we must await the next testable theory and see if it does any better.
     
    Last edited: Jan 12, 2012
  11. Jan 12, 2012 #10
    if collapse occur by observation , how could Schrodinger equation give solution with oscillating probability amplitudes? shouldn't it be just peaks at every point observed(collapse)
     
  12. Jan 12, 2012 #11

    Ken G

    User Avatar
    Gold Member

    I'm not sure what you are asking, it sounds like you are mixing the cases where the outcomes are treated as discrete vs. continuous. The above issues appear in either situation, and either type of probability amplitude can oscillate, but they are different situations that should not be confused.
     
  13. Jan 12, 2012 #12
    no i am not asking that
    what i mean to say is that observation is causing collapses,how could it be that a wave function is evolving as given by Schrodinger equation under non collapsible condition when the only way to see that evolution is to observe it
     
  14. Jan 12, 2012 #13

    Ken G

    User Avatar
    Gold Member

    Ah, you are asking about how we connect the Schroedinger equation to the observations. That's pretty much what all the hullaballoo is about with the various interpretations, right there. It's the main bugbear of quantum mechanics! I won't go into all the interpretations, each one is pretty much a thread of its own, so I'll just say how the theory itself handles it-- via the "Born rule", which converts the wavefunction (from the Schroedinger equation) to the probability of some given outcome (the "collapse"). This means, your answer is not in the Schroedinger equation itself, it is in what we do with the output from the Schroedinger equation. If this seems somewhat ad hoc to you, you are not alone-- many physicists have worked very hard to improve the situation with a more general set of postulates that makes the Born rule seem to follow more naturally, but none have achieved widespread acceptance, so the current state of the art is to simply plug in the Born rule and not worry where it comes from. (The Born rule says that the probability of a given measured outcome is the squared magnitude of the associated coordinate of the wavefunction when expanded on the eigenbasis of that measurement operator.)

    By the way, someone might come along and say that the "collapse" is indeed in the Schroedinger equation, but can be treated with "decoherence", which says that we don't treat the details of the interaction with the macro system, we just kind of statistically average over the possibilities (sort of like replacing the true Schroedinger equation with an average version, and interpret the outcomes statistically, as we would do in statistical mechanics of a gas). However, this misses the point that when you do that, you still only get what is called a mixed-state outcome for your description of a given actual system-- you still don't get an accounting for the fact that we perceive a single eigenvalue of the measurement, rather than a mixture of them. That is the actual place where all the interpretations differ-- in the meaning of a mixed state.
     
    Last edited: Jan 12, 2012
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Observation vs measurement
Loading...