The Naked Scientists

The Naked Scientists Forum

Author Topic: In quantum mechanics, does the observation preceed the event?  (Read 16939 times)

Offline PeteJ

  • First timers
  • *
  • Posts: 6
    • View Profile
Hi folks. My first post here. I have a question, most probably a naïve one...

In QM, whether we favour MW or CI as an interpretation, there is the idea that an observation causes an event, either a 'collapse' or a splitting of worlds.

Does the observation preceed the event, or is it simultaneous with the event?   

Or something else..?

Thanks.




« Last Edit: 08/05/2015 15:50:49 by chris »


 

Offline alancalverd

  • Global Moderator
  • Neilep Level Member
  • *****
  • Posts: 4721
  • Thanked: 155 times
  • life is too short to drink instant coffee
    • View Profile
Re: A Measurement Problem Question
« Reply #1 on: 08/05/2015 12:06:20 »
An observation cannot logically precede the observed event, and given the finite speed of light (which defines the limiting speed for transmission of information) it cannot even be simultaneous.

It is important to distinguish between the process of observation, which could indeed precipitate an event (say by a photonuclear interaction), and the resulting observation itself (an emitted electron or whatever).  Bouncing hypothetical photons off particles does indeed lead to Heisenberg's indeterminacy equation but isn't the cause of indeterminacy.

Beware of confusing a mathematical model (wave function collapse, world splitting, whatever) with reality. The model is merely our best shot at predicting the outcome of a future event.
 

Offline PeteJ

  • First timers
  • *
  • Posts: 6
    • View Profile
Re: A Measurement Problem Question
« Reply #2 on: 08/05/2015 12:12:07 »
Thanks Alan, very clear.

It seems to me that you're right. But it doesn't quite answer my question.

If the observation is not prior to the event and not simultaneous, then what?

How can the process of observation be different from the observation?
 

Offline alancalverd

  • Global Moderator
  • Neilep Level Member
  • *****
  • Posts: 4721
  • Thanked: 155 times
  • life is too short to drink instant coffee
    • View Profile
Re: A Measurement Problem Question
« Reply #3 on: 08/05/2015 12:22:58 »
Process of observation: kick it

Observation: landmine explodes
 

Offline PeteJ

  • First timers
  • *
  • Posts: 6
    • View Profile
Re: A Measurement Problem Question
« Reply #4 on: 08/05/2015 13:28:48 »
Hmm. That does not help. The landmine explodes well before we observe the explosion, and if we sense the kick connecting then that is an observation.

I know my question is naïve but it's proving difficult to find an answer. I've tried elsewhere but had no luck.
 

Offline David Cooper

  • Neilep Level Member
  • ******
  • Posts: 1505
    • View Profile
The following is speculation:-

What I suspect happens is that the wavefunction is only able to be sustained up to a certain level of complexity beyond which it must burst (simplify). By creating data in a brain or computer to represent what is observed, the complexity increases because the computer has to be maintained in multiple states to maintain the data in multiple states, so that is where the simplification is most likely to be forced to take place, after which a matching simplification will be forced to take place in the original object being observed to match up to the data which describes it.
 

Offline lightarrow

  • Neilep Level Member
  • ******
  • Posts: 4586
  • Thanked: 7 times
    • View Profile
If the observation is not prior to the event and not simultaneous, then what?
Usually with "event" in this context, in the sense of "measure event", we mean the physical interaction with a measure device. "Observation" is the subjective act of take awareness of the event from a sentient being.
The first can come without the (unnecessary) second.
Someone has speculated on the possibility that the act of observation could modify the wavefunction, but it's not the main idea between physicists.
Physics, in my opinion, shouldn't be affected by subjective acts, so only the physical measure has a meaning.

--
lightarrow
« Last Edit: 08/05/2015 18:07:13 by lightarrow »
 

Offline chiralSPO

  • Global Moderator
  • Neilep Level Member
  • *****
  • Posts: 1879
  • Thanked: 145 times
    • View Profile
I agree with lightarrow here--I do not think (at least it hasn't been shown) that there is any requirement for a human or computer, or any other entity to "observe" a system for decoherence (wavefunction collapse) to occur. Only an interaction (often called measurement, but doesn't have to involve any type of measurement) is required.
 

Offline David Cooper

  • Neilep Level Member
  • ******
  • Posts: 1505
    • View Profile
When are measurements made without the data ever being processed by a computer or mind? Up until the point where they are processed, the wavefunction could continue to represent multiple possible realities without simplifying. If you then look at the results and that forces a simplification, how are you going to know if you caused that collapse or if it happened back at the time of the measurement? Indeed, it may be that a computer or brain isn't always adding sufficient complexity to cause the collapse, so it might maintain two possible states of that computer/brain which both believe there has been a collapse of the wavefunction when it has not actually occurred yet.
 

Offline PeteJ

  • First timers
  • *
  • Posts: 6
    • View Profile
Great. This is getting interesting.

David - I like your idea and haven't come across it before,. Are you suggesting that there is a law of conservation of complexity? .

Lightarrow - I get the picture and do not see it differently. But I don't think the question of whether only the measurement has meaning is a free decision for physicists. To say that only unobserved events have meaning and then they lose it when they are observed would be a bit odd, and not the usual meaning of 'meaning'.

The thing is, your description leaves a finite time between the event and the observation, and this what I'm exploring. But then...

chiralSPO - You say that an observation is unnecessary, that the event (the detection by the detector) is enough for decoherence. This is not my understanding of things. It would answer my question if it is true, but is it true? It doesn't seem to match with the mainstream ideas that I read about. 

Is it widely agreed that no sentient observation is necessary for decoherence? 
   

   

 

Offline chiralSPO

  • Global Moderator
  • Neilep Level Member
  • *****
  • Posts: 1879
  • Thanked: 145 times
    • View Profile
When are measurements made without the data ever being processed by a computer or mind? Up until the point where they are processed, the wavefunction could continue to represent multiple possible realities without simplifying. If you then look at the results and that forces a simplification, how are you going to know if you caused that collapse or if it happened back at the time of the measurement? Indeed, it may be that a computer or brain isn't always adding sufficient complexity to cause the collapse, so it might maintain two possible states of that computer/brain which both believe there has been a collapse of the wavefunction when it has not actually occurred yet.

I have to dig around to find it, but I'm pretty sure that there have been experiments that show, very explicitly, that a wavefunction has collapsed by looking at successive interactions of a particle--for the first few, it behaves as a wave, and then, after a sufficiently disruptive interaction, it behaves as a particle...

Also, recall that A) nuclear magnetic resonance spectroscopy relies on the half life of coherent quantum states of populations of particles; and B) one of the most difficult obstacles to practical quantum computing is spontaneous decoherence/collapse.
 

Offline PmbPhy

  • Neilep Level Member
  • ******
  • Posts: 2762
  • Thanked: 38 times
    • View Profile
Quote from: PeteJ
Hi folks. My first post here. I have a question, most probably a naïve one...
Welcome to the forum.

Quote from: PeteJ
In QM, whether we favour MW or CI as an interpretation, there is the idea that an observation causes an event, either a 'collapse' or a splitting of worlds.
Where did you get that notion from? When you make a measurement of a system which is in an eigenstate of the system. E.g. if the system is in an eigenstate of energy and you measure the energy then all you get is a measurement of the energy, period. Nothing happens to the system. If the system is not in an eigenstate of the observable you are measuring then when you measure that observable you'll get one of the eigenvalues of the observable. That's all. You didn't cause anything physical to happen. If you write down the expression for the system in an arbitrary quantum state and then measure an observable then the mathematical expression for the system will change to an eigenstate of the observable that you measured. But again, nothing physical happens to the system other than it now being in an eigenstate. But you can't observe the quantum state. It's simply a mathematical expression which represents the system. It's a misconception to think that something physical happens when the system collapses. That's merely mathematical lingo to describe what's going on with the mathematical description of the probabilities of measuring observables.

Quote from: PeteJ
Does the observation preceed the event, or is it simultaneous with the event?
Carefully study what I wrote above and you'll understand that this question is meaningless.
 

Offline alancalverd

  • Global Moderator
  • Neilep Level Member
  • *****
  • Posts: 4721
  • Thanked: 155 times
  • life is too short to drink instant coffee
    • View Profile
If you could obtain information about an event before it occurred, you could prevent or modify it, thus invalidating the information you used, ad infintum.....
 

Offline evan_au

  • Neilep Level Member
  • ******
  • Posts: 4126
  • Thanked: 247 times
    • View Profile
Quote from: ChiralSPO
the event (the detection by the detector) is enough for decoherence.
Quote from: PeteJ
This is not my understanding of things.
I understand that one of the barriers to quantum communication is tiny imperfections in the optical fiber, which result in the entangled photons (rarely) interacting with one of these imperfections.

Nobody is observing all the imperfections in the fiber, and nobody (and no test equipment) has seen that the photon has struck a particular imperfection, but the photon is scattered, and coherence is lost.

So decoherence can be triggered by interaction with the environment.
 

Offline PeteJ

  • First timers
  • *
  • Posts: 6
    • View Profile
Okay. Thanks. Things are becoming more clear. I'm confused as to how the subject changed to a question that isn't mine but no matter.

PmbPhy - You may be giving the answer to my question but I must admit I don't fully understand it. You seem to be saying that the measurement has no effect on anything except our knowledge of the system.
Is that it?   

Evan-au - says that decoherence can be triggered by the environment, but if decoherence is only about  our knowledge of the system then how can this be?   
 

Offline PmbPhy

  • Neilep Level Member
  • ******
  • Posts: 2762
  • Thanked: 38 times
    • View Profile
Quote from: PeteJ
PmbPhy - You may be giving the answer to my question but I must admit I don't fully understand it. You seem to be saying that the measurement has no effect on anything except our knowledge of the system.
Is that it?
Sort of. If you're confused then I'd have to say - Join the club! ;D

Here's what is referred to as the orthodox position: Suppose you have a system which is an particular quantum state which isn't an eigenstate of any observable. Before a measurement of, for example, the position of a particle, is made then The particle wasn't really anywhere.

As Pascual Jordan put it: “Observations not only disturb what has to be measured, they produce it….We compel [the electron]
to assume a definite position…. We ourselves produce the results of measurements.”

If a particle assumes a specific position in space then it might be said that the measurement process "caused" it to happen. If you wish to look at it that way then, in this particular example, cause and effect are simultaneous.

The quote above was taken from the article called Is the Moon There When Nobody is Looking? by N. David Mermin, Physics Today. It's available online at: http://maltoni.web.cern.ch/maltoni/PHY1222/mermin_moon.pdf

You might want to read it. It might help you to understand what you're seeking to learn.
 

Offline David Cooper

  • Neilep Level Member
  • ******
  • Posts: 1505
    • View Profile
David - I like your idea and haven't come across it before,. Are you suggesting that there is a law of conservation of complexity? .

I remind you that what I said is pure speculation, and I'm not suggesting anything about conserving complexity beyond the idea that there may be some limit to how much complexity a wavefunction can maintain before it collapses, but that detecting where that limit is reached may not be possible. If we take Schrödinger's cat as an example, it has not been ruled out that the wavefunction of the cat may be compatible with it being dead and alive. A collapse of the wavefunction could simplify it to eliminate one of those two possibilities, and it may be that the cat is sufficiently complex to force that collapse all by itself, long before anyone looks in the box to see if the cat's dead or alive. When a person looks in the box, again there's the question of whether that person can force the collapse of the wavefunction - it may be that (s)he can't, in which case the wavefunction of the person may have to remain compatible with two contradictory outcomes, one of which registers the cat as alive while the other registers it as dead. At some later point, maintaining the two possibilities may no longer be possible due to there being more complexity than the universe can handle, at which point one of the possibilities must be junked and the cat is then either dead or alive, no longer maintained in an uncertain state where it may yet become either, and at that point the wavefunction of the human will simplify to match (if it didn't simplify first, which is probably more likely), leaving us with a human which knows whether the cat is alive or dead rather than incorrectly thinking (s)he knows while actually maintaining both contradictory beliefs. What we certainly don't see is a human opening the box and finding a cat that (s)he determines is both alive and dead at the same time, but that does not guarantee that the wavefunction has collapsed at that point.

What we do see though are experiments where matter can be seen to be maintaining two contradictory states at the same time, though they aren't really contradictory as the matter hasn't actually been forced to simplify what it's doing. That really is like looking at a cat and seeing it as both dead and alive at the same time, only it hasn't yet been done with anything as complex as a cat. The particular experiment I have in mind involved an item made out of several hundred atoms which vibrated from side to side, but it did so by moving from left to right while at the same time moving from right to left, etc. That experiment showed the effect in something much more complex than a single particle and it opened up the possibility of doing the same thing with bigger, more complex items which are doing more complex things. Perhaps it can be extended to a cat, but perhaps not, and we may never know.
 

Offline chiralSPO

  • Global Moderator
  • Neilep Level Member
  • *****
  • Posts: 1879
  • Thanked: 145 times
    • View Profile
Schrödinger's cat is a terrible thought experiment, and I wish it were not so overused. Schrödinger originally proposed this experiment to expose how ridiculous QM logic is on a macroscopic scale.

Forget about whether or not the cat is complex enough to disrupt the wavefunction, whatever device measures whether or not the quantum event has happened and tells the poison to be released or not has already collapsed the wavefunction.

There are different interpretations of what the wavefunction actually represents, if it is a real thing, or just a useful model etc. Ultimately QM tells us what is knowable and what is unknowable (given our current understanding, but it has yet to fail us)--but this does not mean that we can actually measure anything that is theoretically knowable... Unfortunately the question of whether QM mechanisms are actually deterministic or probabilistic is fundamentally unknowable and cannot be probed.
 

Offline PeteJ

  • First timers
  • *
  • Posts: 6
    • View Profile
I wonder why the thread question has been changed. The OP said nothing about the event preceeding the cause. Not sure I'm happy about this. 

Quote from: PeteJ
PmbPhy - You may be giving the answer to my question but I must admit I don't fully understand it. You seem to be saying that the measurement has no effect on anything except our knowledge of the system.
Is that it?
Sort of. If you're confused then I'd have to say - Join the club! ;D

Here's what is referred to as the orthodox position: Suppose you have a system which is an particular quantum state which isn't an eigenstate of any observable. Before a measurement of, for example, the position of a particle, is made then The particle wasn't really anywhere.

As Pascual Jordan put it: “Observations not only disturb what has to be measured, they produce it….We compel [the electron]
to assume a definite position…. We ourselves produce the results of measurements.”

If a particle assumes a specific position in space then it might be said that the measurement process "caused" it to happen. If you wish to look at it that way then, in this particular example, cause and effect are simultaneous.

The quote above was taken from the article called Is the Moon There When Nobody is Looking? by N. David Mermin, Physics Today. It's available online at: newbielink:http://maltoni.web.cern.ch/maltoni/PHY1222/mermin_moon.pdf [nonactive]

You might want to read it. It might help you to understand what you're seeking to learn.
Your summary here is just fine and very helpful. I'm not really trying to learn QM but, rather, what physicists think about QM. I think you've answered my question. If it is allowable that the measured particle is not really anywhere until the measurement and is 'reified' by the measurement then the problem I was investigating evaporates. I had thought that physics rejected this idea since it would lend credence to idealism of some sort and also mad ideas like David's  ;D

... Which I suspect has some truth in it. It is not quite the same idea, but it seems to make some sense that stable complexity requires a material substrate. I.e that mind requires matter. It's a pet theory. 

No need to go there though. Many thanks for all the excellent replies. On one science forum I know this discussion would have descended into a raging argument within five posts.     
 

 

Offline PmbPhy

  • Neilep Level Member
  • ******
  • Posts: 2762
  • Thanked: 38 times
    • View Profile
Quote from: PeteJ
Your summary here is just fine and very helpful.
Wonderful. I'm so glad to hear that you've got your question answered to your satisfaction.  :D

Quote from: PeteJ
On one science forum I know this discussion would have descended into a raging argument within five posts.     
That's why when I opened my own company I added physics forums to my companies website. Membership is by invitation only. I wait until I get to know a member in one of the forums I frequent and if they have what it takes to be in my forum I send them an invitation. It's simple. I do my best to try to predict whether the person will be able to respect and follow the forum rules. If the person objects to the rules then they're not invited. If they're invited and they join then we can discuss whether any changes should be made to the forum rules. I'm very good about that and am always looking for input from members. Problems sometimes arise when a member doesn't like a rule and then quits because of it. The problem with that is that it was their responsibility to discuss their objections to the rules until we're both satisfied. It's an ongoing process to, not a one time thing. So if I enforce a rule and the member convinces me and other members of the forum that the rule was wrong or didn't apply then I undo the enforcement. The rules are what one typically finds in all forums except that I strictly enforce them.
 

Offline Atomic-S

  • Hero Member
  • *****
  • Posts: 935
  • Thanked: 19 times
    • View Profile
Quote
As Pascual Jordan put it: “Observations not only disturb what has to be measured, they produce it….We compel [the electron]
to assume a definite position…. We ourselves produce the results of measurements.”
.
I find this an interesting concept, which brings up the following question regarding the double-slit experiment. As it is typically conducted, a particle is sent through two parallel slits, behaving wave-like, during which, I suppose, it has no position.  Then it hits a screen as, for example, photographic film, is detected at a specific position, and is deemed to be a particle having that position.  I have wondered if there is any way of detecting it that does not demand that it assume a definite position, and if so, that would be a significant experimental breakthrough showing that a particle indeed not need be in a definite position to be a particle.  It would appear that the nature of the detector is significant in this regard. When photographic film is placed there, it becomes part of the experiment. What if we used something else?  We would have to be clever about what it was in order to elicit the desired result. We want something that would indicate the presence of the particle, but in a way that did not demand that it assume a definite position.  If we were able to do that, it suggests the possibility that the particle could continue to travel beyond there still following much of its original wave configuration, that would become evident in a second detection of some kind. With the conventional double-slit experiment, particles hit the detector with a distribition governed by their wave nature up to that point, but whemn each hits, it is recorded in a specific place, and as such, now has a state function that is different, so that (assuming the particle has enough energy to continue on from that point), its subsequent behavior should be different than its original wave distribution indicated. The point is t hat the first detection, by requiring it to assume definite position, has altered its subsequent behavior.  So the objective would be to show that if we could detect its presence in a way that did not require it to assume a definite position, then its subsequent behavior would be much less disrupted and would largely be the same as its earlier wave function would indicate.  Can such an experiment be done?  The first thing that comes to mind as a possible way of doing so is to change the chemistry of the detector so that it is characterized by large molecules having electron orbits that run throughout their entire length and thus do not have narrowly localized locations. You would also want to try to set this up so that the molecules had a dearth of closely-confined electron orbits that would be sensitive to the particular particle to be detected.  The result would be that when the particle interacted, an electron from a wide-ranging orbit would be kicked up into another higher-energy wide-ranging orbit, which could triger other phenomena that would be recorded.  But because the process did not pin the detected particle to a particular location, its wave function would not have "collapsed" as much as would ordinarily be the case, allowing the particle, with such residual energy as it had, to proceed further on and be detected a second time by a conventional detector. The result of that would be, executed on many particles, a diffraction pattern more closely resembling what would have come out of the slits (probably some weak fringes), rather than what would have come out of a narrow-position detector (probably diffuse). 
 

Offline PmbPhy

  • Neilep Level Member
  • ******
  • Posts: 2762
  • Thanked: 38 times
    • View Profile
Quote from: Atomic-S
I find this an interesting concept, ...
And it might be interesting to read if you didn't write it as one long sentence. It's easier to read when you break things up into paragraphs. If you choose to do so I'd like to read what you wrote. Otherwise it's a strain on my eyes.
 

Offline Atomic-S

  • Hero Member
  • *****
  • Posts: 935
  • Thanked: 19 times
    • View Profile
To make a long story short, we do the double slit experiment as follows: There are the slits, and there is a conventional detection screen. It produces an indication of where each particle hits. Between the slits and the screen we place another detector.  It is to detect the passage of the particle without detecting its position.  When the particle passes through it, its presence is recorded but its position is not. Because its position was not, it remains indefinite.  The purpose is to test to see if, by leaving it indefinite, the wave characteristics that emerged from the slits remain in force.  We find the answer to that by allowing the particle to strike the final detection screen and record a position. We repeat the experiment with many particles. If the statistical pattern on the final detection screen is consistent with the diffraction expected from the slits, then we would know that the wavelike behavior was not nullified by the earlier detection as a particle.  That would be interesting because this experiment would have demonstrated an object possessing both particle-like and wave-like properties at the same time.

The detector to be first encountered might be constructed of a semiconductor having two conductive energy bands separated by a small energy difference.  Because such energy bands are characterized by electrons free to wander extensively,  their electrons are of indefinite position. If the energy difference is substantially less that that of the incident photon (for example), the amount of energy lost from the photon when it kicked an electron from the lower band to the upper band would not greatly affect it, so that as it proceeded from that point on toward the final screen, the photon's wavelike character would undergo only limited change. This change would probably appear as an increase in wavelength. The result of that could be that the diffraction pattern would widen a little.  But the experiment should still be doable.

 

Offline PmbPhy

  • Neilep Level Member
  • ******
  • Posts: 2762
  • Thanked: 38 times
    • View Profile
Quote from: Atomic-S
To make a long story short, we do the double slit experiment as follows: There are the slits, and there is a conventional detection screen. It produces an indication of where each particle hits.
So far so good.

Quote from: Atomic-S
Between the slits and the screen we place another detector.  It is to detect the passage of the particle without detecting its position.
What you just said is impossible. One can't merely make such statements and expect them to be taken as a valid statement. You have to state exactly how such a thing is possible by stating at least one way on how to do it. And detecting a passage of something means that you've taken a position measurement. You don't appear to understand that those two things are identically the same thing.

Quote from: Atomic-S
  When the particle passes through it, its presence is recorded but its position is not.
Again, impossible. To demonstrate what I mean please draw a diagram of what you're saying and the actual physics of how it would be accomplished.

Things like this are the main problems with ideas in this forum. The ideas presented are so vague as to appear possible but so vacuous that it can't be accomplished in practice.
 

Offline Atomic-S

  • Hero Member
  • *****
  • Posts: 935
  • Thanked: 19 times
    • View Profile
The position of the particle has three components.  The component in the direction from the source to the final target is indeed recorded to the precision of the position of the central detector, but the position in the other two dimensions is not.  It is those transverse measurements that matter in this experiment.
 

The Naked Scientists Forum


 

SMF 2.0.10 | SMF © 2015, Simple Machines
SMFAds for Free Forums
 
Login
Login with username, password and session length