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

What does to observe mean?

  1. Oct 14, 2011 #1
    What does "to observe" mean?

    I am pretty sure that you will recognise this video :
    about at 3:50, they claim that, when we observe an electron passing through a double slit, acts as a particle, as opposed to its wave property. Does that mean, if we place a 8 mega pixel digital camera in between the slit and the light source, will electrons behave as particles? I hardly think so. SO my question is, what is the context of observation in this video?
    Last edited by a moderator: Sep 25, 2014
  2. jcsd
  3. Oct 14, 2011 #2
    Re: What does "to observe" mean?

    See here:


    An electron is a particle AND a wave, but both are not simultaneously observable.

    It's a reflection of complementarity....or wave-particle duality.

    What you observe depends on how you measure!!!!

    If you think any of the is wierd, you are correct...as compared with classical large scale observations..... check out "quantum erasure" and "delayed choice" for even stranger
    observer/system relationships.
  4. Oct 16, 2011 #3
    Re: What does "to observe" mean?

    An observation interferes with the current state of the electron. If the electron is in some (initially superimposed) state, then performing a measurement will "throw" it into one of the constituent Eigenstates. This is what "destroys" or changes the behaviour of the electron as it passes through the slit.

    If you think about it, an electron just going through the slit is not the same as the electron being detected and then going through the slit, because the latter involves an interaction process in between (with the measuring apparatus).
  5. Oct 16, 2011 #4
    Re: What does "to observe" mean?

    Observation means that one of the eigenstates become entangled with the measurement instrument, which then become the observed value. It does not require a humean being observing the result to collapse a wavefunction. That is just a stupid philosophy. Quantum states are not particles, they just decohere into smaller waves.
  6. Oct 16, 2011 #5
    Re: What does "to observe" mean?

    We see an electron as a wave before we observe it because, at the quantum scale, you can argue that an electron is here and there so the concept of an electron being a wave is philosophical. When we observe the electron we exert energy on it and it appears that the electron "realizes" it is being observed and acts up. So to observe is simply to exert energy on something.
  7. Oct 16, 2011 #6
    Re: What does "to observe" mean?

    I'd say it's more than just philosophical since the electron shows interference patterns which charecterizes a wave.
  8. Oct 16, 2011 #7
    How exactly do we exert energy on an electron by looking at it?
  9. Oct 16, 2011 #8
    Re: What does "to observe" mean?

    We're not "looking" at it, it is reacting with the measurement instrument. Probably with a force. Allthough I have to admit I'm not aware of exactly how such a measurement is performed or with what kind of measurement instrument.
  10. Oct 16, 2011 #9
    Re: What does "to observe" mean?


    The classical reasoning is as follows: what does it mean to see something? You don't just see things by default, it's rather a process that ends with your eyes taking in photons, photons that scattered of the thing you "see". Without the photons scattering of the object, you wouldn't be able to see it, so you see a measurement really entails some kind of physical interaction with the system (and you also see it's not necessary for you to actually register the photons to do a measurement[*]).

    Now that's a classical idea, however, since QM doesn't talk about particles being hit by photons or what have you. QM is a bit more abstract, and to describe the measurement of seeing a particle (passing a certain slit) you have to do some math, which in the end suggests that the measurement process has unnegligibly altered your system.

    But what "actually" happens is still up for discussion:

    I hope this helps somewhat.

    [*] This sentence has to be taken with care, it depends on what interpretation of QM you're using... The truth value of my statement is a big part of the so-called "measurement problem".
  11. Oct 17, 2011 #10
    Re: What does "to observe" mean?

    What counts as an observation, there is no agreed upon definition.

    By putting a camera by each slit, the camera either (a) collapses the wave function of the particle to a definite state or (b) the camera and the particle become entangled and no collapse occurs. The apparent loss of interference is because two interference fringes add up to create a scatter pattern (see Quantum Erasure experiment).
  12. Oct 17, 2011 #11
    Re: What does "to observe" mean?

    This is a very misleading comment, it gives the camera a powerful status: the camera is just there to register photons. The collapse will only occur / has occured (depends on your interpretation...) if there are photons for the camera to see. The camera is just one element in a long chain of interactions that constitute "measurement", not a key element as your quote suggests (and while you might understand what you meant, I don't want the OP to get confused).
  13. Oct 17, 2011 #12
    Re: What does "to observe" mean?

    "Now that's a classical idea, however, since QM doesn't talk about particles being hit by photons or what have you. QM is a bit more abstract, and to describe the measurement of seeing a particle"

    That is incorect. Richard Feynamn goes into great detail how electron/photon interactions cause interferance collapse regarless of "seeing" a particle. Yes maths are used to describe the process but it still requires a physical interaction ot collapse a wave function.
  14. Oct 17, 2011 #13
    Re: What does "to observe" mean?

    @Thenewdeal38: I get what you're saying, but that's just one interpretation of the math, it's not in the math itself. If it were really that easy, then we wouldn't have a "measurement problem".
  15. Oct 17, 2011 #14
    Re: What does "to observe" mean?

    There isnt a measurment problem between real physicists. Only between real scientists and sudo quack new agers who misenterpreted bohr's original use of the word "observe". He even stated that type of language would obviously be misenterpreted by the layman and that QM has nothing to do with cousciesness.
  16. Oct 17, 2011 #15
    Re: What does "to observe" mean?

    I think you misunderstand the idea of the measurement problem, it has nothing to do with a consciousness.
    Granted, one suggested solution to the measurement problem uses the concept of consciousness (advocated by nobel laureate Wigner[*]).

    If you denote "experimental physicists" with "real physicists", then I'd agree, but if you also include "theoretical physicists", then I disagree.

    To get a better view I'd suggest reading the entertaining and insightful article "Six possible worlds of quantum mechanics" by John Stewart Bell, a great physicist.

    [*] not that it has to be true because it comes from a nobel laureate, but just as an indication that he isn't a "quack"
  17. Oct 17, 2011 #16
    Re: What does "to observe" mean?

    How can measurement device somehow affect the result just by watching? Photons bounce of the object to the measurement device and that's it.
    A bounce of a ball isn't influenced by a person who catches the ball 3 seconds after.
  18. Oct 17, 2011 #17
    Re: What does "to observe" mean?

    I would say an observation is an interaction between the system (observed) and an external apparatus causing the wave function to collapse. I know it's pretty vague but I think that the real question concerning quantum measurement is who/what can observe? Is it only human beings, conscious beings, animals, environment? :s
  19. Oct 17, 2011 #18

    Ken G

    User Avatar
    Gold Member

    Re: What does "to observe" mean?

    That isn't really quite it, though-- it's more tricky than that. Take for example this experiment. You have a particle in a 1D box, in its ground state. You shine a super bright light on one-half of the box, and get no detection-- no light bounces off the particle in the box, the particle is not found to be on that side of the box. What is the expectation value of the particle energy now, is it still in its ground state?

    No, it is not. The new information we have about the particle wavefunction is that the particle wavefunction is zero on one-half the box. The rest of the box must hold the full wave function, but with the same shape it had in its ground state-- we just scale up the normalization. Now a truncated ground state wavefunction is not a ground state wavefunction any more-- the energy of the particle has increased (in expectation value). Where did that additional energy come from? Information is important in quantum mechanics, not just interactions.
  20. Oct 17, 2011 #19
    Re: What does "to observe" mean?

    None of the above. The observer is an uncounsciess meausuring device. And to observe is to physiclly interact with something where mathematical deductions based on "the bounce back" of the interacting subatoms reveal location and or position. It does not mean to look at and examine. OVERSIMPLIFICATIONS OF THE OBSERVER EFFECT ARGHHHHHHHH!!!!

    Also define in precise terms of QM: Information?
  21. Oct 17, 2011 #20

    Ken G

    User Avatar
    Gold Member

    Re: What does "to observe" mean?

    Even this is an oversimplification. If we could really get away with such a simple meaning, we would not need any interpretations of quantum mechanics at all (some people feel this is in fact the case, but in practice, some interpretation is always needed). The problem with the above is it is not consistent with quantum mechanics to simply define observation as interaction. This is for two reasons:
    1) as I mentioned just above, some types of information updates involve no interactions with a measuring device at all, and the absence of interaction then becomes a type of interaction, so it's a bit more subtle, but worse:
    2) according to quantum mechanics, interactions only produce entanglements. So we could think of observations as entanglements with measuring devices, but that only produces mixed states, again according to quantum mechanics. The "measurement problem" actually comes next-- what does the mixed state mean? The "problem" here is that we never actually perceive mixed outcomes, we perceive definite outcomes. So entanglements with measuring devices really doesn't cut it-- we have to go beyond the entanglement into the definite outcome.

    That last step, that many people simply don't recognize, is also where interpretations come in-- to some, no "collapse" ever occurs, and all we have is an entanglement, not a definite outcome-- but that's MWI. To others, a collapse does occur because it is seen to occur, and that has to be tacked onto QM separately-- that's CI. To others, the collapse was always there, in the initial conditions, we just have no way to see it until the experiment is over-- that's deBB. To still others, QM was never intended to provide a complete description, only a statistical one, so there is simply no need for QM to account for collapse-- that's the "ensemble interpretation." Some might even be fine with all 4 interpretations, and others, they simply see them as different angles from which to view QM and not to be taken very seriously (that's my own personal stance).

    But one thing is clear-- the only reason that final step ever comes up at all is because we are conscious beings that think and do science. This is simply an undeniable fact-- if we were not conscious thinkers (I don't attempt to parse the differences in thinking and being conscious), neither an interpretation of QM, nor even QM itself, would ever be necessary or ever exist.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook