What Is Surprising About Wave Function Collapse?

[abrogard]

Excuse my ignorance. I've been googling trying to understand but they always seem to just state it without explaining why.

They say the electron could be anywhere within this area of probability but we don't know why until we look and then the wave function 'collapses' and we know where it is.

And that's a big surprise. A big 'strangeness'. Evidence of the 'weirdness' of the quantum world.

Why?

Because until we looked it wasn't there?

Or because until then we didn't know where it was?

I see nothing surprising about either of those two. If it's moving it is not there until it is there.

And of course we don't know where it is until we look.

You don't know where I am until you look.

I've obviously got a completely naive understanding of what they're trying to say. Can someone please shatter my naivete?

 

[atyy]

In quantum mechanics, there are two rules for time evolution. One is deterministic and goverend by Schroedinger's equation, and the other is random and is called collapse. They are not contradictory because they apply at different times. However, who decides which rule is applied when? Apparently, an external observer is needed to decide when a measurement outcome is seen, and the random collapse occurs. This is fine in practice, since we always know when we get a measurement result. But it is unsatisfactory that our theory of physics doesn't seem to make sense when applied to the whole universe, since there is no external observer of the universe. This is the problem of collapse or the "measurement problem".

 

[bhobba]

All this is interpretation dependent. The formalism, that is the math without any interpretation or a very minimal one, says QM is a theory about observations that appear here in an assumed common-sense classical world. What's going on when not observed the theory is silent about. Its that silent bit that to a large extent interpretations want to elucidate.

The other thing about the theory is only probabilities of the results of observations are predicted. And those probabilities are a generalisation of ordinary probability theory:
http://www.scottaaronson.com/democritus/lec9.html

Thanks
Bill

 

[Nugatory]

And of course we don't know where it is until we look.
You don't know where I am until you look.

You are describing it as if the the electron is always somewhere and it's just that we don't know where until we look. If that were what was going on, it wouldn't be at all surprising at all - we all know people who have this problem with their car keys every day.

But orthodox quantum mechanics says something different. It says that the until the electron position is measured, it has no position... not "it has a position but we don't know what it is", but rather that there is no position unless we measure, the same way that I don't have a lap unless I'm sitting down. Have you looked at the double slit experiment for electrons? If not, google will find you plenty of pointers.

 

[abrogard]

I am still considering, with the help of some googling atyy's response.

bhobby also has me thinking and studying - his link seems to indicate clearly quantum mechanics belongs to another realm, not observable reality, not maths, but a third place.

I certainly had not thought orthodox quantum mechanics said the electron has no position and cannot currently see how that can be. It describes an area of probabilities, does it not? Well surely a thing must 'be' in the first instance to have a probability of 'being' here or there?

Yes I've seen the double slit experiment numerous times and listened to and read numerous descriptions/explanations of it. And I think I've seen reputable physicists dogmatically state 'it is a particle' and 'it is a wave' - i.e. the two of them contradicting each other.

It all leaves me not knowing. It doesnt' solve anything for me. I get the impression no physicist expects it to, I frequently get the impression they show us this experiment in order to induce a state of bewilderment. In fact they virtually say as much, don't they?

I do have a query arising from that, though. Why isn't the particle wave question easily decided on the basis of waves propagating out from source in spherical fashion, whereas particles propagate out like bullets?

But I don't wish to complicate this thread. I've asked a simple question and am happy to stay with that until I've arrived at some 'answer' that will satisfy me.

 

[Derek Potter]

I've asked a simple question and am happy to stay with that until I've arrived at some 'answer' that will satisfy me.

Hmm, you might be here a long time. :)
I agree though. The wierdness of quantum mechanics is not very obvious if you are just measuring the position of a single particle.

 

[bhobba]

It all leaves me not knowing. It doesnt' solve anything for me. I get the impression no physicist expects it to, I frequently get the impression they show us this experiment in order to induce a state of bewilderment. In fact they virtually say as much, don't they?

We have met the enemy and he is us - Pogo. The quantum world is not part of everyday experience. It requires a different mindset.

I do have a query arising from that, though. Why isn't the particle wave question easily decided on the basis of waves propagating out from source in spherical fashion, whereas particles propagate out like bullets?

This wave particle stuff, while common in popularisations and beginner texts, isn't really correct and is not part of more advanced treatments:
http://arxiv.org/pdf/quant-ph/0609163v2.pdf

Here is a treatment of the double slit from the more advanced perspective:
http://arxiv.org/ftp/quant-ph/papers/0703/0703126.pdf

But I don't wish to complicate this thread. I've asked a simple question and am happy to stay with that until I've arrived at some 'answer' that will satisfy me.

In my experience until you let go of ingrained classical beliefs that may prove elusive.

Thanks
Bill

 

[abrogard]

http://arxiv.org/pdf/quant-ph/0609163v2.pdf[/URL]
Bill

I have just opened the above link and read the intro. It looks very promising to me indeed, looks like something targeted at such as myself. Excellent.

Causes me to ask one little question if I may before I even read it though.

In the light of all that first para - the 'abstract' - says what is it exactly that IS generally accepted and proven about quantum mechanics that causes numerous populisers of the science to claim it is 'The most successful theory ever devised.' ?

Tried and tested to numerous orders of precision. Making predictions to the same astounding accuracy.

What is that then? Some set of maths? That does what exactly? Predicts particle decay products? Just what? I am totally in the dark as to what it might be. That was my best guess just then. There must be a big difference between it and all the things mentioned in that introductory abstract, obviously, if they are so much debated, contested, disputed.

Do I make my question clear? I'm bad at that. I'm just asking what is this great strength and indisputable truth and fact and science that is employed every day and works so well if it is not all those things - those things being the very things that are generally presented to the general public such as myself continually as being, in fact, 'quantum science' or 'quantum mechanics' itself.

No wonder we're so confused if we're told on the one hand this thing is IT and on the other hand that same thing is widely disputed.

:)

p.s. Edit. Perhaps I should have waited a little. Maybe the first para of the actual intro is the answer I'm looking for:

quote:
On the technical level, quantum mechanics (QM) is a set of mathematically formulated
prescriptions that serve for calculations of probabilities of different measurement outcomes.
The calculated probabilities agree with experiments.
unquote:

Is that it? 'mathematically formulated prescriptions for calculations of probabilities' ?

 

[bhobba]

What is that then? Some set of maths? That does what exactly? Predicts particle decay products? Just what?

Everything at the level of the very small is explained by QM - everything without fail. That includes how transistors work, the very strange behaviour of liquid helium, otherwise inexplicable phenomena like black body radiation, even why solidity itself exists - the list goes on. It has done everything that has been asked of it. Even things like its unification with general relativity you may have read where it failed has recently been shown to not be the issue it was once thought:
http://arxiv.org/abs/1209.3511

The accuracy thing is most likely referring to QED which has been tested to breathtaking accuracy:
http://scienceblogs.com/principles/2011/05/05/the-most-precisely-tested-theo/

Thanks
Bill

 

[atyy]

I am still considering, with the help of some googling atyy's response.

bhobby also has me thinking and studying - his link seems to indicate clearly quantum mechanics belongs to another realm, not observable reality, not maths, but a third place.

I certainly had not thought orthodox quantum mechanics said the electron has no position and cannot currently see how that can be. It describes an area of probabilities, does it not? Well surely a thing must 'be' in the first instance to have a probability of 'being' here or there?

Yes I've seen the double slit experiment numerous times and listened to and read numerous descriptions/explanations of it. And I think I've seen reputable physicists dogmatically state 'it is a particle' and 'it is a wave' - i.e. the two of them contradicting each other.

It all leaves me not knowing. It doesnt' solve anything for me. I get the impression no physicist expects it to, I frequently get the impression they show us this experiment in order to induce a state of bewilderment. In fact they virtually say as much, don't they?

I do have a query arising from that, though. Why isn't the particle wave question easily decided on the basis of waves propagating out from source in spherical fashion, whereas particles propagate out like bullets?

But I don't wish to complicate this thread. I've asked a simple question and am happy to stay with that until I've arrived at some 'answer' that will satisfy me.

The wave particle duality is not a problem. All physicists agree quantum mechanics is a perfectly coherent theory.

The disagreement is: is quantum mechanics potentially complete, or does the fact that we have difficulties using only the deterministic evolution of the wave function to describe the whole universe indicate that quantum mechanics is incomplete?

In general, we don't expect our theories to be complete, and we expect that more experiments will show our current best theories to be wrong in some way. But in many cases, experiment is essential for showing our theories to be incomplete. The question in quantum mechanics is: is there an indication, even before any experiments prove quantum mechanics to be wrong, that quantum mechanics itself indicates that it is incomplete?

 

[vanhees71]

Just to answer the question in the title of this thread: The only surprising thing about collapse is that it still is thought as being necessary to use quantum theory as a physicist. Everything else is subject to "interpretation" beyond the "minimal interpretation", and that's just a matter of taste, what you like to believe, but it has little to do with physics (but this is also, of course, a personal opinion) ;-)).

 

[atyy]

Just to answer the question in the title of this thread: The only surprising thing about collapse is that it still is thought as being necessary to use quantum theory as a physicist. Everything else is subject to "interpretation" beyond the "minimal interpretation", and that's just a matter of taste, what you like to believe, but it has little to do with physics (but this is also, of course, a personal opinion) ;-)).

I should stress the vanhees71 uses "collapse" to mean "physical collapse", whatever that means. The standard usage of the term collapse is not what vanhees71 is talking about. In the standard usage, the wave function and collapse are not necessarily physical, so "physical collapse" is not defined.

Collapse in the standard usage is a standard part of quantum mechanics. The idea of physical collapse usually refers to approaches like GRW or CSL, which are not standard quantum mechanics and predict deviations from standard quantum mechanics.

 

[lightarrow]

My personal intuition about collapse under measurement is that QM' standard interpretation claims to separate the system from the measurement apparatus but this is just a first approximation. If an atom's electron is used to measure a photon (through photo excitation) is really the atom which measures the photon, or is the opposite, or both things? What makes a measurement apparatus a macroscopic, instead of a microscopic one? I believe something is still missing here about what phisically is a measurement. I'm not learned in all the concepts as decoherence theory, etc, but we have to explore the problem further, in my opinion.

--
lightarrow

 

[vanhees71]

If collapse is not meant to be physical, you don't need to introduce it in the first place. At least I have no use for it whatsoever. Then it's simply the update of knowledge of an observer due to a measurement.

 

[Demystifier]

Is that it? 'mathematically formulated prescriptions for calculations of probabilities' ?

Yes, that's what QM is about, if we concentrate only on non-controversial aspects with an undeniable success.

 

[atyy]

If collapse is not meant to be physical, you don't need to introduce it in the first place. At least I have no use for it whatsoever. Then it's simply the update of knowledge of an observer due to a measurement.

But you do need to update the knowledge of an observer in order to calculate a conditional probability. So you do need collapse.

In the standard interpretation, collapse is updating the knowledge of an observer. The standard interpretation is agnostic as to whether it is physical or not. This is the position of Cohen-Tannoudji, Diu and Laloe's textbook.

 

[vanhees71]

Sigh. Why do you want to call it collapse instead of just stating that you calculate a conditional probability? The word "collapse" is loaden with so much metaphysical balast (where Bohr and particularly Heisenberg are the main culprits ;-)) that I try to avoid it whenever I can!

 

[atyy]

Sigh. Why do you want to call it collapse instead of just stating that you calculate a conditional probability? The word "collapse" is loaden with so much metaphysical balast (where Bohr and particularly Heisenberg are the main culprits ;-)) that I try to avoid it whenever I can!

Then it's just terminology, which is pointless to argue about. The important point here is that an ingredient beyond the Born rule and Kolmogorov's axioms is needed, and it is good to have a handy name like collapse or if one is more formal one can use "state reduction".

 

[vanhees71]

I precisely deny that there's "something" necessary beyond Born's rule and Kolmogorov's axioms to make sense of quantum theory as a physical model of the real world. Which "something" should that be, however you call it.

 

[atyy]

I precisely deny that there's "something" necessary beyond Born's rule and Kolmogorov's axioms to make sense of quantum theory as a physical model of the real world. Which "something" should that be, however you call it.

That's great! I'm glad we haven't been having a pointless terminology discussion. So can collapse be derived from the Born rule and Kolmogorov's axioms?

I have an argument that it cannot. In the most general form of collapse, the rule of collapse is not unique, even for a given observable. It must be calibrated according to the measuring apparatus. So one at least needs a postulate beyond the Born rule and Kolmogorov's axioms to derive the rule of collapse.

 

[andresB]

OP,

Look at (just for example)

http://arxiv.org/pdf/quant-ph/9610033.pdf

and you will see that the collapse is not so simple.

 

[bhobba]

and you will see that the collapse is not so simple.

If collapse is simple or not, or even exists, is very interpretation dependent. One can not make generalisations.

Thanks
Bill

 

[lightarrow]

If collapse is not meant to be physical, you don't need to introduce it in the first place. At least I have no use for it whatsoever. Then it's simply the update of knowledge of an observer due to a measurement.

I need a clarification here: a "measurement" is intended only as a physical act or not? This "update of knowledge" happens even in other cases or just in the case of a measurement?
Thanks.

--
lightarrow

 

[naima]

QM founders highlighted that there is no QM without classical Mechanics. Classical Mechanics is about a word where we neglect microscopic details. You have heat pressure, mean values and probabilities. The is no Schrödinger cat in this word. But you have to use CM to describe the apparatus in a laboratory, its environment and so on.
So we need a frontier. The problem is not to find where this frontier is: You put it where you want! It may include an observer who looks at the apparatus.
This frontier has to be seen as a boudary in space time. You can choose it to wrap only the particle between two moments or the whole laboratory between 2014 and 2015.
Once you have this frontier QM tells you that this boudary is a black box. Not a black hole but not so far. Inside the box you have amplitudes of probabilities that you have to sum Outside you have probabilities. Inside you have Schrödinger equation outside you have
https://en.wikipedia.org/wiki/Diffusion_equation
The is no collapse in QM. Seen from the outside QM has given a probability to the boundary that YOU have chosen. I think that the question of when did the collapse occurred has no sense.
Collapse is an interpretation of QM for observers who live in a classical word where there is no amplitudes to collapse.

 

[stevendaryl]

If collapse is not meant to be physical, you don't need to introduce it in the first place. At least I have no use for it whatsoever. Then it's simply the update of knowledge of an observer due to a measurement.

I think that this updating is the same thing as collapse. Of course, if the wave function is viewed as expressing our subjective knowledge about the true state of the system, then such a collapse is unremarkable, but I don't see how that interpretation is viable, given Bell's theorem.

 

[vanhees71]

It's not subjective precisely because of Bell's theorem. Why should it be? For me quantum theory is a mathematical concept to predict probabilities for the outcome of measurements given a state defined by a preparation procedure of the system under study. Thus doing a measurement is always a random experiment in the sense of probability theory. Whether there is some other theory that is as successful as QT in describing matter on a microscopic level and at the same time deterministic, we don't know. At least, so far there is none known. To realize the value of a measured observable in a random experiment is just this recognition of this outcome, but nothing on the system's state collapses. It's not even clear to the collapse proponents, when this process might happen. Is it enough that a measurement apparatus stores the result somehow or does one need a "conscious" being to "collapse the state"? If so, what's "consciousness"? To ask like Bell: "Is an amoeba enough to collapse the state?" To me collapse is totally superfluous, leading to severe conceptional problems rather than contributing to our understanding of Nature.

 

[stevendaryl]

In the "consistent histories" interpretation of QM, instead of thinking of the wave function as giving probabilities for what's true NOW, you can instead think of the wave function, together with its evolution equations as giving a probability distribution for complete histories of the universe. With that interpretation, the "updating" following a measurement is exactly the updating that happens in classical probability when you acquire new knowledge.

The down side of this way of looking at it, it seems to me, is the following two thorny issues:

1. To interpret QM as a probabilistic theory on histories, you have to make a choice as to which set of distinguishable histories you're choosing from. This seems subjective, although maybe decoherence helps to give you a preferred set.
2. Viewing QM this way is actually equivalent to a hidden-variables theory, in which the "hidden variable" is which possible history is the "real" one. But as a hidden-variables theory, it's nonlocal.

 

[stevendaryl]

It's not subjective precisely because of Bell's theorem. Why should it be?

If it's not subjective, then it's physical, so the updating is a change in a physical quantity.

For me quantum theory is a mathematical concept to predict probabilities for the outcome of measurements given a state defined by a preparation procedure of the system under study.

If a measurement is a physical interaction like any other, then what makes a measurement result have a definite outcome, when something like "the z-component of the spin of an electron" need not have a definite outcome (that is, an electron can be in a state that has no definite value for that quantity)?

 

[vanhees71]

A measurement has a definite outcome because you constructed your measurement apparatus to give you one.

More complicated is the question whether the probabilities of QT are "objective" or "subjective". First of all one has to define the meaning of these words. For me the statement that probabilities are "subjective" is describing a situation, where you use probabilistic descriptions, because you have incomplete knowledge, as in classical statistical mechanics. You don't know the location and momentum of any gas molecule in a container of an equilibrated gas but describe it by some "coarse-grained observables" giving you the relevant description as a macroscopic system like pressure, density, internal energy, temperature, and so on. According to the classical theory in principle each molecule has a determined position and momentum (and perhaps also some orientation if not monatomic). Within classical physics all probabilities are subjective.

In quantum theory, however you have a different kind of probabilities, because within quantum theory you cannot prepare a system to have determined values for all observables, and thus a pure state, representing the complete possible knowledge about the system, leads only to probabilistic statements. Even for one particle, it's not possible to determine its position and momentum precisely at the same time (Heisenberg uncertainty relation), but you describe its position's and momentum's (and any other observable's) probability distribution (or probabilities for discrete quantities like spin components etc.) by it. The state, however, is uniquely defined by (an equavalence class) of preparation procedures, and you cannot gain more information about the system (within the realm of quantum theory). Bell's theorem tells us that, if there is a deterministic theory, where these irreducible quantum probabilities are "subjective" in the above stated sense, it must be a (I suppose very complicated) non-local deterministic theory, and so far we don't have such a theory nor do we have any glimpse that such a theory might exist or is consistently stateable at all.

 

[atyy]

To realize the value of a measured observable in a random experiment is just this recognition of this outcome, but nothing on the system's state collapses. It's not even clear to the collapse proponents, when this process might happen. Is it enough that a measurement apparatus stores the result somehow or does one need a "conscious" being to "collapse the state"? If so, what's "consciousness"? To ask like Bell: "Is an amoeba enough to collapse the state?" To me collapse is totally superfluous, leading to severe conceptional problems rather than contributing to our understanding of Nature.

Does removing collapse solve this problem? You still need an observer to say when an observation occurs (ie. when do you apply the Born rule), or you need something extra beyond standard quantum theory.