Simulation theory

  • #1
3
1

Main Question or Discussion Point

Hi guys, something has been bugging me for a while now and I thought I’d just ask it here in the hope someone can explain it to me.

Ever since Elon Musk brought it up, I’ve been thinking about the simulation theory (I know it’s not his original idea, it’s just the event that brought it to my attention). One of the things that really got me is when I thought about what limitations in a simulation would mean for the inhabitants, which basically brought me to quantum mechanics. When I look at things like the double slit experiment and quantum tunneling, it really fits nicely with a theory where the world around us has a default resolution and can locally increase fhe resolution when measurements at quantum level require it. Using functions (like wave functions) on a larger scale to save computational resources and only using additional resources when needed (due to an observer) seems to make total sense to me where QM without such an additional explanation makes no sense to me. Quantum tunneling for instance seems to me like the rendering of two bodies in a low resolution which causes them to overlap (think about low res games where this also happens). In the low res it would look like certain pixels have passed certain boundaries, but if a high res rendering of the full motion was created it would never appear so. The parallel here would be that an observed particle with a collapsed wave function would not tunnel as opposed to its low res cousin.

I’ve been googling things here and there but have not really found this idea to be properly looked at in a paper. Since I’m very novice at QM I assume there is some huge mistake in my understanding of these QM phenomena which will be fhe reason why physicists don’t really entertain this simulation theory linked to QM. My hope is you guys can tell me what’s wrong with my reasoning.

Thank you very much in advance.
 

Answers and Replies

  • #2
34,043
9,891
We can simulate the motion of billions of particles in classical mechanics. With quantum mechanics our computers struggle simulating 10 particles.
Quantum mechanics makes things more complicated to simulate, not easier. Superposition increases the effort exponentially with more particles.
 
  • #3
3
1
Thank you for the reply. What if it’s a storage limitation instead of a computational one? Like how a vector image for a 10mx10m poster is a lot smaller than a fully rendered pixelated image?
 
  • #4
3
1
Thank you for the reply. What if it’s a storage limitation instead of a computational one? Like how a vector image for a 10mx10m poster is a lot smaller than a fully rendered pixelated image?
But in this case every particle needs its own function so the comparison is flawed. Unless there is a way particles could share (parts of) functions?
 
  • #5
34,043
9,891
Same result: Storing a general quantum state needs much more storage than storing a classical state.
 
  • #6
Nugatory
Mentor
12,614
5,165
But in this case every particle needs its own function so the comparison is flawed. Unless there is a way particles could share (parts of) functions?
You are wasting your time speculating about this until you've learned enough quantum mechanics to set up and solve Schrodinger's equation for a simple two-particle system. Compare that solution to the classical description of the two-particle system and you'll start to see the problem.
 
  • #7
DrChinese
Science Advisor
Gold Member
7,243
1,053
The simulation theory is more of a speculative hypothesis rather than an actual theory. Is features no specifics, and actually does not in any way explain a single element of quantum mechanics. The reason time is not spent by professional physicists (past a few papers here and there) is that it doesn't hold much promise to resolve current outstanding questions.

On the other hand, standard QM continues to make an incredible number of predictions which are being analyzed and tested every days. Hundreds of new papers appear each week on these new ideas for confirmation and extension of standard QM.

As Nugatory mentions, you might want to learn more about the standard stuff before getting into the extreme side of speculation. Let me assure you: standard QM is as incredible as anything you can see in a sci fi movie.
 
  • #8
727
168
I can't find a quote from Musk but I don't think he was proposing simulation as a theory, he said that if it was a simulation you couldn't tell.

Cheers
 
  • #9
LURCH
Science Advisor
2,549
118
I think perhaps the term “Simulation Interpretation“ or “View” would be more appropriate. It is unlikely that any such state of affairs could ever be tested, because the results of such testing would be simulated results only. If it cannot be tested or falsified, it is not a theory, and this, I believe, is why there are so few researchers looking in to it.

Might sound like just semantics, but I think it’s an important distinction.
 
  • #10
132
16
I am new to learning quantum mechanics, but I am also interested in the simulation idea. More specifically I am interested in the idea that there are some simple mathematical rules that can simulate the universe in a way that is deterministic and so could be simulated on a computer.

One idea that got me interested in the simulation idea is the spooky action at a distance of entanglement. I like to think of entanglement in terms of a simulation because the entangled particles seemingly must share and communicate a state variable even though they may be very far apart. The delayed choice quantum eraser experiment is one such experiment that readily demonstrates this idea. The gist of this experiment can be simulated with the basic idea that the first of the entangled photons to interact immediately (FTL) changes the state of the other entangled photon with the result of its interaction.
 
  • #11
727
168
I am new to learning quantum mechanics, but I am also interested in the simulation idea. More specifically I am interested in the idea that there are some simple mathematical rules that can simulate the universe in a way that is deterministic and so could be simulated on a computer.
You may be interested in reading Stephen Wolfram's book "A New Kind of Science", this explores the construction of physical theories from Cellular Automata. I'm not sure if his ideas are popular amongst physicists or not and are probably considered speculative at best, but it does develop ideas that you mention above.

Cheers
 
  • #12
Boing3000
Gold Member
332
37
More specifically I am interested in the idea that there are some simple mathematical rules that can simulate the universe
I very much doubt so. For once the "simple rule" would need to include the "simple simulation rule". I think there is an infinite recursion right there.

in a way that is deterministic and so could be simulated on a computer.
Well maybe if you are ready to wait the answer for 7.5 million years :smile:

One idea that got me interested in the simulation idea is the spooky action at a distance of entanglement. I like to think of entanglement in terms of a simulation because the entangled particles seemingly must share and communicate a state variable even though they may be very far apart.
Me too. You can have a look at this. But please note that there is no communication possible this way, and even less "action" (thus not "spooky" at all)
 
  • #13
9,331
2,306
One idea that got me interested in the simulation idea is the spooky action at a distance of entanglement.
There is no spooky action at a distance in QM.

To really understand this, its one of the myths of QM promulgated in popularizations, you need to go back to Bell's original paper:
https://hal.archives-ouvertes.fr/jpa-00220688/document

All Bell is, is simply a correlation like Bertlmann's socks - and nobody would suggest spooky action at a distance between the strange behavior of his socks - rather just a quirk he had. But Bell showed something interesting - QM when dealing with entanglement and correlated particles it has statistical properties different to normal probability theory that describes Bertelmann's socks. This is not really surprising since we now know QM is what's called a Generalized Probability Model - these are theories that are generalizations of ordinary probability theory:
https://arxiv.org/abs/1402.6562

But lets say we want it to behave like ordinary probability theory - can we do that? Yes we can - by introducing FTL communication. But you do not have to - you can just take QM at face value and accept it is different to classical probability theory. That is Bohr's position mentioned in Bells paper - although I would not have explained it the way he did - Bohr unfortunately had a bit of tendency to be philosophical in his writings and somewhat obscure - but we will not go into that here. He also had the other characteristic of speaking softly, mumbling, and it was not easy to grasp exactly what he was saying. I will not go as far as to say it was 'gibberish' because his good friend Einstein, while not agreeing with him, I am sure understood it - but to us mere mortals its not as clear. Even Bohr's brother, Harald Bohr, a very good mathematician and friend of Hardy, found it difficult.

I personally am a bit of a maverick in that I don't even discuss locality in relation to correlations in QM due to something called the Cluster Decomposition Property, which is the careful formulation of locality in Quantum Field Theory - our most fundamental theory:
https://www.physicsforums.com/threads/cluster-decomposition-in-qft.547574/

For this fundamental property of Quantum Field Theory to make sense correlation must be excluded. So once you exclude it from discussion of Bells results you are left with - QM is simply as QM is ie nothing weird going on (other than of course the usual weirdness of QM - and there is a fundamental issue with QM that remains to be resolved - or not depending on your view - but that is not for here - start a new thread if interested - its got to do with the nature of measurement - in purely QM terms what is a measurement?).

There are a number of myths of QM - it would probably be a good idea starting out to read about this now so you do not get confused - which is very easy in QM - I have read a lot on QM including my bible - Ballentine - QM - A Modern Development (it is considered others bible as well - if you really want to know QM that is the book to get - but beware - its graduate level and needs to be worked up to) and I held myths even after studying that book, but they have been slowly whittled away at during my positing here:
https://arxiv.org/abs/quant-ph/0609163

Thanks
Bill
 
Last edited:
  • #14
63
8
TBH simulation theory is laughable ..it is not even a scientific theory it is sci-fi or pseudoscience in best case.
I dont understand what is the point of simulation theory! And why it is considered less nonsense than "aliens beought us here" theories!
 
  • #15
132
16
There is no spooky action at a distance in QM.
Many thanks for all of the links! I have taken a quick look at each and they all look really good to me. I intend to read through and understand each of them. I will let you know what I think when I am done.

Your first claim that "There is no spooky action at a distance in QM" is already contradictory to what I know and so I am immediately suspicious that this is an interpretation view and not really here nor there when it comes to reality. I am interested in the interpretations, but I am more interested in the what is actually going on and I guess that is an interpretation in itself, but I fail to see why we should let ourselves drop reality at this point. The more I learn about QM the more I am persuaded QM is simply something that describes an underlying reality using probability. Maybe your links will convince me of something else. I am doubtful, but I will try to keep an open mind.
 
  • #16
727
168
Your first claim that "There is no spooky action at a distance in QM" is already contradictory to what I know... I am doubtful, but I will try to keep an open mind.
Then "what you know" is incorrect. A correlation like Bertlemann's socks does not require action at a distance. Similarly a quantum correlation is just a different type of correlation with different statistics but also does not need action at a distance.

Make sure when you keep an open mind that nothing falls out.

Cheers
 
  • #17
DarMM
Science Advisor
Gold Member
2,370
1,396
The more I learn about QM the more I am persuaded QM is simply something that describes an underlying reality using probability. Maybe your links will convince me of something else. I am doubtful, but I will try to keep an open mind.
Note that hard constraints on such a underlying reality have been found. Consider these axioms:
  1. Measurements have a single objective outcome
  2. Spacetime is not locally highly nontrivial (i.e. there aren't billions of microscopic wormholes)
  3. The choice of what measurement one makes isn't strictly dictated by the system you're measuring (absence of superdeterminism).
I'll just call these the fundamental axioms. So if there is a underlying reality that can be mathematically described and assuming it doesn't violate the axioms above:
  1. Bell's theorem tells us it would have to be nonlocal
  2. Kochen-Specker tell us it would need to be contextual (i.e. the predicted value of a quantity depends on what it is measured with)
  3. The PBR theorem (https://arxiv.org/abs/1111.3328) tells us that either something in it obeys the Schrodinger equation or you can't prepare systems independently. So the underlying reality would be similar to QM in many respects.
  4. The Pusey-Leifer theorem (https://arxiv.org/abs/1607.07871) tells us that if the underlying reality keeps a certain time symmetry seen in experiments it has to have retrocausal signals*.
Copenhagen would reject the underlying reality being amenable to mathematical description (not that there is no such underlying reality).

QBism and Many-Worlds would reject the first of the fundamental axioms. Many-Worlds the "single" part and QBism the "objective" part.

't Hooft's cellular automata theory would reject the third fundamental axiom, in it QM is incorrect but the initial conditions of the Big Bang determine we will always happen to not choose the measurements that would expose QM being incorrect.

DeBroglie-Bohm just swallows all these results and retains the fundamental axioms.

*For anybody reading DeBroglie-Bohm doesn't have retrocausality and thus looses operational time symmetry at the fundamental level. However it possess it "effectively" from the same choice of initial conditions that mask the nonlocality (quantum equilibrium).

Everettian Many-Worlds also lacks operational time symmetry at a fundamental level and hence would need some fine-tuning in the initial conditions to mask this. I distinguish here Everettian Many-Worlds, where reality "is" the wavefunction and it alone, from Many-Worlds theories in general which would have additional elements in their ontology besides the wavefunction. The latter may possess operational time symmetry at a fundamental level as the Pusey-Leifer theorem doesn't cover them.
 
  • #18
132
16
Then "what you know" is incorrect. A correlation like Bertlemann's socks does not require action at a distance. Similarly a quantum correlation is just a different type of correlation with different statistics but also does not need action at a distance.
If I instead called it a "spooky correlation at a distance" would you say I was correct? Is it the use of the term "action" versus "correlation" that you deem incorrect? Because then yes, I would agree that it is better to use the "correlation" term which we know is accurate. I used the term "spooky action at a distance" because I thought it was a general slang term for the phenomenon we see in the EPR experiment and I was not trying to be specific to whether it was an action or something else. If it was more than this then I am lost to what is being called incorrect.
 
  • #19
9,331
2,306
Your first claim that "There is no spooky action at a distance in QM" is already contradictory to what I know and so I am immediately suspicious that this is an interpretation view and not really here nor there when it comes to reality.
The mere fact you say interpretational view is a clue that you have not thought about it deeply enough. All interpretations, every single one, is in accord with the QM formalism. There are interpretations with it (eg BM), and those without it (eg Ensemble) - so the logic is simple - QM does not imply spooky action at a distance any more than their is spooky action at a distance between Bertlmann's socks. Even Bell said so in his original paper - read the bit about Bohr and Copenhagen.

My personal view is a bit subtle. I believe, due to the cluster decomposition property, you should preclude discussions of locality for correlations. You can keep it if you want, but like the discussions about the Aether IMHO is ultimately pointless. Still, just like the the Aether it has one advantage - to some the world seems more sensible - but that is a personal thing.

Thanks
Bill
 
  • #20
9,331
2,306
If I instead called it a "spooky correlation at a distance" would you say I was correct?
I would say more correct. The correct statement is its required if you want QM to behave like correlations here in the classical world. Or, if you want the world to be more 'commonsense' like then its required. It's your choice. But remember the cautionary tale about the Aether. I can't prove it doesn't exist - but for reasons of simplicity most current physicists reject it. But its perfectly valid, and I can even point you to some theories about it eg:
http://ilja-schmelzer.de/gravity/

Note - by forum rules we do not discuss LET, aether, etc except like I did in passing or in a historical context - I used it simply as an example - so you cant pursue it any further on this forum:
https://www.physicsforums.com/threads/rules-in-the-general-discussion-forum.954263/

Thanks
Bill
 
  • #21
63
8
Although I respect everyones choice in his favorite interpretation, but I dont like the idea that you are free to pick what suites your taste.. the fact that there are many interpretations that are perfectly consistent with QM predictions doesnt mean they are all correct, they cant be.
Either:
- one of them is correct
- or all of them are wrong and we are yet to discover the true mechanism that describes reality more accurately.

The problem is that you cant prove interpretations by experiement, because they all give same predictions, which means that this field will stay wild open and any one can come up with new perfectly valid interpretation even if it was a weird one.

That being said I believe we should focus less on interpretations and focus more on actual QM, and I believe at the end the picture will be more complete making it more easy to connect the dots and find out the correct interpretation.
 
  • #22
34,043
9,891
Either:
- one of them is correct
- or all of them are wrong and we are yet to discover the true mechanism that describes reality more accurately.
Or (c) the question is meaningless and all interpretations can be seen as equally correct or not.
 
  • #23
494
292
but I dont like the idea that you are free to pick what suites your taste..
Well, facts don't care about what some people like or don't like.
 
Last edited:
  • #24
DarMM
Science Advisor
Gold Member
2,370
1,396
The problem is that you cant prove interpretations by experiement, because they all give same predictions, which means that this field will stay wild open and any one can come up with new perfectly valid interpretation even if it was a weird one.

That being said I believe we should focus less on interpretations and focus more on actual QM, and I believe at the end the picture will be more complete making it more easy to connect the dots and find out the correct interpretation.
Many of the interpretations do give different predictions, unfortunately it tends to be in extremely obscure scenarios that are difficult to test at the moment. There are scenarios where Copenhagen, Many Worlds and DeBroglie-Bohm for example will predict slightly different macroscopic behaviour, but you would need measuring the devices the size of the Sun to resolve the differences.

However, investigating the interpretations has revealed a lot about QM itself, much work in Quantum Information theory has been motivated by Foundational studies in interpretations. Also, there are now far less interpretations known to be viable now due to no-go theorems than there were forty years ago. Also many surviving interpretations have had unpleasant features (e.g. fine-tuning) discovered that make them seem less natural than previously.

Or (c) the question is meaningless and all interpretations can be seen as equally correct or not.
Perhaps I'm not understanding, but I don't fully get this. Many-Worlds and QBism for example make contradictory claims about reality. Is it truly meaningless to ask if one is correct? Of course both are compatible with current no-go results and hence can currently be seen as equally likely to be correct, but I don't think that extends to saying the question of which one is correct is meaningless.

To my mind Deepblu's two options are valid, what am I missing?
 
  • #25
63
8
Or (c) the question is meaningless and all interpretations can be seen as equally correct or not.
That does not make sense
 

Related Threads on Simulation theory

Replies
13
Views
1K
Replies
10
Views
591
  • Last Post
Replies
7
Views
925
  • Last Post
Replies
6
Views
2K
  • Last Post
Replies
1
Views
596
Replies
9
Views
1K
  • Last Post
Replies
2
Views
2K
Replies
2
Views
477
Top