Why is quantum mechanics weird? The bomb experiment
Vložit
- čas přidán 19. 06. 2024
- Check out the physics courses that I mentioned (many of which are free!) and support this channel by going to brilliant.org/Sabine/ where you can create your Brilliant account. The first 200 will get 20% off the annual premium subscription.
I have done quite a few videos to demystify quantum mechanics. In this video I want to explain just why quantum mechanics is weird. Is it entanglement or superpositions? Schrödinger's cats? Not quite, really. But the famous Elitzur-Vaidman Bomb Experiment demonstrates well just why quantum mechanics is weird indeed.
You can support us on patreon / sabine
0:00 Intro
0:55 Psi
2:09 Dead-and-Alive cats
3:43 Entanglement
5:20 The Bomb Experiment
9:36 Sponsor Message
#physics #education #science - Věda a technologie
![Eminem - Houdini [Official Music Video]](http://i.ytimg.com/vi/22tVWwmTie8/mqdefault.jpg)
Hi everyone. As several people have pointed out, the audio doesn't quite sound right. I can't fix this -- I can only take the video down and replace it in the next couple of days. As it seems to at least be clearly understandable I'll let it up. Will make sure that next week we're back to the normal quality. Sorry about that.
It's weird that I didn't notice it until someone pointed out.
@@nomizomichani Ditto
@@nomizomichani Well, neither did I...
I didn't notice a problem at all, even at a second hearing. So the audio track is in a superposition of weird/not weird. But there is nothing spooky about it.
@@SabineHossenfelder Dont worry about the sound Sabine.. When will you physicists finally get real and admit that we're all living inside a super-advanced, hyper-realistic holodeck complex super-structure.. That would explain emergence, fine-tuning and the holographic nature of our reality. The organic big bang theory doesn't explain any of that 😲😲😲
I had a quantum mechanic once. He was definitely weird. And he was never sure whether he’d actually fixed my car or not.
The car was fixed and not fixed at the same time. But you could never tell where it was and where it was going at the same time.
@@paryanindoeur which wouldn't be so bad if he didn't leave his dead cat inbtge back seat.
That is a joke. But I don't know if a laugh or not.
@@fadiluca , lol
I am both laughing and not laughing but someone is both watching me and not watching me so I am doing neither and both.
I think its pretty amazing that quantum mechanics is both weird and not weird at the same time.
I believe interpretations are that which is weird. In quantum physics particles are NOT in multiple places at once, we just cannot say for certain where a particle is until measured.
Well yes and no. That's the whole idea of the superposition. The particle is located at all of the possible places at the same time before the observation is made. In short - all that is possible, IS possible. It's just about how probable it is. If you were to repeat the measurement over and over again, you would get a probability chart.
Without the observation the particle isn't at any single position, instead the particle is the tautology of "it being" from the sum of all the probabilities (where it could be when observed).
Bomb experiment seems to indicate that we can find out information that COULD be possible without actually it happening, which is really neat.
@@Fex. The idea of it being at all possible places at the same time before the measurement is made..is cool but ultimately bollocks.No different to placing a ping pong ball in a box and shaking it.Would you say the ping pong ball is in superposition?
@@philip851 by your hypothesis, interference pattern should not exist. A single electron can exist at multiple places at the same time and interfere with itself and produce interference pattern.
Well, only until you observe it
I'm happy that my teacher of quantum physics was Paweł Horodecki, he showed us in 2008 at one of the first lessons this insane topic of Elitzur and Vaidman bomb experiment. I still treat my notes from it like some sacral artifact. I talked to all my friends about it even when they had nothing to do with physics .Great memories
Find someone how whats to talk Physics all the TIME ;)
well, you are not wrong about treating this like that. After all this is the precursor for magical tech
The bomb experiment may not actually be very weird at all. It may just come from the fact we arbitrarily choose to treat photons differently than other states.
As shown with the second beam splitter, beam splitters are sort of like a logic gate with _two_ inputs and _two_ outputs. It is basically equivalent to the Givens logic gate with the angle π/4. If both inputs are 0, it outputs 00, which is the analogue to no light on the beam splitter, then no light comes out. If both inputs are 1, it outputs 11, which is the analogue to light on both angles of the beam splitter just producing light on both angles of the output. If only one of the inputs is 1, meaning you only shine light at it from one angle, then the output is 50% chance 01 or 50% chance 10, meaning it has a 50% chance to redirect the photon to either path.
Since it is a logic gate and all quantum, logic gates are unitary, applying it twice cancels itself out, so if your input is 10 and you apply it twice, your put is 10. If you apply a phase shifter on one of the paths, basically the Pauli-Z logic gate, then it ends up flipping the final path the photon takes, so if you pass in 10 into the first beam splitter, apply a phase shift on one of the paths, then 01 will come out the second beam splitter. This also replicates what happens if you make a measurement, the state undergoes decoherence after the first beam splitter and becomes either exactly 01 or 10, and thus when it hits the second beam splitter it will have a 50% chance of leaving as a 01 or 10.
If you implemented this circuit into anything _else_ besides photons, all the mystery immediately disappears. For example, if you assign 1 to an electron with spin up and 0 to an electron with spin down, two tangible electrons both take the two paths. There isn't much of a mystery here because both electrons are tangible objects that carry a bit of information as well as some phase related information, so when they recombine on the other end they can interfere based on that information, and making a measurement to the tangible electron causes decoherence. This can be explained in entirely classical terms, you don't even need to resort to quantum mechanics.
The potential fallacious reasoning arises from the fact that we treat photons differently from other quantum states. We assume that a photon in the 0 state does not actually exist and thus cannot carry any information at all. But if we don't treat it differently, if we treat it like any other state, then a photon in the 0 state could indeed carry information and propagate through the system. It would show up on any detector as a 0 because it only carries phase-related information, but it may or may not interact with a detector (depending on whether or not the bomb is a dud), may or may not causing decoherence, and then when it recombines with the other photon, it could alter how they recombine. That would mean it is not an "interaction free measurement," it is an interaction with a photon in the 0 state.
Just food for thought.
I started watching quantum physics a few years ago because I thought it was weird...no what if it's not weird it's no fun
Guess I will go back to the why files and heckelfish to get my daily weird...😊😊
By the way, I really appreciate your putting the relevant papers up on the screen. It helps a lot.
Quantum mechanics: how I learned to stop worrying and love the bomb.
"how I learned to stop worrying and love the bomb." Finally. Something even I can understand! Thankyou Kimmo.
Haha, good one :)
Lol
ABDULPls
Loved the movie
I love you show! Binge watching after watching the last video you posted, You are a true scientist!
Excellent open question posed in title. Good job!
"Quantum mechanics isn't weird."
"Oh, that's nice."
"However, quantum mechanics is terribly weird."
"OH NOOOOOOO"
It is weird and not weird or in an undefined state.
It's a superposition.
Hah - Schrodinger's weirdness!
@@Sturzfaktor2 They call the undefined state the superposition? That's like weird.
Let’s ask Douglas Adams about this; he’s still around in a non-Marvel multiverse. 😇
I thought the main weirdness of entanglement was the idea of no hidden variables. That is, it’s not that the two particles have correlated values the whole time that are simply measured at some point for particle A, thus implying what B’s value was all along, but rather that A and B do not in any way have those values, or have them in a non-privileged way along with every other option in superposition, and only when you measure A does B in fact take on the correlated value at that instant, where prior to the measurement it could not have been said to have that value at least not exclusively.
You are absolutely right. Entanglement is incredibly weird, I am always disappointed when people try to "demystify" it. The fact that we have a mathematical model to describe it does not mean that things that it describes are not completely in contrast to our normal understanding and preception of the world.
@@TelekinesisT yeah.. its so misleading because, i can understand that the math is simple enough, but that doesnt mean..i dunno
To be precise, the violation of Bell inequalities breaks local realism, you can keep locality if you throw realism down the drain, that is the idea that it is possible to stitch observers experiences into one global description of the universe.
If all systems are quantum systems, Alice/Bob can't tell that Bob/Alice has chosen a measurement basis before they communicate their results/measure each other, and that can only be done in a way that preserve locality.
That's the idea behind Relational quantum mechanics : en.wikipedia.org/wiki/Relational_quantum_mechanics.
Yeah, the weird thing about entanglement is that if one particle has the value X, and the other -X, you can in fact change the value of one to Y, and when you measure the other it will be -Y... You do not really know what is X or Y either, but you can show with an experiment they are not the same. It is weird.
@@NetAndyCz correct me if I'm wrong but, entangled "particles" must always originate locally. Then, the entangled particles are separated by an arbitrary distance and remain "connected" in so far as their properties must compliment each other. Why is this weird? If particles are really waves in a quantum field(s), then the entangled particles are really just complimentary parts of a wave which remains consistent at any distance. The peak of the wave (particle A) remains a peak, and the trough (particle B) remains a trough, always. If we measure A as a peak, of course B will be the trough, and vice versa. Why is this weird?
your videos take me to the underground of thinking about scale, spectrum, perspective, reference. thanks for that.
7:32 - is this correct, though? There's a 50% chance it gets blocked by the dud, and in the rest of the cases it either goes into detector A or detector B, with 25% chance each. Exactly the same as if the bomb were live.
Unless the wave function somehow magically passes through the dud and recombines with the upper ray to cancel out...?
I looked up the experiment on Wikipedia. The dud does not absorb any light. Protons just pass through.
My thoughts exactly!
You can’t have your cake and eat it…?
The triggers on the dud bombs have no photon sensor, so any light incident on the bomb will not be absorbed and will instead pass straight through.
Sabine: QM is not weird, just unintuitive.
Me: Ok...
Sabine: But it is weird!
Me: Oh, the plot twist!
And then the villains become the cops and the cops become corrupt. Not surprised to be surprised, it is the plot of all and every American nanar.
Weird and non-intuitive are basically synonyms how most people use them. When they say it's weird, they mean it isn't intuitive...
I got a thought - isn't this just consequence of photon being a wave ? Since wave takes both paths and only collapses after interaction, it travelled a little in the other path that was'nt detected, so it can have some information about it ?
glad I found this channel
looks like you've misinterpreted the work of the CZcams's recommendations
So, you glady embrace the religious cannotations of modern science, where guesses and beliefs have become large part? Seems so.
@@josephjohnson3738 What? Mrs. Hossenfelder actively battles against that mentality.
"Two paths diverged by a beam splitter, and I took the one without the bomb.... and that has made all the difference." - Robert "Photon" Frost
Wow...you are rather good at giving intuitive explanations for this
Some high 🍞
Things like delayed-choice quantum eraser always bugged me the most in QM. Maybe you could make a video about it?
Even better, I'm writing a paper about it (like, right now). Hopefully also another video in the near future!
@@SabineHossenfelder That sounds interesting, is there a preprint available?
@@taw3e8 Not yet! Working on it, working on it...
@@SabineHossenfelder Waiting in superposition...is it good, is it not?
@@SabineHossenfelder I love you :)
I tried to understand that experiment earlier, but couldn't wrap my head around it.
After your explanation, it is much clearer (but still totally weird). Thank you (again) for being such a great teacher!
Amazing show can’t wait for next episode
Thank you for explaining this!
God, I love good experimental design. There is something so satisfying about seeing a complex idea and teasing it apart well enough to test it.
I have the DEFINITIVE solution to the Schrödinger's cat "paradox" I can with significant statistical certainty say that after 10 weeks in a box THE CAT IS DEAD.
This is empirical evidence after a hell of a lot of boxes (and cats)
and you can skip the whole "radioactive" part of the experiment, it makes no differense
@@qinby1182 how do u know that
@@goldenwarrior1186 it starves.
About the "weirdness" of wave functions: I think it's worthwhile to remember that wave functions are NOT the particles themselves, rather they are mathematical models for some aspects of the particles' behavior. It's easy to lose track of the distinction, but it is important. Among other things, it means that there may be aspects of a particle's behavior that the wave function doesn't model very well, and that's fine. It just means we need to understand the limits of our model.
We've been down this road before incidentally. Remember the Bohr model of the atom, and how it was derived with the concept of an electron traveling in a circular path around the nucleus? The model was initially held to be correct, since it proved useful for many purposes. In fact it was based on some faulty assumptions about how electrons function, but it so happens that the math worked out pretty well anyway, and the Bohr model was accepted as the correct model. Until we understood that it wasn't as correct as we thought and discovered its shortcomings.
Wave functions seem to be holding up much much better than that, and they remain a useful model. But it's only a model.
Good comment.
Much of the confusing, so-called "weirdness" of QT comes from making a literal exposition out of the Statistical Interpretation of the Wave-function ("Nature is Uncertain/Indeterministic"). If we regard the Statistical Interpretation as it was first set-out - in Max Born's 1926 letter to Einstein - it is merely a pragmatic, absolute limit on empirical observation, not a statement about the Nature of Reality.
@@donthesitatebegin9283 Well said. And, I suspect that some ideas (like the Many Worlds Interpretation) start with the mistake of confusing the wave function with reality. Maybe there are a billionty jillionty new universes created every nanosecond, or MAYBE it's simply that our model is imperfect. I know which one I'm leaning towards.
Well, wave-particle duality is not weird, it's misleading thinking that only one entity is involved with two roles; one a diffuse and wavy but at the same time compact. The way to see this is to visualize two entities that coexist together, one the wavy quantum package space and the other the compact particle that exists inside the package. This existence changes aleatorily inside the package between valid solutions, on each fluctuation only one eigenstate express. Now, over time the quantum package will have inside a probabilistic distribution of all the particle eigenstates i.e., Psi wavefunction describes space fluctuation but particle physical values depend on a probabilistic distribution of particle existence. So, taking Psi one will conclude that the position will be where it exists, momentum will be where the particle exists, energy will be where the particle is, and so on... all the parameters are developed by an operator over the existence Psi... weirdness has diminished
some people view the wave function as something like a physical wave that pushes the particles around. this interpretation is as valid as any other, unless we find something more fundamental about reality
@@heliusuniverse7460 Yes, but two entities that coexist together solve wave-particle duality, it also explains how communication is achieved in entangled particles, besides, it solves the trajectory problem between eigenstates, explains the collapse situation, and eliminates superposition ideas of simultaneous existence of all the eigenstate with the plausible ultra-high oscillation between them, one at a time idea... many fresh ideas that can be read in a short amazon book "Space, main actor of quantum and relativistic theories."
Thank you Sabine. I'm an utter layman and if not for your videos, I very may have fallen into the "quantam woo" trap. You are doing an invaluable service with your videos, it cannot be overstated.
I disagree with her re entanglement. Check it out for yourself. She over simplifies entanglement. I more or less agree re the quantum eraser experiment.
@@deandeann1541 What's wrong with her explanation of entanglement?
A very, very important comment.
The average person example is so simplistically brilliant. Thank you for making it understandable.
Your channel is wonderful, a much needed counter to the Deepak Chopra type folks who keep misusing Feymans's "nobody understands quantum mechanics mechanics" statement. Also- I never had heard of the bomb experiment till this time. You are a true educator in the finest way.
My brain can tell me something that didn't happen, which is, understanding this lecture. Whenever I think I'm overly intelligent, I simply watch a video on this channel, and I'm immediately brought back to reality.
yeah i didnt understand so much so i went to the comments for reassurance or a tldr lol i feel so dumb
@@pandapower5902 LMAO same, by the time I finished the video the inside of my head was just like ?????????????
If you scanned my head right now you wouldn't see a brain, you'd just see an endless question marks filling up an otherwise empty skull
I came to the comments hoping to find an ELI5, or at the very least, reassurance
I found the latter so I'm content.
Don't be afraid of learning. It isn't hard. What is hard is breaking down your own faulty beliefs.
Not necessarily, there are more possibilities than what you think.
Too many incoherent interpretations in QM. Nature doesn't work with probabilities and uncertainty. The only reason we tolerate all those speculations is the fact that they assert to have more or less 5% knowledge of the functioning of the universe; so, not bad for the superior beings of a small planet called Earth.
I thought "entanglement" was different from a ripped photo because the particles don't decide which property they have until they are measured. IIRC there was an experiment where you measure spin axis direction which shows a result different to the expected result for a "ripped photo" example.
The quantitative results are different because we aren't dealing with objects, but the basic idea about entangled pairs, be they classical or quantum mechanical, is fairly similar.
This video prompted me to subscribe. Excellent comparative examples and explanation for easy understanding. Excellent! Thank you Sabine!
It would sound less weird if you had mentioned what being a dud means, i.e. there is no functioning detector in the "bomb" path. So the weird thing is that the presence of a detector, or anything that can interact with the photon, leads to the change .
Maybe talking about bombs is distracting, but that's the way the original paper described it. For experimental purposes, using a detector that may or not interact with photons works the same.
If I understand the point correctly, what is being described here is essentially the same as the double slit experiment where the attempt to detect which slit the particle went through results in a corruption of the wavefunction such that the interference pattern no longer appears. Which I think is perfectly understandable; introducing "detectors" in the path of the interaction most certainly modifies the wavefunction of the overall physical arrangement.
@@kenlogsdon7095 Sure, but the difference here is that in the double slit, the detector interacts with the particle, or at least could potentially interact with it.
Here, we have it setup in such a way that even if the particle never goes on the path that leads to the detector (and thus shouldn't possibly have the information of its existence) and yet it behaves differently *anyway* simply because it could have interacted with it, even though it didn't. That's what is the crazy thing. The particle somehow knows if it's live or a dud before it interacts with the bomb and changes its behaviour accordingly.
Given that a single particle will exploding the bomb, how does a single particle tell you anything? A detection at B means the bomb went off, which is the same info you get from just trying to set it off directly. A detection at A tells you nothing.
@@zyrain Existing a live bomb, Photons only reach B if they go the "up" path.
Thanks for the (as usual) clear explanations! It would be nice to hear about the quantum Zeno effect as well, if you ever get the chance.
Wieder mal ne tolle Erklärung, bei der ich viel gelernt hab - danke!
These videos are mind-boggling - I have to watch them a few times - but they are excellent.
Oh boy, I have to re-watch this.
Same here, I am having a very hard time wrapping my brain around this. But as even S abine classifies this as "weird" that was only to be expected...
Thank you! I first learned of this a couple years ago, and I have periodically read the wikipedia page on it multiple times since then, but I was never able to properly wrap my head around what actually is happening in this experiment. Seeing the process built up step by step finally made it click for me!
If you understand QM, it is easy. If you do not know QM, it is genuinely weird. It represents a kind of non-locality different from EPR, which is not really that weird. It is based on correlations. Interestingly, EPR shows that QM does not obey the ordinary rules of probability as Bells Theorem showed (amongst other things).
@@bhobba Why then conceive and perform such an experiment by those who obviously understand QM? What did they want to explore?
@@nikoszaronakis1862 Many, many people understand QM. But there are many different interpretations of what it means, largely IMHO because we do not have direct experience with the Quantum World, so do not have an intuition to guide us. It is to emphasise this non-intuitive nature that some come up with such thought experiments - they understand very well how QM explains it. Remember, QM is a model of the QM world, a map if you want to use that sort of language, but as the saying goes - the map is not the territory.
@@bhobba Thank you! So they understand how it works (which I don't fully) but they don't understand why it works that unintuitive way (so there comes interpretation). But it seems like this experiment doesn't add to what we already know about how QM works. To me all these experiments sound like "what would be the analogue of QM behaviour in the macroscopic world and are there really any interactions/ impact on it".
@@nikoszaronakis1862 The way I look at them is to hammer home you can understand something and know why it works, but it still is weird. That occurs not just in QM but in many other areas as well. Take 1+2+3+4...... Any ninny can see it is infinity - but believe it or not, it is -1/12. I know why (it boils down to something in complex analysis called analytic continuation without going into the details), but it is still counterintuitive and weird. The root cause of the problem here is we make an unwarranted assumption the integers are just part of the real numbers - in fact, they are also part of the complex numbers, and powerful theorems from that area of math can be used. So one reason for these 'experiments' is to flesh out the unwarranted assumptions you are making.
Enjoyed your video very much thank you. 👍👍
Wow. Really makes you think. I frequently have to watch some parts more than once
Thanks for the video. Brilliant experiment! It illustrates the power of the quantum wave function to describe the real world.
I have spotted a trend in QM: anytime there is a disagreement between the math of the wave function and our intuition (or other ideas), the wave function wins out. In Ψ we trust.
The wave function plus a bit of epistemological autonomy. It's not an easy subject.
"If the bomb is a dud, nothing happens. The photon splits, takes both paths..." But surely the bomb (or its detector) would still block the photon even if it's a dud?
In the scenario of the article, with a 100%-efficient 100%-absorbing detector in one of the arms, the odds are: 25% D1 clicks (let's say D1 is the detector that always clicks when there is nothing in the paths), 25% D2 clicks, 50% neither D1 or D2 clicks and the photon is absorbed in the bomb's detector.
Forget the bomb, the scenario is about what happens along the photon paths and tells us nothing about the quality of bombs eventually wired to a detector along the photon paths.
In fact, the "detector" itself might be damaged and it doesn't register anything. We have no information about this by the click in D2. What we know is that the state of the photon has been altered by the presence of something along the paths, at least one of them. Could be the lab assistant's elbow.
@@ThePinkus Yes, I realise it's not about the bomb. But the detector would still block the photon no matter if the detector works or not. Why does she say "If the bomb's a dud, nothing happens"? 07:27 I guess she should have said "if the bomb is not there..."
@@Tom_Quixote I rechecked the article to see if I was missing something before answering.
Btw, You can find it on arXive, though I am not posting the link since the last time I did YT deleted my comment, or at least that was the correlation between link and deletion...why!?!?
The authors specify the sense in which the bomb is a dud at page 8 of their article.
A bomb is a dud when it doesn't have the detector that would absorb or scatter (prevent it to go on through the interferometer) the photon, so, they mean "dud" = "no obstacle".
When "dud" means just that, then the reasoning goes on as described.
Ok, in this way it makes sense.
You made quite the right question.
No other type of "dudness" can be detected by the apparatus, so it is very important to make this clear.
Thank You for asking!
@@ThePinkus That is not what a dud means in English.
@@yecril71pl I am of course well aware of it. ;)
That is why the authors of the article take care to explicitly state that by "dud" they mean just that, so that they can make the subsequent reasoning. They also are explicit on the fact that all of the bomb narration is just a dramatization of their reasoning, and I would add, so that it is not relevant.
It is totally obvious that the apparatus does not divine the quality of things in the common sense of parlance, but it only detects an obstacle along one of the paths.
So, given that the article is written in that manner, if we don't specify that by "dud", in this case, we have to intend nothing else than "no obstacle" we cannot follow their reasoning.
To get this I think you need to know the phase shift of the photon at beam splitter and why in dud case you can't see detection in both A and B (Hong-Ou-Mandel effect).
What I understand is that the experiment uses single photons. At the first beam splitter this single photon enters superposition. It is going both ways (lower and upper pathway), acts like 2 photons!
Beam splitters cause phase shift to photon and the arrangement is such that destructive interference is detected in direction to B (i.e no detection). Constructive interference is detected at A. So, if there is nothing in lower or upper pathways, then the photon is always detected in A (dud bomb can be counted as nothing).
If there is live bomb in lower pathway, follows 3 different possibilities.
- No detection at all means live bomb detonated and photon "really" went lower pathway. Why no detection in A? Must be because bomb itself acted as an observer and caused the superposition to collapse (i guess)
- Detection at A: Photon "really" went upper path and when in second beam splitter it had 50/50 chance to be observed in A or B. This scenario is indistinguishable from a dud bomb by the way.
- Detection at B: Photon "really" went upper path and again at the second beam splitter it had 50/50 chance to get detected at A or B. Detection at B can only happen can in the absence of interference!
Detection at A means bomb is a dud at 50% probability
Detection at B means bomb is live ad 100% probability
Detection at B means we know that bomb is live even tough we never went to peek there!
Gentleman, you dropped this: 👑
It was a whole lot better explanation than the video.
What would happen when we put the bomb midway when the detector is just about to interpret the result.
Hmm, you mean no bomb at all at start and then putting a bomb midway at lower pathway just before detection is made? I think it's giving result for the original setup. I mean changing "post photon" setup does not affect detection result. Like pouring oil to a racetrack after race car has passed does not affect it. I might be wrong...And thanks for reply!@@Jhakaas_Jai
The thing I don't understand is that why the beam splitter split the photon into two for the dud but doesn't split the beam, instead only create one path for the live.
@@edward3190 Hi! Photon enters superposition when passes 1st beam splitter on both cases (live or dud). After beam splitter the same photon goes both upper and lower path. Dud case is easier to understand. Live bomb case is harder. I understand what you mean (why no explosion always if photon goes both ways as said). In live bomb case we sort of look back what really happened. Detection at B happens only in the absence of interference (no photon lower pathway). Detection at B kind of destroyed photon in lower pathway, so is it influencing backwards in time? There is many world interpretation, but that, I think, makes problems for dud case (we should then see also detection at B 50/50, but we don't, so photon must really go "both ways", not one way in this world and other way in other world). I have not seen good fundamental explanation to this detection at B case. I am a layperson.
I just found this channel. Wow, my answer to the question "what do you find most strange about QM?" has forever changed. Thank you, this is awesome!
Sabine Hossenfelder: Her science is up-to-the-minute up-to-date---and her clothes are from the future.
I thought it was a Star Trek uniform from the 1990s
@@fillemptytummy :)
@gustavo champoski You could be right or you could be wrong. You are in a superposition of right/wrong.
I hope I can find stuff like she wears in Sydney, some serious outfit shopping is in my future.
It's a bit weird to say, "it's not weird, it's just counter-intuitive." Counter-intuitive is a plausible definition of 'weird,' and since quantum mechanics also violates long held, informed suppositions about physics, it seems counter to even informed intuitions. To then say you can't even imagine what would satisfy the equations, but it's not weird... I'm having a hard time imagining what 'weird' must mean.
Nothing is intrinsically weird or not weird. Things are what they are. We can put whatever labels onto them that we want but nature doesn't care what we call things.
Yes Bubba, even more... one can say all valid solutions are almost classical (eigenstates). The great difference with classics is that quantum existence is in an oscillatory situation with a frequency depending on its energy. The other great difference is that on each fluctuation, nature doesn't have defined information on which solution is, so it will assume aleatorily a temporary valid solution (one eigenstate per fluctuation). These two quantum realities are the so call weird QM; not just... because we can imagine a world with these two additional conditions over the classical ones. Hope this will reinforce your comment, regards
For me weird is paradoxical, counter-intuitive perhaps less so.
Maybe she thinks of "weird" in physics as something that defy explanation whereas "counterintuitive" would be something that have an explanation that seems illogical to most people. I don´t know, but her language would make more sense if she defined it that way.
Will, I think that some words as weird just express a reaction that gives interest to the reader to continue... the important issue is the reasoning in quantum world, how we must adjust it to the real nature atomic behavior and not only with the classical experience.... the arguments over just a sensational word.
Came here from the Delayed Choice Quantum Eraser video. Was gonna suggest you cover this topic, but I figured I'd check to see whether or not you already had first.
It took me ages to get my head around this. You did say ‘when the bomb is live and it doesn’t explode then you know the beam is in the upper portion only. I couldn’t see how this extrapolation would in its self collapse the wave function and I had previously thought that detection and collapse of the wave function only worked positively ie when you observe you see a distinct discrete position. I didn’t realise that a negative observation (nothing observed) also collapsed the wave function to an alternate discrete location.
I was thinking about it the other way around, like, it is the (theorical) collapse of the wave fuction through one path that results in a negative observation.
"The photon goes through one path, so the results tells you something about the path it didnt take"... i think it is the change in the wave fuction that actually causes the result, so i would say it is the wave fuction that tells you something about the possible path it could've taken
IIRC the success chance of detection without going boom can also be boosted quite a bit with more elaborate setups, right? (Though the chance that the bomb blows up in your face can't be brought to 0)
Adding more beam splitters after the bomb part should do it. Each time the photon splits and moves the direction or A or B2. Put a bunch of beam splitters (n) and the chance for the photon to always move to A is (0,5)^n. I could be wrong.
@@leolafortune1255 I think once you do the split, the particle will always take the same path in subsequent splinters. It's like a particle can have a probability less than 100% to be in a certain location, but after you measured it there, you can be certain that is where it is and it's not going to jump to one of the other probabilities in a later measurement.
What you want to bias the setup for is a situation where the probability of the photon to go to the bomb is near zero and the probability of going to the detector B when a bomb is live to be near 1. This requires more beam splitters in series before the bomb to increase the probability of the photon not blowing up the bomb. Then, when we go to where the beams recombine, we can sum all wave functions and get something converging on 25% and 75% for B and A respectively. To further increase the accuracy would require increasing the number of times you sum "the path not taken" - and I don't think there is a way to do this. You simply reduce the odds the path taken is the one that blows up the bomb and allow for multiple sample cycles - if you shoot 100 cycles and don't get a response from B or an explosion, then you have whatever the probability of flipping a coin with a 75% chance to be heads is 100 times and getting only heads. Or... 74.999x - or whatever the probability has been reduced to.
I think ... It may converge on 50/50, but I don't think so.
Bear in mind I have had no formal education or experience with this; I am well beyond my math.
@@Aim54Delta en.wikipedia.org/wiki/Elitzur%E2%80%93Vaidman_bomb_tester#Improving_probabilities_via_repetition
Of course, if we take the suggestive scenario too seriously, the better strategy is to detach any explosive or otherwise dangerous part from the apparatus, since the experiment doesn't really tell us anything about their quality.
So, the wave function travels down two paths, then interferes with itself to determine the probability of the particle being detected in a given location. Place an obstruction/detector along one of the paths, and the wave function collapses so that the original interference doesn't occur and the particle arrives where you would classically expect it to. This is just a fancy double slit experiment. In my opinion, that's plenty weird enough, but I don't see how it's conceptually different.
- Each detector adds a potential boundary condition on _probabilities_. You could do much of quantum mechanics using classical particles and known transition probabilities (i.e. a beam-splitter will bounce particles one way or the other with certain probabilities). Given a detection rate (a probability), you have a boundary condition that allows some path probabilities while excluding other path probabilities. To say that "a photon takes two paths" omits important information - "the photon can take either of two paths" - and also omits the reality of the beam-splitter's actual performance statistics.
- However, what is weird and non-classical is the guide-wave represented by the complex-valued psi function in _parameter space_. In the case of a photon, we can relate the wave behavior back to an electron oscillation and Maxwell's equations. But, when we do a similar experiment with electrons, the origin of the wave behavior is mysterious.
This is second part of Entropy, which includes entropy in terms of arrangement and probability.
Suppose there are three color balls, r(red), g(green), b(blue) arranged in three places available for them. So they arranged like; rgb, rbg, bgr, brg, grb, gbr. There are six ways in which they can arranged this is permutation. If one more different color ball or place is added, pernutation or number of arrangement increases to twenty four, that is four multiply to six previous arrangements.
Now as there is no preference of any arrangement and all are equally likelihood, so probability of any one selection is 1/6. Thus we see that probability of any selection decreases with increase in permutation or arrangements, and which is related to number of particles or participants which is ball in this case. Decrease in probability is increase in uncertainity or randomness or chaos.
Now if in above case if two of ball are of same color, suppose there are three balls of two colors r(red) and b(blue). Then above six arrangements reduces to three; rbb, brb, bbr. So when particles becomes indistinguishable, permutation or arrangements decreases and thus probability of any one arrangement is increase. This type of permutation is equivalent to combination of choosing two balls from three balls of different colors.
Probability distribution function of maxwellian particles which are considered as distinguishable is given by suppose, 1/X. Where X is permutation of particles. Similarly permutations of fermions and boson are X+1 and X-1. Both fermions and bosons are considered as indistinguishable particles but their probability distribution function is higher than maxwellian for boson is okay but lower than maxwellian for fermions shows that fermions are distinguishable particles and that is indicated by their spin half property which is basis for exclusion principle.
Does there are three kind of particles, two of them are governed by quantum statics or there is one kind of particle given as classical one and there are three kind of distribution density states.
Suppose permutation of particles having given higher energy is X, then its probability density function is given by, 1/X. This is known as Maxwell-Boltzmann distribution function where it gives probability of a particle having given energy at temperature. On increasing temperature, probability of particle having given energy increased.
Probability of a particle having given energy is 1/X and probability of a particle to not have given energy is,
1 - 1/X or (X - 1)/X. Now ration of a particle having given energy to a particle not having energy is, 1/(X - 1). This is known as Bose-Einstein distribution function and it tells about probability of a particle to have given energy if there is no particle have that given energy before or say ratio of probability of a particle to have given energy to go higher energy level to release given energy to come back to lower energy level. In textbooks it is interpreted entirely different.
Again probability of a particle having given energy is 1/X, and probability of another particle to have that same given energy is, 1 + 1/X or (X + 1)/ X. Now ratio of a particle having given energy and another particle to have same energy is given by, 1/(X + 1). Thus the probability of a particle having same energy as by another particle is decreased to if that energy is not occupied. This is known as Fermi-Dirac distribution.
So we see that there are no more two other kinds of particles obeying quantum statics but conditional probability distribution of same kind of particles.
I was following for a while but the grammar got me confused at the end >< are you saying that the particles have to be distinguishable, because if they weren’t, you could lower the probability distribution by having particles change into different particles?
Neil you are clearly very intelligent and your point seems good but the detail is lost because of grammar twists and turns.. if you could rework it … well I’d love to try to understand cause i think i know what your saying but its just not so clear for me to really get it…. Appreciate.
@@5ty717 May be my words stumble because I want to write in brief and second thing, to write against established conception is difficult due to people misunderstand that I lack understanding. Also this topic is tedious.
In the case of entanglement the conservation of angular momentum is preserved after the interaction where the beam splits. So part of the correlation can be considered local but the particles don’t contain all the information to predict what will happen in the future. That is established when you measure the first particle and it’s non local.
This is a great example of Sabine thinking like the highly trained physicist and mathematician she is rather than like the ordinary person she is addressing. "Weird" in this context means essentially that we don't know what it means (which Sabine recognizes too), but we ordinary folk can't help but still try to make sense of what it means *in non-mathematical terms* and when we try to do that we fail. Sabine, I suspect, is so at home with the mathematics that she does not need to "make sense" of it in non-mathematical terms. "Weird" is just this tension between the ordinary person's inability to make sense of such things as superposition in concrete terms and their impulse to still try to make concrete sense of it. The fact that such things can be handled perfectly simply (and non-weirdly) in mathematical terms does nothing to take aware that weirdness. When ordinary people try to make sense of superposition they are not thinking of it as the addition of wave functions - they are wondering what that addition of wave functions in the equation represents concretely in the world. "I think I can safely say that no one understands quantum mechanics". - Richard Feynman. That's why it *is* weird.
In college I never got credit for being 25% right - no one detected my brilliance
You were credited with a -75%. That was brilliant enough that the college still got their money! 👍
It's not easy being a photon after all :)
This is probably because you never 50% detonated.
I bet they didn't, you kept blowing up the school half the time!
Sabine, amazing video as always. I love how your videos get to the realistic and accurate visual models of these experiments like how you pointed out the misrepresentations of the wave pattern on the double slit experiment. For the bomb experiment with the Mach-Zehnder interferometer I think showing the mechanism behind the interferometer and why it works the way it does is the profound part of the way a photon behaves for this purpose. At the same time, I think it's worth pointing out on this video that the bomb is another act of measuring where the photon is forced to make a choice at the second detector. Without the detectors do we see something totally different? Has anyone ever tried the interferometer with photosensitive screens and a combination of this with the photo sensitive bomb?
This is one of the most intriguing and brilliant videos on quantum mechanics I've ever seen. Brava!
Also completely false. ;-)
@@schmetterling4477 really! In what way?
@@WeirdMedicine There is no quantum mechanics here. The bombs behave in a perfectly classical way and so does the interferometer. ;-)
@@schmetterling4477 interesting! So the interference would happen with, say, water? I believe this has been debunked, but I’m no expert. Thanks!
@@WeirdMedicine I don't know what you mean by "debunked". This is simply a boring like drying paint physics experiment which doesn't tell us anything about quantum mechanics.
That is amazing I wonder what application this could be used it it's like a weird logic gate
That are efforts targeted at using this in medical imaging, the idea is that this technique could help tissues be damaged less (and ideally, not at all).
However, I have to add that to my knowledge this possibility is currently extremely far off, due to the enormous experimental challenges it poses
The strangest thing about QM, are the physicists that act like their interpretation is the correct one.
No interpretation is perfect. They fail to see all the implications, and consider only the ones that fulfill their expectations.
Ah ha…. The nuance of being “correct “ (certainty) and the probability that you are correct
In science, any interpretation is only an analogy and nothing more than that, as long at it is trying to communicate more than the experimental facts. And interpretation will, more often than not, bring much more to the table than just that.
Yes, this is a good example of how mathematical physicists interpret the world. Unfortunately, when they do this, they give us incredibly bad ideas, like Quantum Computers, based on the mathematical idea of the superposition of "Quantum States." Math should be applied to physical reality, instead of creating new fantasies about nature and then using propaganda to get people to believe in these fantasies.
But only in a non local way.
If the wave who goes trought the uper path hits the second beam-splitter a long time before the wave who goes trough the bottom path hits it, then should we expected the detector B lights up (with 25% chance of happen) regardless what happens in the bottom path?
Oh - your videos are an absolute joy of clarity :
OK that doesn't mean I necessarily understand EXACTLY what you're saying in terms of outcomes, but I do know what I need to look at again, as many times as necessary, to (maybe) get to your level of understanding.
You recreate that same enthusiasm and insatiable curiosity as the very best of teachers who inspired me to achieve my potential in the things that I was already (fairly) good at.
You make me feel young again (despite the fact that I am now very old) - thank you so much for that.
Quantum mechanics has only a fraction of the weirdness of the people who dislike these great videos.
I'm suspecting that Michio Kaku creates fake accounts to dislike Sabine's videos 😂
Another great video Sabine! Just one correction - @ 8:51 the bomb can’t go to Detector B, it’s the photon that can.
That's not the only glitch does anyone remotely understand where A and B are and what oath is destructive and which is constructive??
@@leif1075 This is essentially a double slit experiment. If the photon really took both paths, it will combine with itself and create an interference pattern. You place one detector in the bright band of this pattern, and the other detector in the dark band. 50% of the time it will be constructive interference and land in a bright band, and 50% of the time it will be destructive and NOT land in the dark band. If the photon only takes one path, there will be no interference pattern, and 50% of the time it'll land where the bright band would have been, and 50% of the time it lands where the dark band would have been, making it bright. Therefore, if you ever actually detect a photon in the dark band area (and that will happen 25% of the time), that means the photon could only take one path, and the bomb is real and didn't go off. If you detect it in the bright band area (50%) or don't detect it at all (25%) then you don't know if the bomb is real or not, except in the 25% of cases where the bomb does go off.
If this isn't clear, imagine a regular double slit experiment, and you're placing the bomb in front of one of the slits. If the bomb is a dud, the light passes through the bomb (she didn't actually mention this, and I had to go look at the paper to understand that bit) and both slits forming an interference pattern. If the bomb is real, it intercepts the photon, thus blocking one slit, and you get a regular single peak distribution instead of an interference pattern. Though since you're doing this one photon at a time, you can only tell the difference in the patterns if the photon happens to land where normally there'd be darkness in case of interference.
@@stargazer7644 see she did NOT explain it that way in the video..sidnt thr video confuse you too..and you don't mean the photon can literally take both paths at the same time right..since it's not possible for a photon to be in 2 places at once so why did you say that?
@@leif1075 The photon can be in two places at once - it takes both paths at the same time - actually it takes all paths and interferes with itself. That's fundamental to Quantum Mechanics. That's why you get an interference pattern in the double slit experiment instead of just a band of light behind each slit. Look for some videos on the double slit experiment.
@@stargazer7644 but isn't I just the PROBABILITY that the photon cam be in one or the other path right? Think about it a photon is a tiny particle it is not large enough to be in both places at once? It's just the probabikity..the double slit interference can be due to multiple photons interfering with each other
For the entanglement in 4:50, is it different, however, for quantum mechanics since they are not just statically correlated, but as it were beyond-space/time correlated such that if I change one the other one changes too (as if are the same object, but actually not the same, and both at the same time)?
"if I change one the other one changes too" WHAT. Once you have made a measurement, you cannot unmeasure it. No change is possible.
Your videos are simultaneously exciting and soothing.
Unfortunately the bomb did break my brain at the end.
Isn't this equivalent to putting a wall between the first splitter and one of the mirrors? It also deletes one of the paths and prevents destructive interference.
Also, are we experimentally sure about the premise to the experiment, aka that it would only activate detector A without the bomb?
That's what i have a problem with. if superposition is a sum of two possible outcomes then even without a bomb, the same possibilities should exist.
Yes, this is exactly the same result as the classic double slit experiment except the output is binary not a distribution. The weird part is that with the entanglement apparatus described by Sabine you can perform stuff like nondestructive testing on a sample. Using Sabine's example and assuming 50/50 live vs dud population of bombs, you can separate 25% of the live bombs from a mixed population on each pass (meanwhile 50% will blow up and 25% will remain in the mixed population, which can be sampled again). This could have major implications in many areas of science,.medicine, computing, etc.
@@everfree2532 Superposition is a sum in the sense that two entangled states are combined to represent the entire system. Consider a simple oscillation (as a.proxy for the wave function); a superposition would mean the single oscillation is divided into two (or more) oscillations that can be added back together to derive the initial oscillation. Without a "live bomb", they are always added back together. But the presence of the live bomb sometimes "collapses" the wave function whereby the divided oscillation cannot be recombined, and this fundamentally changes the observable behavoir of the quantum object that the oscillation defines (in this case, a photon).
if i understand correctly, the dud is like empty space and does nothing. the bomb is like a detector and a wall and collapses the wave function and destroys the entanglement. this seems similar to the interference pattern disappearing when you introduce a measurement into the double slit experiment.
i think you could do something similar rig a bomb/detector or a dud-nothing to the double slit apparatus. if you detected a photon in the "dark" region of the interference pattern you would have a much better than 50% confidence that there was a bomb/detector attached to the apparatus. if it was in the "bright" region of the interference pattern, you might get into a monty-hall style debate about whether there was a still 50% chance the thing was a dud.
This is explained so well in so few words, and with a scheme, that I don't get what is supposed to be weird. It sounds and looks quite intuitive. What I am missing?
the second beamsplitter should let 50% through, but she says it does not, it would reverse the effect of the first. But what is the reason for the second beamsplitter to work totally different all of a sudden.
Its only possible when the photon wave duplicates at the splitter, and then interferes at the second splitter to go back to a non splitted wave.
I think the weird thing is that we still think of light as partickes, though they obviously are a wave, as the inerference pattern of the double slit shows.
I wonder how these experiments change using polarized lighhtwaves
A flowery way to talk about sums. Non- local correlation. That truly does make it not weird. Thank you.
9:00 "and that means you know something of the path the photon didn't take"
But that's similar to the interference on the double slit. The photon hits the screen, it interferes with its wave from the path it didn't take.
The difference is, in the "bomb" case, when it's live, it interacts with the wave function without interacting with the photon.
Very good video. But there is a mistake in the video at 4:00: "there's nothing weird about non-local correlations because they are locally created". However the really weird thing in the quantum entanglement that they can be created without any local interaction. So it is possible - based on quantum mechanic - to create quantum entanglement between distant atoms without any interaction between them in the past! (E.g. see Lucien Hardy version of Elitzur-Vaidman bomb testing Gedankenexperiment and the Quantum Liar Experiment.)
Even more, it is very common that distant atoms become entangled state without any local cause in the past, see e.g. Yakir Aharonov et al. - Interaction-Free Effects Between Distant Atoms
Good references!
In a QLE, isn't the entangled state the one between the two atoms and the photon? Each atom interacts locally with the photon state (its local component), and it is the photon state that is itself not local, thus mediating the correlations between the distant systems? This would not contradict that entanglement is established locally.
Am I missing something? (Very possible!)
@@ThePinkus I have written a reply for your comment but I cannot see where is it (youtube system just deleted it?)
Anyway I have found that there is more article QLE, so I have to clear which one I thought: R. E. Kastner - The Quantum Liar Experiment in Cramer's Transactional Interpretation (2009).
This thought experiment is very similar to bomb testing, only here there are two independent atoms in two branches of the MZI: one in each branch. Each atom is in superposition according to a given spin so that if its spin is in a given direction, it is in the branch and absorbs the photon (if it goes that way), and if it goes the opposite way, it does not. (It is much better explained in the article cited above, check it out there please!)
Now if in this arrangement the photon passes through the MZI to get to detector B (in the article it is "D"), then we know that one branch was blocked by one atom and the other was not, and the photon went to the direction where the branch was not blocked. It follows that the spin of the two atoms will be the same, so they will now be quantum entangled! Moreover, Bell's inequality will be true for them.
So in short, the two atoms were not entangled, and then the photon brought them into an entangled state without interacting with any of them AND event the atoms had not interacted with each other! So a entanglement is “generated” between atoms via an interaction-free method.
I quote specifically from the article: “
There are now some apparently strange and paradoxical implications of this result for the atoms for which photons were detected at D. If we choose to open one of the boxes, this will constitute a measurement of the z spin of that atom, and according to [equation], the other atom must always be found with the same spin. The usual account is that this means that one atom blocks one of the MZI’s arms while the other lies outside of it, and therefore the photon must definitely have traveled along the unblocked arm; but that would mean that it didn’t interact with the atom on the other arm, so how could it have brought about any kind of correlation between the atom”
The whole scenario is very similar what is written in Yakir Aharonov et al. - „Interaction-Free Effects Between Distant Atoms” article.
@@janosmadar8580 Thank You very much. I managed to find the articles. Perchance You tried to post a link? I tried in another comment to post a link to arXive and my comment got deleted by YT several times, until I posted without the link.
I find all of these scenarios very interesting respect to the reasons You described above.
One way to regard them is as benchmarks for interpretations or narrations (perhaps, we can consider narrations as weaker forms of interpretations).
E.g., "interaction-free" is interesting. On one hand, we know the QM description and in it the very possibility of certain results is a consequence of the fact that the photon state interacts, according to the QM formalism, with the obstacle. But it is the component of the photon state that it is not coupled to the obstacle which yields the possibility of that same result. Thus, on the other hand, this interaction between the photon state and the obstacle has those characteristics that we want to narrate as "interaction-free". We are using "interaction" in two different meaning, one in immediate reference to the formalism of QM, the other referring to something more specific.
For instance I would disagree, and at least recommend caution, on the statement that "the photon must definitely have traveled along the unblocked arm". This is interpretational, and I have doubts that this statement is legit. It uses backward conditionalization, which, while it is natural in classical mechanics, it is not explicitly part of QM.
Quite the contrary, according to QM there is no doubt, by the design of the scenario, that the photon state evolved into a superposition of the states propagating along both paths, then one of the components was coupled to an obstacle and because of this interaction this component was prevented from evolving into reaching the second beam splitter.
But a point that I think is important is that while only one component was coupled to the obstacle, this is one component of a superposition, the state is just one state, and it is this one state that is evolving.
There are interpretations that make the statement valid (I think, e.g., a pilot wave interpretation does imply from the result that the photon "must definitely have traveled along the unblocked arm", though it remains that the pilot-wave itself interacted with the obstacle to make it possible to yield the otherwise impossible result), but this is not a consequence of QM itself (unless one proves QM requires some of those interpretations).
There are two topics, perhaps related, that I find interesting.
One, is what we want to mean by "interaction-free" in relation to the QM formalism. And the other is the distinction of correlating entanglement and docoherence, since that distinction does change the statistics of the results.
Tentatively, I would say that "interaction-free" refers to factual occurrences of effects of the interaction, or factual causation. "Factual" becomes the critical term, and what it means weights for interpretations. These non-trivial problems aside, I am not surprised that we can have quantum-mechanical-interaction and still talk, in some sense, of "interaction-free" effects. A quantum state encodes more than facts, e.g. correlations, thus the evolution of these states (the mechanics of their interactions) might well be about the evolution of their correlations without necessarily implying the occurrence of facts.
In the scenario presented in the video, I would not say that, when B clicks, the photon went along the unobstructed path, but I would say that the event/result (which I assume to be a fact) of a click in B correlates with no observable occurrence between the photon and the obstruction, no factual causation between the two. It is, tentatively in this sense that I intend the result to be an "interaction-free" effect, while very much intending that its possibility is the consequence of the QM interaction as formalized in the treatment of the scenario.
As Sabine is, I think, pointing out, are also very interesting the implications of the non locality of the photon state combined with the establishment of entanglement with other systems it interacts with. I think that the scenarios, also in the reference You made, shows how a single non-local state can interact with different systems, depending on the localization of its components.
Now, this is also interesting also in relation to relativity, since the interactions are local but they affect the state as a whole, i.e. the description/formalization of the state in spatially separated regions. While EPR presents this issue for conditionalization (collapse), we have the same issue already for the evolution of the state. It should be very interesting to analyze how QM manages the consistency with SR.
@@ThePinkus Thank you very much! (And yes the links to arxiv server must have been the source of problem, thanks.)
I agree with you, especially with this part: “For instance I would disagree, and at least recommend caution, on the statement that "the photon must definitely have traveled along the unblocked arm””. I knew when I wrote this sentence that it was not a very correct statement, especially because the quoted article goes on to explain the problems with this statement too. (OK it is true that from the point of view of a specific interpretation (TI), but that is not so important now.) In fact, it cannot be said that the photon have traveled along the unblocked arm because there is not a definitive unblocked / blocked arm in this thought-experiment (against the bomb testing experiment) just only a superposition of unblocked/blocked arms even after the photon arrived to detector B. Rather, this is just a clue to see why it follows from that the photon arriving at detector B the fact that the spin of the two atoms are the same (quantum entangled).
So yes, we can say that the term "interaction-free" sounds a bit more mystical than needed. After all, if all that happens, which has always been characteristic of QM: that the overall experimental arrangement as a whole determines the outcome of the experiment (the probabilities) even if the individual (spatially separated) parts of the experimental arrangement do not participate in physical interaction/interaction. it is usually called "Quantum contextuality".
So, You are probably right that the point is somewhere in there that "quantum state encodes more than facts, e.g. correlations". After all, the article I cited at the end (Interaction free effect) is exactly about this: that correlations arise without factual effects, the formal reason for which is precisely that in QM the state vector is more than actual (realized) facts, because it also encapsulates possibilities. Bohr's interpretation that the wave function is just a function describing probabilities ends up with this. It is a different question how the possibilities eventually become concrete observed fact(s) when measured. This is the problem of measurement, which Sabine has made some very good videos about.
As for the relativity theory, it is difficult to judge. If we accept that there is an objective reality, then it can either be non-local or super-deterministic (and the last I think is no better at all). If it is non-local, then it is not compatible with relativity on some level since relativity theory is a locally realistic theory. But I do not want to go into this very deep problem. What I think that even if assume that physical reality is totally non-local it does not necessarily contradict relativity theory, because the actual facts (actual physical interactions) never contradict it (at least in principle). It seems that non-localization always remains in the "realm of possibilities", it cannot be used to get to specific real non-local physical interactions or non-local information transfer (faster than light or retro-causal information transfer).
@@janosmadar8580 "It seems that non-localization always remains in the "realm of possibilities"", this is exactly how I (tentatively, of course!) approach the question of QM being non-local and yet consistent with relativity.
(Note: the following is more in the interpretative context of QM, and in certain perspectives, than strictly about QM formalism and practical employment, of course there are alternative approaches and it is debatable)
Relevant is the fact that the notion of correlations, as the information about the combination of systems, is intrinsically non-local, thus, already in classical physics. It is the property of classical logic allowing the reduction of the information encoded in the correlation to that ascribable to each system separately (in the limit of uncertainty removal, i.e. in the limit of the reduction of the probabilistic formulation of the physical theory to the deterministic formulation, where probabilities, and with them correlations in particular, are eliminated), plus the ideal of local systems, that renders the theory entirely local. The terminology I tend to use is that correlations of classical logic are trivial, meaning the information is reducible to that of the separate systems, where instead the correlations of quantum logic do not satisfy this (which incidentally is the EPR claim), and are in fact holistic (according to the meaning that there is "more" information in the combination than what is available from the systems separately).
In this perspective, QL correlations are holistic, and consequently the theory is non-local.
We also have that a quantum state is itself non-local, but as we are discussing, it is a legit consideration (I am being perhaps euphemistic, I feel it is a lot more than merely legit) that a quantum state encodes first "possibilities" then, eventually (and non-trivially), facts. Then the requirements of locality that we ask from these possibilities might be significantly less stringent than what we can ask from facts.
(Fair warning -- long off-topic rambling coming!)
To express this dichotomy of possibilities and facts, I often end up talking about potency and actuality, co-opting the "ancient" philosophical terminology, and intending it as modal logic. It probably can get close, in general terms, to the notion of propensity (e.g. Popper), but also find its correspondence in the parallel duo of objective and subjective probabilities. I want to emphasize that I give priority to subjective probabilities because these are those that can be matched to physis, i.e. what is given to experience, for the simple consideration of the subjectivity of our experience. But I do not exclude that the structure of these probabilities, namely their logic, could indicate, as the implied content of our physical knowledge, the notion of objective probabilities. This is a general problem of relativity of any theory of subjective knowledge: how the structure of the subjective perspectives encodes the objective content of our knowledge. From this point of view, QM, with it probabilistic formulation, deals with the relativity of the subjective limitations of knowledge just as GR deals with the subjective perspectives on spacetime -- in that, the "objective content of our knowledge" is mediated by the structure of the subjective perspectives. It is following this idea that I consider that quantum logic indicates potency, or "objective probabilities", in the same sense in which (pure) classical logic indicates determinism. I am not aware that the traditional discussion on the distinction of, and relations between, objective and subjective probabilities recognize that their difference is not merely notional, but corresponds (could tentatively correspond) to a radical difference in their logic, which makes their difference and their relation determinable analytically.
Now, any attempt to identify an "objective content of knowledge" by this strategy renders that content to some extent metaphysical, having noted that physis is given subjectively. This helps emphasizing an issue that any theory of "objective probabilities" should deal with, how those "objective probabilities" connect to the subjective probabilities that match our physical experience. I had one of those revealing moments (as in "1+1=2, eureka!!! Ok, how is it I didn't see it before?") reading David Wallace "The Emergent Multiverse", where he discusses the topic of objective and subjective probabilities, and Lewis' "principal principle" which is intended to establish this connection between the two. Wallace rightly lament, as I would put it, hopefully not to much apocryphally, that that is an abstract conjecture, plausible perhaps, but still an out-of-the-blue guess. The extent of my realization is that, given the notion that we have to deal with objective probabilities not out of abstract considerations but because our physical theory indicates them, then it is this physical theory that sets the question of how objective probabilities connect to the subjective ones. More specifically, when that physical theory is QM, the form that that issue takes is the measurement problem. By that time I had already made the considerations that objective probabilities have the issue of explaining subjective probabilities, objective and subjective probabilities are distinct by their logic, the measurement problem of QM is the problem of the relation of quantum logic to classical logic (given a certain approach to it). So, 1+1+1=3, eureka! Followed by a serious scolding at me not seen it before. We can by this make sense of Heisenberg's intuition that the probabilities (the "true" probabilities, meaning just the epistemic and subjective probabilities) come from the "divide between quantum and classical" (not a quote, just vague phrasing).
Ok, please forgive this unnecessarily long and divergent rambling, back to the consistency of QM and relativity.
Another point that I think it is relevant for the strategy of making sense of the consistency between QM and relativity, is that correlations are actually "observed" by combining the results (e.g. determining coincidences of various detectors), meaning that we have to physically collect the information of the results obtained in different places. This process is limited by the speed of light (or, more prosaically, by the speed of the charted flight we are using to bring a well packaged, and fully back-upped, hard-disk containing the terabytes of data we collected to the other laboratory so that we can make a beautiful picture of a black hole), and we conclude that we can analyze the combined results to test our theory of correlations only in the intersection of the future cones of the events yielding the results themselves. This implies that the consistency of correlations, as physically testable, does not assume, nor requires, any time ordering of the events of the results, and, more generally, any relative localization of the events, as long as they do not causally affect each other (being then necessarily one in the future cone of the other, or at the same event, thus with a well defined time ordering). At some point, it might perhaps turn out that the distinction between causation and correlation, as relations in broader generality, boils down to this dependence or independence from time ordering or localization. This physical consideration leaves open for the theory the possibility of being non-local (e.g. with holistic correlations) and yet consistent with relativistic locality (local causality).
can we say the wave function is kind of field ? Fields as well as wave functions are described in each point and generally they spread out everywhere to infinity.
That’s one possible way of interpreting it, however this field would have to be extremely weird, in comparison to anything that we are familiar with, for the following (interrelated) reasons:
(a) as the experiment shows, this field apparently can give us information about events that would have happened, but didn’t (eg. the bomb exploding);
(b) the field seems to be affected by objects (eg. the bomb) without in turn affecting them in any way (like a “ghost” field) thus it seems to violate Newton’s third law (action-reaction);
(c) when you try to observe the photon’s position, the whole field collapses instantly; it’s impossible to observe the field gradually shrinking into where the photon is supposed to be;
(d) it’s impossible to observe what value this field has in one particular region of space; in fact, this local value doesn’t even give us the probability of finding the photon there, because only the distribution of the field over the whole space has any physical significance.
so, any attempt to describe the wave-function as an actual wave in a physical medium proves itself so strange that it doesn’t really help that much
I got a thought - isn't this just consequence of photon being a wave ? Since wave takes both paths and only collapses after interaction, it travelled a little in the other path that was'nt detected, so it can have some information about it ?
"And that's what we'll talk about today"
I love the little smile that goes with this every time.
I didn't understand the bomb allegory until realized that the "dud" bomb does not detect anything IE it does not exist. And the live bomb is detecting the electron. Thus decohering the system. So all the thought experiment is doing is detecting the presence of a detector in one of the possible paths without triggering the detector....25% of a time.
Exactly. It's a fake paradox. If the the 'Dud' were REAL then it would ALSO collapse the wavefunction (just like its 'Live' twin).
@@adamjondo The dud could simply take the signal and reemit it and the entangled photon would be retarded. Depending on the inherent delay in that scheme, it would not be measurable and so 'nothing happens'. Thinks... a Fresnel rhomb would pass the electric field and absorb the magnetic component, such that the emission is colinear with the source. Such a system existing in the path of one of two entangled beams would retard the phase of both of them due to inductance at the source (not magical or instantaneity (which still introduces a phase shift, a problem for single photon interference), simply inductance). Failing that a Fresnel rhomb could be introduced into both paths. It would take a long age explain just how rigged it is in the favour of generating the paradox, but you have to give them credit for putting in the work.
Then why does the live bomb collapse the wavefunction? it doesn't explode. So 'nothing happens'. It can superposition. All the jack-in-the-boxes have been defused and can only be sprung by a trainee.
This brings to mind something that I’ve wondered about.
Regarding photon emitters: Is it really possible to just emit 1-photon at a time?
I believe that I read somewhere that those photon-emitters actually emit a very small quantity of them, but that technology isn’t good enough yet to emit just one.
They often use sources that actually emit a bunch of them and then filter them until there's only one left.
@@SabineHossenfelder Wow! Thanks for clearing that up.
@@SabineHossenfelder Thanks, but that phrasing needs a little correction. "... until there's only one left according to the accepted theories (in particular QM)".
Is there any way to prove that there only one photon at the end without using all of our theoretical framework? Using just an intuitive experiment? That is where one can see the distance and possibly "weirdness" from our 'normal average' human experience (which are in them selves just as much assumptions).
As an aside i will add that I personally do mainly agree with your demonstrated pragmatic attitude in this video. That is more like what i think as science.
As for my earlier case, it is not that it is wrong to assume anything, but it is wrong to forget that it is an assumption. That is how basic logic works after all. It is just that we must work with some axioms, but that doesn't mean that they automatically apply outside our argumentation.
Here the outside could be some kind of 'reality' vs our theories.
@Sabine Hossenfelder the problem is scientis made assumptions from their beliefs of how things should work and use experiments to proof their assumptins
the experimets dont proof at all, actualy the experiments brings more problems, but they continue with their ideas making other teoris to try to explain the results of the experimets
so when new experiments can prove those teories them think the initial assumption was rigth
if I assume that
1- universe is filled with minuscle particles that are slightly repeled by electrons, the think that qe call vacuum
2- the movement of the electron produce a wave in this particles, the phenomenon that we call light
3- those waves interfer in the movement of the electrons
so a lot of problems in qunatum mecanics will be solved an experiments will prove this assumptions. it will bring a lot of other problems too
what it means? nothing
i just think that scientists shouldnt trust so much in the outcome of experiments and the matematics.
@@estranhokonsta There are actual sources called single-photon emitters, name self-explanatory, and there are single photon detectors or SPD sometimes called single photon counters. The SPDs are different from basic light detectors which measure the flux density of light and are DESIGNED to detect one photon. The construction and theory into building SPDs goes back decades and are sold ubiquitously. Nothing tricky about detecting light. Solar panels do it. The only difference is scale. If you measure the electrical pulse you've detected the photon using whatever math to balance equation when going from what your input is "light" to what that should yield at the output: the electrical which represents a detection.
Some things to note, your eye is sensitive enough to detect a single photon. Also if you're thinking of the Young Double Slit experiment just know that the same quantum results has been done with not just photons but with molecules. Last I read it was with the "bucky-ball" molecule: fullerene so I wouldn't get too caught up with dissecting the peculiarities of quantum physics with light.
For one thing, you can't say something is in multiple states at once until it is observed because you can't observe it in multiple states.
Thank you! :)
Can someone tell me how the 2nd beam splitter results in constructive interference to detector A, but destructive interference toward detector B?
What is happening at this beam splitter to cause the light to not come out the top?
Phase shifts occur when light is reflected on the front side of a mirror. This is described by the Fresnel equations. See the Wikipedia article on the Mach-Zehnder interferometer, for instance.
yh she didn't explain that part well at all
@@MrCrystalm8yeah, and because of thst i asked why a hundred times, than i tried to explain why could that be, i think it's because if the photon reflects at the beam splitter nothing happens but instead if it goes through the splitter some properties of the photons change
I am confused.
When you say : "We can't see wave function" does not that contrast with the double slit experiment? When I don't watch for single photon what I see on the screen, isn't it a wave described by the wave function? (and not a specific particle)
Yes, but you never see the wave-function itself. You see the photon on the screen with a probability that can be computed from the wave-function.
@@SabineHossenfelder so it's just math, just as I cannot "see" the number 2 physically but I can see two objects that can computed with 1+1 whereby 1 doesn't really exist but is a property of an object. LoL
@@notlessgrossman163 You're talking about the philosophy of how math is created in the first place...ALL of math is just an abstraction of the real world, and we use math to describe it. The only thing that you could colloquially think of as being "real" are finite positive integers...because you can count 1, 2, 3, 4....etc. and you can relate those things to another set of things in a 1 to 1 correspondence. But just because you can still count something doesn't make that number system real, because the things you are counting, are based on what you believe to be separate and identifiable objects...where Quantum Mechanics might have you fooled on what objects are actually truly separate and distinguishably countable things.
People talk bad about Stephan Wolfram, but he has a 2 hour talk about the concept of numbers and whether they even mean anything (called "How universal is the idea of numbers")...its a good talk that will scratch your itch.
9:10 Sabine still ignores the question and refuses to admit that life is not born out of chemicals. Life is a different non-chemical energy.
love the introduction XD
This seems to be a different description of entanglement than the one I was given to understand (quite some time ago and it may have changed, I guess)
In my scenario the entangled particles are in unknown states & are constantly evolving through possible values of whatever you would want to measure.
Until one observer makes a measurement, no one knows the exact state of either and that state is not unknown just because no one has looked at it. It's unknown because no observation has been made & (an important 'and', back then) the states of both particles, even if correlated, are changing. The act of observing A not only determines A's state at this point in space and time, but also fixes B's state no matter how far it may be essentially instantaneously (or faster than light at least).
If this is incorrect and we're just talking about having put a red ball in a bag and a blue ball in another bag and then placing them far apart and but then are to find that when we look at our bag and see a "red" ball, it means we know the other bag has a 'blue' ball, that is child's play. Big deal.
Yes my understanding is that entanglement involves non classical physics, since no signal can travel faster than the speed of light. It wouldn't be a lot less weird if a signal could be sent between entangled partials.
I'm not a physicist but I've watched a video talking about some scientists proving Einstein wrong in that there are no hidden variables in entangled particles, so wouldn't that mean that even if the correlation between the entangled pairs was locally created, their future states should not be dependent on each other (but they are)?
It is attributed to Einstein that he thought quantum mechanics is not complete, and a more complete description would entail further variables. But we don't exactly what he meant since his famous paper was written by one of his students, and of course he has a more subtle take. He saw what nobody then saw, and in addition he was essential in the discovery of quantum mechanics. Since then there have been the theorem of Bell and the experience of Aspect that showed quantum mechanics is complete. But that doesn't mean it is right. Actually it is weird and doesn't fulfill the criteria of a scientific theory.
That’s correct, the Bell theorem experimental results prove (independent of any possible theory) that there can not be any local hidden variables that would reproduce the correlations predicted by quantum mechanics. Sabine’s analogy of tearing a photo and sending them off, is not in accord with QM,… which is to say, it is invalid to presume that the measured attribute/results exists before a measurement is made. [The act of measurement supplies the conceptual form, as a condition for observability,…. so the attribute is created at observation]
@@clmasse : QM fulfills a non-naive criteria of a scientific theory perfectly well.
@@clmasse *
@@noumenon6923 A scientific theory must be consistent and predictive, quantum mechanics is neither.
Entanglement is said to be weird not because the entangled particles are simply like a left shoe and the matching right shoe each inside a sealed box and we just don’t know which is which, rather the weird part is that we have a method and choice to change one shoe to either left or right and the other shoe will always be instantaneously opposite, so reality is not grounded until observed and when observed, that information seems to propagate instantaneously (faster than light).
I am not an expert. But no "we have a method and choice to change one shoe to either left or right", we don't have a way to change one shoe to either left or right. It's just that if one shoe, when observed, turns out to be right, then other will turn out to be left.
@@abhay8437 Not true. You can flip the spin of one particle (without knowing what it was to begin with), and the two particles will still have opposite spins when observed. That is one aspect of entanglement that makes the universe not “locally real” and is pretty weird.
If it helps, there is no way to check that the entangled flip actually happened until an observer physically travels to and confirms the observation on the other distant particle, so speed of light limits verification, but the flip still happens instantaneously. We have done lab experiments where we sometimes flip one particle and other times don’t flip it, and confirmed that the entangled particle instantaneously flipped to the opposite with the same probability distribution as in the setting without any flips and did so before light could have had time to travel the distance between the two particles.
The above can’t be used to transmit information faster than light because you don’t know what the spin was to begin with and flipping it immediately collapses the wave function, so for information transmission purposes, the left shoe and right shoe analogy holds, but it’s an inaccurate analogy otherwise.
@@abhay8437 yeah, and what's so weird about that? I thought they were able to change the entangled particles or move them around and teleport them etc. didn't they do that with quantum mechanics? What about that?
This is not true. We cannot change the spin of one entangled particle and instantaneously change the spin of the other entangled particle.
Thanks for the video! Just found your channel and watched some more with joy! I have one question here however: Why do we act in the bomb experiment as the bomb isnt gonna "swallow" the particle and thus ending the experiment at that stage? I feel like in the examples before you succesfully demask the mystifying of quantum mechanics by highlighting the difference between a probabilistic prediction and their actual state (which for us humans seems to be only possible after the event). But somehow do not apply it to the bomb experiment.
At least that perceived inconsistency by me reflects my understanding and that might, or most probably will be wrong. Im asking still as I think this brings me closest to understand it correctly. So thank you, if you could take the time and consider my thoughts! :)
Life is weird,but death may be even weirder.So much to grasp,so little time. Great info,please keep it up,no nonsense physics.Thanks.
How death may be even weirder?
This video exploded my supposed understanding of quantum mechanisms.
If you're still having trouble figuring out why this is better than a coinflip (after all it still explodes 50% of the time), think about the problem this way: we want to test if the bomb is live so that we can store it for use later. In classical physics this is impossible, there is no way to ensure that the bomb is live without blowing it up. But in quantum mechanics, using the method shown in the video, there is a 25% chance it will NOT blow up but we WILL know that it's live.
Thank you
this is a far better explanation than in the video, thanks!
But I still don’t understand even from the video why we know it’s live. It seems like she is saying that the beam splitter in some cases directs the photon in one direction vs the other and in other cases it splits it into two (half photons? Different type of of photons?) and half a photon won’t blow it up?
@@JayMutzafi Yes, this is confusing in the video because of some of the words she uses (splits/beam splitter/recombines). If I understand it correctly, it's not a beam "splitter", instead it causes the photon to choose a direction with a 50% probability. The superposition property allows the photon to interfere with itself at the second "splitter" so if the bomb isn't detected the probabilities of travel (+50/-50) add back up to direction of travel to A. If detected by the bomb being triggered, the superposition property of the photon is removed and the photon is then forced to choose a path at the second splitter with 50% probability once again. In my mind, this still doesn't explain the superposition property.
Yeah it took me 2 hours to figure this out because half of the key terminology Sabine used in this video were misnomers. While beam splitters tend to be used in the context of splitting up a photon, in the video it is used to indicate a split in the photon’s decision tree. At least per my understanding.
This material is still too advanced for my head to wrap around properly at the moment 😅
The odd thing is not that the we gain knowledge about a path not taken, the odd thing is that the behavior of the photon changes at the splitter depending if there is a bomb and depending on whether it's live or not. If there is no detector, it seems to take both paths, if there is, it takes only one of them. That's the only odd thing here and it may not be really odd if we one day find out what space really is.
Consider this: Space is not empty room, it consists of threads and energy traveling through space must travel along one of the threads. E.g. energy always is a wave on a thread and thus it can only travel along a thread. For simplicity, ignore that threads may join, split up, be bend, or many threads may be knotted together. For a photon arriving at the splitter, there are two threads its energy could take; so it could take one or the other one or it split up and part of its energy could go either way. Nothing odd so far.
But what happens if you put an active detector in its path? How about this: An active detector changes the tension of the thread that it interrupts or forces to "go around it". This change of tension is detectable all the way back to the splitter. If now a photon arrives at the splitter, there are two threads it could take, but unlike before, these threads don't have an equal tension level anymore, their tension level is different and the photon can detect that already at the splitter, long before will hit any detector. This may change things a lot and maybe different tension levels, or let's call them stress levels force a photon to make a decision to take one thread or the other thread, as it will only split up if the stress level of both threads is equal.
It's like an electron in an electric circle getting to a point where it can either pass through a 100 or a 200 kOhm resistor. It will have to make a decision which way to go as both paths will be used, even though the 100 kOhm one will be used a lot more. How it makes that decision? Sure, that would still be a mystery. I'm talking about the photon here, we know it for the electron. And an electron will also not split up if both paths are 100 kOhm, so the photon behavior is different here but adding those threads (or paths or whatever you want to call them) to the mix and giving them a stress level, it's less spooky that the photon changes behavior at the splitter depending on whether there is an active detector or not if only an active detector changes the stress level.
How is an active detector different from an inactive one? How about that: An active detector forces energy, at least certain kind of it, passing by to interact with it, while an inactive one would energy just allow to pass by. Maybe anything that is reactive to at least some form of energy changes the stress level of a path and as path may not be limited in length and could spawn from one end to the other one of the universe, assuming for a second there is such a thing as an end, this could influence the behavior of energy that could take this path and is million of light years away from us.
In reality these threads would rather by a super complex asymmetric 3D grid, you could also say a field and this brings us back to quantum field theory. If our universe is made up out of fields then an active detector would just change these fields in some way and this will change how energy is traveling through these fields, wouldn't it? And from that perspective, I don't think this experiment is weird at all. It's not intuitive to us as the world we live in is full of detectors. E.g. there is matter all around us pretty much all the time that will interact with photons and thus behave like a detector. The stress level of all possible paths is pretty much always different and that's why we see particles only taking distinct paths in our world. Yet that is like if someone who was born on an island completely covered by forest and who also staid there his entry life would believe the entire world must be covered by forest which isn't the case but that's how the only world he knows and has ever seen.
Your videos are just best
So there is a way for a quantum particle (light) to interact with a sensor (bomb) in a way that does not change the particle in any way? How so?
A quantum state describes more than just determined events. E.g., it describes correlations (in the form of entanglement). Since (quantum) mechanics is the evolution of these states, we can have a physically significant (quantum) mechanical evolution of correlations that does not entails the occurrence of any determined event.
In the scenario of the article, the photon's state enters the apparatus "going through both arms", and it always "interacts", in the above sense, with the system placed in one of the paths. This is necessary for the possibility of the second detector to click, it wouldn't if the state went through unaltered.
Yet, when this detector clicks it never happens together with any other determined event occurring between the photon and the system placed in the arm. In this sense, which is entirely different than the previous one, the scenario is narrated as "interaction-free detection".
@@ThePinkus Let me rephrase my question, since it seems to me that you did not adress at all: The sensor (bomb) that (clearly has to) interact with the particle does not change the particle with this interaction?
@@oceanliketeacher The interaction changes the particle state, and one could say that the particle is changed or affected while meaning just that.
The particle going into the EV apparatus with an obstacle ("bomb", for dramatization I guess, but quite irrelevant what it is) is always affected/changed by the presence of the obstacle on one of its path in this sense, whatever the end result.
Formally, this is probably the more correct description -- the state changes, and when we say that a particle changes we just mean that its state changes.
But with this meaning, it is already impossible to give a positive answer to Your question "the sensor that interacts with the particle does not change the particle with this interaction?". By definition, the interaction does change the particle. There is no way to interact without changing the particle, in this sense (a trivial interaction that doesn't affect the particle is the identity operator, eventually multiplied by a complex scalar, but we probably want to call that "no interaction at all").
But! One might want to reserve "interaction", "change", "being affected" (and I am not suggesting that one should, in fact, I'd rather suggest that one shouldn't) for a stricter meaning, such as the particle being absorbed, or scattered from one of its undisturbed paths, or exchange some amount of energy, or some determined occurrence in general (which is vague...). When one says "interaction-free detection", he/she refers to this second meaning. And this meaning is open to a positive answer to Your question.
I'll try the "how is this possible?".
The first thing that the "bomb-detector" does to the particle state is decohering its state in the basis resolving the two paths.
For the first meaning the particle is changed by this, but not for the second (for a very ideal, decoherence-only, effect on the particle state).
If the "bomb-detector" does nothing else than this to the particle, both states along the two paths propagate to the final beam splitter and end up to the interferometer detectors.
The particle has not been absorbed, scattered, it had no exchange of energy, nothing whatsoever (of this sort).
Maybe, according to the second meaning we want to sat that the particle has not changed due to its interaction (which occurred!) with the obstacle?
Maybe. But is its behavior at the end of the apparatus the same? No, its different because its state has changed.
Statistics? No bomb-detector yields odds for clicks in the two detectors that are 100% D1 and 0% for D2. Bomb-detector yields odds that are 50% for D1 and 50% for D2. So the change is physically significant.
Do we want to call this change in behavior a change of the particle or not?
I guess it depends on preferences, context, what one wants to express.
For certain reasons that I have, to me this is already a very dramatic and significant change.
Formally the particle going into the apparatus can be represented as a superposition of the states going through the separate arms of the interferometers.
If it is not disturbed, i.e., there is no "bomb-detector", this choice of basis is just a matter of convenience for the computation of the effects of beam splitters and mirrors. It exits directed toward D1 with 100% probability. In the basis picture, there is a constructive interference toward D1 and a destructive interference toward D2.
If instead there is our ideal decohering "detector" in one of the arms, no matter which or both, the state is changed from a superposition into a mixture of the states going through each path. The mixture (a matrix, not a vector as a pure state) is diagonal in the basis resolving the two paths, by the construction of the scenario. So this basis is no longer just a matter of preference.
You can practically think of this mixture as a classical statistical distribution over alternatives, i.e. that the photon state is travelling in either one of the paths, but not both. Or You can just say that the mixture has the interference terms which prevented the photon to reach the D2 detector suppressed. What we get is that now the photon clicks the two detectors with equal probability.
Note that in the article setup the bomb-detector prevents the photon on its own path to reach the other side of the interferometer.
When it clicks, the click occurs together with a change of the particle in both of the previous meaning.
When it doesn't click, we are back to the question of the two meanings we can intend for a "changed particle".
I don't know if this makes it clearer how I am trying to answer Your question, or if I intended it as You meant it.
Hi Sabine, I am confused about something from your video. When you first start the bomb experiment, you imply that the photon has a 50% chance of going to the top path and the bottom path, but then you only show the results of the photon taking both paths at the same time. In the case where the photon takes both paths at the same time, the split photon will either interfere with itself destructively (nothing detected) or it will interfere with itself constructively (detected only at A). My first question arises here. Why is it that it is only ever going to detector A when it constructively interferes, yet later you imply that if the photon makes it through only the top path, there is an equal chance that it is detected at A or B? Next, you show the bomb scenarios where if it is a dud and the photons go through both the top and bottom paths, the bomb doesn't explode and therefore we should get a detection at A, same as if there was no bomb. When you add the live bomb, somehow the photon is allowed to take only the top path, which makes it possible for the detection at B. As far as I can tell, this has absolutely nothing to do with the bomb being live or a dud. It has everything to do with the inconsistency between two variables. First, whether or not the photon is allowed to take only one path, or if it has to take both paths. If it is allowed to take only one path in any scenario, then if it takes the top path only, you will get the same result whether the bomb is live or a dud, since there is no photon at the lower path to determine if the bomb would explode, and no lower photon energy to interfere with the upper photon energy prior to detection. Second, it makes no sense that the energy making it through the final splitter only has a chance of going to detector B when it only takes the upper path. If energy going through has a 50/50 chance to go to detector A and B, then the entire nature of the experiment changes, and it should have a 50/50 chance of being detected at B when it is constructively interfered with after taking both paths. If the energy going through the final splitter doesn't have a chance of being detected at B when it is taking both paths and is constructively interfered with, it shouldn't have a chance of being detected at B when it only takes one path. If detector A only detects energy that was constructively interfered with, then it wouldn't detect the photon taking only one path, because that energy wouldn't be constructively interfered with. Either way I look at it, the results aren't based on whether the bomb was live or a dud, the results are based on when you do or do not permit the beam to take both paths vs a single path, and when the detectors do and do not have an equal chance of detecting energy that makes it through the final splitter. Can you please help me understand?
A live bomb acts as a measuring device, and takes a measurement of the photon after the beam is split. When you measure the photon, then it can only have taken a single path. But when you don't measure the photon, then the photon takes both paths and interferes with itself at the second beam splitter.
Also, when the photon interferes with itself at the second beam splitter, it's always detected at A. It's not either interfering with itself destructively or constructively. It's at the same time interfering with itself destructively in the direction of B, and therefore nothing is detected at B; and interfering with itself constructively in the direction of A, therefore leading to a detection at A.
Hope this clears things up.
@@Fred-gs1ur It clears up the fact that it's a poor analogy. The live bomb is being treated as a measuring device, but the dud bomb is essentially being treated as if it doesn't exist at all. Why bother including the dud bomb, then? It only muddies the water. Just give the analogy with the live bomb only. In reality, the photon could care less whether you hear the bomb go off. That would be like saying if a detector had an alarm attached to it to let you know when it detected a photon, but the alarm broke, that the photon would completely ignore the detector. The photon is detected because it's interacting with a system that changes the nature of the photon, not because you hear the alarm go off. It's silly to assume that a dud bomb wouldn't cause the exact same result as the live bomb, but it doesn't matter, I get what was intended. Thank you for your time.
@@adorableinsect unless dud bomb not only can't explode, but it dosen't even measure the foton at all
A + for effort from me even though I didn't fully read your question due to you not adding linebreaks and me not understanding the explanation anyways.
you say “it’s silly to assume a dud bomb wouldn’t create exactly the same result as a live bomb”…it doesn’t create the same result. In the live bomb scenario, the photon cannot interfere with itself, and is then capable of being detected at B.
If there is a "bomb" (detector) in the lower path, and it's live but doesn't go boom, do we really know that the photon takes the upper path? Could it not be that photon is still taking both paths with 50% probability, but that the detector, even if it does not go boom, disturbs the anticipated interference pattern, so that the photon has a 25 % chance of being detected at A and a 25 % percent of being detected at B?
Not because the photon "took the upper path" - (in some clear sense as if it was a classical setting), but because the interference pattern is disturbed?
I think the point here is that if the bomb is unable to explode, it will not interfere with the photon at all.
A flowery way to talk about sums. Thank you
Normally, particle fields are continuous waves and are nonlocal. Only the interactions between two fields are discrete and local (particle).
*This* is really weird
This makes quantum mechanics sound a lot like being an electrician.
only that the amount of actual knowledge you gained is zero with 100% probability.
Amazing, thank you.
Amazing. Thanks.
What you described was classical superposition; it would be disingenuous to call quantum superposition just 'simple addition and not weird when it has extra properties due to the matrix math involved which makes the 'simple addition' non commutative.
I don't remember addition of wave functions being non-commutative. I only remember that the product, which is what expresses successive measurements, is non-commutative. That is also the case for generic (square) matrix math.
@@histreeonics7770 And you would be right 😉
A common confusion with quantum mechanics is people are taught to view particles as a point like object, which there is no evidence for. Hence wave duality "paradoxes". In reality they are more like fuzzy balls that can spread, which explains why an electron can form superposition bond with two protons symmetrically apart, as it spreads. We cannot assume every mathematical description has a realistic counterpart.
Elementary particles are point-like in a certain sense. The probability function is not the same as the particle being spread out. You can actually tell the difference. There is something called a form factor in particle physics which accounts for extended structures of particles. The electron has a charge, and when a particle interacts, the way it is deflected depends on whether the charge is spread out, or located at an infintessimal point in space. It's mathematics, so I can't really explain it any better than that. It is one of the reasons we know that electrons are point-like, but protons are not.
@@alienzenx There is absolutely no evidence that it is located at an infinitessimal point in space and it's not necessary to interpret it that way from the mathematics, please refer to "No Evidence for Particles" by Casey Blood to see the myths answered surrounding the conception of particle.
That's not true. Experiments in colliders have given an upper bound of the size of the electron, and it is much smaller than a proton, _a fortiori_ of an atom. The corpuscular aspect of a particle is that if it is observed (as a point say) at a position, it can't be observed at another position. When the position or the momentum of an electron in an atom is measured with a good enough precision, the atom ceases to exist.
@@clmasse I know that. The point is that the spread of the wavefunction is not the same as the particle itself being spread out, which we can measure.
@@JL-fh4qw We can never measure with infinite accuracy and there are fundamental physical limits to what can even be theoretically measured. Never-the-less, the electron is as far as we can tell point-like, and the spread of the wavefunction is not the same as the particle itself being spread out. The wavefunction has a finite value at every point in space, so you would have to consider particles to be of infinite size.
"You never see the average person." Quite true when talking about the statistics of averaging and it also occurs to me that it'could also be true of any individual at the psychological level. Sometimes I don't know who will show up today disguised as my office colleague.
On second note, there was no need for Planck to do sum of all modes as oscillator to gave energy density function of black body emitting radiation. Multiply it with function similar to Maxwell-Boltzmann distribution function was sufficient, but he want to consider or to show that considering density of energy different from particle density in Maxwell-Boltzmann distribution. Thing is that without particles, energy can't be stored in cavity, storing energy as modes in empty region is incorrect view, energy can only be stored or exchanged with particles, without them there is no meaning of quantization.
Imagine a single photon, whose wavefunction is split into two (see: beam-splitter) and one half is sent to Brazil and the other half is sent to London.
What this experiment shows is that a bomb (or, a "detector") can actually affect and interact with local parts of a photon's wavefunction in Brazil, ...EVEN THOUGH THE PHOTON WAS ACTUALLY OBSERVED IN LONDON!
In other words, we can physically interact with "something" (aka part of wavefuntion), that has something to do with the physical reality of the photon, EVEN THOUGH the actual photon is actually detected a million miles away! What did we interact with in Brazil then? Where did this "thing", that we interacted with, go when the photon was actually observed in London? Disappeared? Why did it disappear? How did it know that its "mother" (the photon) was detected in order to know to disappear and stop keep interacting with the physical world?
When the photon takes the upper path and gets detected at B, what happens is: the lower part of the wavefunction gets "blocked" by the bomb in a manner that doesn't detonate the bomb.
And this is what's paradoxical:
- If we fire a whole photon (the whole wavefunction) to the live bomb, it will always get "blocked" by the bomb (i.e., won't continue travelling on a straight line) and will ALWAYS DETONATE the bomb 100% of the time.
- But if we fire PART of the photon's wavefunction to the live bomb, it will again get "blocked" by the bomb but now it won't necessarily detonate it! This weird state of blocking (aka interacting) and not detonating (aka observing) is only possible when using PARTS of wavefunctions as probes (aka Quantum Superposition)!
- Thus, parts of the wavefunction can interact with the physical world even though "detection" (or, "detonation" in this context) doesn't always take place on that part of the world.
In other words, interacting with stuff and being detected by stuff, is not the same thing. The wavefunction can interact with stuff but not necessarily get detected by them, and this is what allows for the so called "Interaction-free measurements" (en.wikipedia.org/wiki/Interaction-free_measurement)
Another paradoxical thing is this:
- When the lower part of the wavefunction interacts with the bomb and its path gets "blocked" but without detonating the bomb, the upper path of the wavefunction now can travel to B and get detected by detector B. If this happens: i.e. if we detected the photon at B, then the lower part of the wavefunction, that had a life of its own down there, completely disappears from the world. In other words, it doesn't keep travelling around and doing its own thing, but vanishes from existence. (Or does it? -- open question) When measurement takes place at B on the part of the photon's wavefunction that has travelled to B, then all other parts of the wavefunction WHEREVER THEY HAVE TRAVELLED IN THE WORLD completely vanish as if they never existed. But THEY DID EXIST, and that we know because of the fact that detection took place at B, and that would only be possible if the lower part of the wavefunction had existed in the lower branch and got "blocked" by the bomb there!
#mindblowing
Something physical is traveling that path. No doubts.
spooky stuff
Also quite interestingly, apparently there's a way to keep improving the results further and get closer to probability 1 of truly being able to tell if a bomb is live! 🤯
When I first heard of this, many years ago now, I concluded that you could do something like that to determine the color of unexposed photographic film because you can determine if a photon of a particular wavelength would be reflected without the photon actually needing to hit whatever it is.
@@jonathanguthrie9368 That actually sounds even crazier!
Also there was another thing people did recently, which was to communicate using something similar. They managed to communicate and send information, without actually transferring any particles between two regions!
@@factsheet4930 *facepalm* You can do that easily already by tapping Morse code on a wall and having someone on the other side listening. There is no particle transfer, but there is information transfer. You guys really need to stop sensationalizing QM.
@@Elrog3 Nope, you will transfer sound waves. which is to say, movements inside a medium that propagate all the way to the listener... You transfer vibrational energy and also particles.
@@factsheet4930 Yes, there is energy transfer. There is no particle transfer.
In the case where the photon is detected by B with a live bomb that does not go off, this means that the superposition collapses without interacting with the live bomb, what exactly makes it collapse? does the bomb observe the superposition without exploding? Is it possible to create a bomb that explodes if it observes a photon ie breaks the superposition without necessiraly interacting with it?
omg! Thank you!