How/why are pocket universes created during eternal inflation?

[phyguuy]

Hi, I'm looking for an explanation of how eternal inflation leads to the creation of universes in a multiverse.

I've read papers and watched videos on the topic, but I can't seem to get my head around it. I've heard words like decay, expansion and inflation in the same sentence to explain the phenomenon. This really confuses me, and I'd like to understand the process.

So, could someone please provide me with an explanation that would make sense to a 10th-11th grader in high school like myself who has only knowledge of basic physics and very basic cosmology?

Thank You.

 

[zoki85]

Why "pocket"?

 

[Torbjorn_L]

Hi, I'm looking for an explanation of how eternal inflation leads to the creation of universes in a multiverse.

I've read papers and watched videos on the topic, but I can't seem to get my head around it. I've heard words like decay, expansion and inflation in the same sentence to explain the phenomenon. This really confuses me, and I'd like to understand the process.

I'm no cosmologist, but I haven't seen anyone explain it in a sentence. In fact people use a whole lot of text to describe it. Here is one example, I hope it helps: http://scienceblogs.com/startswithabang/2011/10/28/why-we-think-theres-a-multiver/

Another resource that describes it (at length) is Susskind's youtube Stanford Lectures in cosmology, highly recommended. (You would want the latest of several for most up to date info.)

Let me try to excerpt the main parts though.

During the inflation process, the system that is a local volume of spacetime acts like a ball rolling down a potential well:

Akin to how people describe harmonic oscillators:http://osxs.ch.liv.ac.uk/java/model/physchem/Images/SHO-PES.png

[ http://osxs.ch.liv.ac.uk/java/spectrovibcd1-CE-final.html ]

Secretly, what we have sloppily called inflation comes out of some physics, the simplest such is a quantum field. (That is why the inflation potential has field strength instead of a physical dimension such as bond length on the x axis.) Such fields have quantum fluctuations:

Some parts of the inflating spacetime we looked at will fluctuate towards lower potential energies and stop inflating earlier. Those are the volumes that make local "pocket" or "bubble" universes.

Some parts of of the inflating spacetime we looked at will fluctuate towards higher potential energies and remain inflating. As it happens, the remaining volumes will end up larger than the volume we started with, despite some parts dropping out of the process. Hence the process can be named "eternal" inflation.

I hope that helps!

 

[bapowell]

So, could someone please provide me with an explanation that would make sense to a 10th-11th grader in high school like myself who has only knowledge of basic physics and very basic cosmology?

During inflation, you can imagine that the universe is filled with a special kind of energy density that causes the universe to accelerate in its expansion. In the classical universe, this energy density would be perfectly uniform across the cosmos. As inflation progresses, this energy density slowly drops, eventually reaching a value too small to support more inflation. When this happens, the universe stops inflating and undergoes a more leisurely decelerated expansion, known as the standard big bang cosmology.

However, we do not live in a classical universe -- quantum mechanics leads to fluctuations in this energy density across space. In this picture, the energy density still falls as inflation progresses, but it does not do so in a clean uniform fashion across the universe. There will be regions where it is higher, or lower, than average. Quantum fluctuations are usually very tiny; very rarely, however, they can be very large. The effects of these are dramatic: even when the average energy density is high enough to support inflation, there will be rare regions where a large fluctuation has driven the energy density too low to support inflation. This region will stop inflating and subsequently undergo the standard big bang cosmology. Such a region might correspond to our observable universe. As inflation progresses and the average energy density drops, more and more regions of the universe will stop inflating. However, there are other rare parts of the universe where a large quantum fluctuation might result in a region with much larger energy than average. These regions will continue to inflate. Even though these large fluctuations are rare, they result in a space that is inflating and therefore growing exponentially. In any given volume of space, they therefore come to dominate the volume.

This is the idea behind eternal inflation -- that such regions in which quantum fluctuations have driven the energy higher than average are plentiful. Even though our region of the universe has "dropped out" of the inflationary expansion, there are exponentially many regions that are still inflating. And there always will be. Those regions that stop inflating are sometimes said to "decay" or undergo "bubble nucleation". These terms derive from the physics underlying the inflationary expansion based on phase transitions. I can explain them in more detail if you want.

 

[Chronos]

One issue that eternal inflation suffers, as well as virtually every other multiverse conjecture I know of, is all 'alternative' universes are causally disconnected - which is to say they are observationally inaccessible and unable to affect any other universe. Their existence, therefore, smacks more of faith than empirical evidence. Mathematics is the only tool that appears to lend the concept any credibility. I find this unsatisfactory and eminently worthy of suspicion. I believe many people would agree that not all things mathematical possible are necessarily realized in nature - e.g., Boltzmann brains. See Alan Guth's paper http://arxiv.org/abs/hep-th/0702178, Eternal inflation and its implications, for further discussion. Sean Carrols discussion here http://www.preposterousuniverse.com...boltzmann-brains-and-maybe-eternal-inflation/ might also be of interest.

 

[bapowell]

Chronos, I agree. If, however, we come to find that polynomial "chaotic"-type inflation is consistent with observations, then eternal inflation is much easier to swallow. Sure, we won't have any empirical evidence of the separate Hubble patches, but eternal inflation is a mathematical consequence of a theory that would be otherwise well corroborated.

 

[Chronos]

bapowell, I am totally sympathetic. My point is math is a powerful tool and not to be ignored. I merely think it is like any other tool - it has equal power to be enlighten and deceive. It has led us to discoveries our ancestors never imagined, but, also down more than a few blind alleys. Math, unvetted, reveals possibilities, not reality.

 

[phyguuy]

Thanks for the explanations, you've helped me a lot. :)

 

[Chalnoth]

One issue that eternal inflation suffers, as well as virtually every other multiverse conjecture I know of, is all 'alternative' universes are causally disconnected - which is to say they are observationally inaccessible and unable to affect any other universe.

There's no reason for this to count for or against a theory. The observable consequences are what matter. That it has other consequences that aren't observable is irrelevant.

 

[Torbjorn_L]

Their existence, therefore, smacks more of faith than empirical evidence.

There is a lot of physics like that though.

The insides of black holes are also causally disconnected so the most analogous example. It seems odd to think that the event horizon would encompass nothing - they have a radius and the black hole has a mass, a spin and a charge.

Other examples I can think of is (the size of) point-like particles, the wavefunction, et cetera et cetera.

 

[zoki85]

Other examples I can think of is (the size of) point-like particles, the wavefunction, et cetera et cetera.

...with big difference that concept of a wavefunction is very useful in science while concept of a "pocket universe" is very useful in science fiction

 

[Chronos]

Which examples do you have in mind that have observational tension?

 

[julcab12]

... Most(if not all) of the argument on multiverse is associated with the interpretation of the visual representation of the indeterminate statistical distribution of particle--faraday wave. It would then be treated as multiple state with disconnected realities--Copenhagen. This type of formalism is later used to established a universal interpretation(cosmology) leading to prediction such as Multiverse.

But nature is tricky. There is also a 'possibility' that QM is deterministic-- that can raise doubt on discontinuity. 'IF' the mimicry of experimental pilot wave theory is directly interpreted or holds consistent. We can assume that such multiple state will have 1 unitary state that appears to be smirred around. The convenient thing about this notion is that we eliminate inconceivable huge inserts. IMO, If the prediction is somewhat dubious like multiverse/MWI and so on. It is only natural to check whether we are interpreting the statistics right. BTW I'm not saying that multiverse or any version of it is not possible since the notion is perfectly reasonable relative to the interpretation. I'm just casting a doubt on the interpretation of the premise.

 

[Torbjorn_L]

...with big difference that concept of a wavefunction is very useful in science while concept of a "pocket universe" is very useful in science fiction

Well, the concept of inflation is very useful and the predictions that can't be tested is still accepted. "Pocket universes" is very useful in inflation and multiverses are useful in the string theory landscape, FWIW.

But more than that, such notions doesn't append to the black hole example.

 

[bapowell]

... Most(if not all) of the argument on multiverse is associated with the interpretation of the visual representation of the indeterminate statistical distribution of particle--faraday wave. It would then be treated as multiple state with disconnected realities--Copenhagen. This type of formalism is later used to established a universal interpretation(cosmology) leading to prediction such as Multiverse.

This discussion is not about MWI or multiverse theories arising from possible interpretations of quantum mechanics. The OP is asking specifically about multiverse models within the context of eternal inflation.

 

[julcab12]

Sorry for stressing the issue a bit. I'm aware of eternal inflation. We already knew the conventional wisdom(Guth)--as a quantum scalar field w/or out tunneling. Inflation field produces tiny variation of densities from the poking of Quantum fluctuation but instead of slowly rolling down on its potential during inflation. The poke causes the field to go higher than low in eternal fashion-- inflation continues forever. Hence, we get eternal inflation.

On the side-note. I'm just making an argument of which the basis for all this model comes from the statistical approach of how we interpret quantum fluctuations in the first place. Of course we won't have problems if the fluctuations is considered a standard actual event in time and get along with it on whatever the consequence might be. But what if we have experiment that challenge that approach. Should we just ignore it?

 

[zoki85]

Well, the concept of inflation is very useful and the predictions that can't be tested is still accepted. "Pocket universes" is very useful in inflation and multiverses are useful in the string theory landscape, FWIW.

The concept of inflation is very useful for cosmological theory of our universe I agree. OTOH, concept of a pocket universe brings nothing of importance and isn't testable per definition. And if you're saying that parts of one completely unproven theory (string theory) allow the existence of a completely esoteric concept of multiverse, than fine, I have no further objections.

 

[Chalnoth]

... Most(if not all) of the argument on multiverse is associated with the interpretation of the visual representation of the indeterminate statistical distribution of particle--faraday wave. It would then be treated as multiple state with disconnected realities--Copenhagen. This type of formalism is later used to established a universal interpretation(cosmology) leading to prediction such as Multiverse.

But nature is tricky. There is also a 'possibility' that QM is deterministic-- that can raise doubt on discontinuity. 'IF' the mimicry of experimental pilot wave theory is directly interpreted or holds consistent. We can assume that such multiple state will have 1 unitary state that appears to be smirred around. The convenient thing about this notion is that we eliminate inconceivable huge inserts. IMO, If the prediction is somewhat dubious like multiverse/MWI and so on. It is only natural to check whether we are interpreting the statistics right. BTW I'm not saying that multiverse or any version of it is not possible since the notion is perfectly reasonable relative to the interpretation. I'm just casting a doubt on the interpretation of the premise.

Pilot wave theory reduces to MWI with one of the "worlds" labeled as "real". It has all of the exact same predictions as MWI, except that it adds the concept that one of the states is special. All of the other worlds exist within pilot wave theory. They're just not labeled real, as if that somehow makes a difference.

As I've said a number of times, "I don't like what this theory predicts, therefore it probably isn't true!" is not a valid argument.

 

[Torbjorn_L]

concept of a pocket universe brings nothing of importance and isn't testable per definition.

There is no such "definition". And it would be a wrong definition anyway, since there have been several ways proposed how to test it in the literature. (Say, testing weak anthropicity, bubble collisions, et cetera.)

if you're saying that parts of one completely unproven theory (string theory) allow the existence of a completely esoteric concept of multiverse, than fine, I have no further objections.

I didn't say that. I said that multiverses have found use in the exploration and potential testing of string theory as a physics of branes.

Here is another instance of making common but false claims: string theory is not "unproven". It has passed tests (say, predicting QCD flux tubes or the correct entropy of black holes), but its prediction has been surpassed by simpler theory (QCD) and postdiction doesn't make it competitive but catching up.

The correct question to ask is perhaps if it is physical. I'm reminded of energy concepts, that start out in some results of pure mathematics of differential equations and just happens to map to physics. It's even worse than string theory, because the math is without units. ... but that doesn't say that string theory is physics.

 

[zoki85]

And it would be a wrong definition anyway, since there have been several ways proposed how to test it in the literature. (Say, testing weak anthropicity, bubble collisions, et cetera.)

For metastable vacuum bubble collisions. Not for pocket universes that remain intact, growing, to constitute "multiverse"

 

[julcab12]

Pilot wave theory reduces to MWI with one of the "worlds" labeled as "real". It has all of the exact same predictions as MWI, except that it adds the concept that one of the states is special. All of the other worlds exist within pilot wave theory. They're just not labeled real, as if that somehow makes a difference.

As I've said a number of times, "I don't like what this theory predicts, therefore it probably isn't true!" is not a valid argument.

... Lol. My personal taste has nothing to do with it. Wording. :(

I'm not looking backwards on the subject. Instead I'm making an argument before the proposed prediction(mulitverse). Basically on the literal part of the formulation---treatment or view of quantum states having each of its own 'worlds', parellel or whatever. Should we be cautious and very literal in treating a physical system--QM simultaneously exist in all possible states or form? Yes! Of course. That is the reason we treat QM as statistical bec. of it's indeterminacy. I'd rather think that state of the system is "blurred or smeared" (which is in fact direct, 'more natural' and akin to observation) than proposing inserts on the variable like 'worlds' which is unnecessary .. IMO, the uncertainty doesn't need such inconceivable variable. It is perfectly reasonable to view it as fundamentally linear and has continuity all the way through collapse than parallel/casually disconnected containing volume of huge universes and so on.

 

[Chalnoth]

than proposing inserts on the variable like 'worlds' which is unnecessary

The other worlds are not inserts or extra variables. They drop out of the equations as long as you don't add in an extra assumption about wavefunction collapse.

I don't think the language of blurring or smearing is very accurate.

 

[julcab12]

The other worlds are not inserts or extra variables. They drop out of the equations as long as you don't add in an extra assumption about wavefunction collapse.

I don't think the language of blurring or smearing is very accurate.

I don't think the language of blurring or smearing is very accurate.

Isn't it a basic notion to think that a state of the system is blurred? --There is only 'one state' of the system, but it is not uniquely defined, the state of the system is probabilistic. Besides, Jittery/blurring effect is firstly associated with visual aspects/conditions before even considering the next framework. That's what i meant with direct. It is like the analogy of blurred picture of a person bec of out focus shot; not bec the person is naturally blurry :).

Anyways. If multiverse is a natural consequence of the equation, then it is what it is. I'll rest my doubt for now.

 

[Chalnoth]

Isn't it a basic notion to think that a state of the system is blurred? --There is only 'one state' of the system, but it is not uniquely defined, the state of the system is probabilistic. Besides, Jittery/blurring effect is firstly associated with visual aspects/conditions before even considering the next framework. That's what i meant with direct. It is like the analogy of blurred picture of a person bec of out focus shot; not bec the person is naturally blurry :).

Anyways. If multiverse is a natural consequence of the equation, then it is what it is. I'll rest my doubt for now.

If you have an experiment which measures a two-state quantum system and reports either A or B with equal probability, then every time the measurement is reported there is a part of the wavefunction where the system is in state A and the measurement apparatus reports that it has measured state A. There is likewise also a part of the wavefunction where the system is in state B and the measurement apparatus reports that it has measured state B.

I don't think this feature is captured by the idea "blurriness". One way that it is described that makes more sense is that information is lost to the environment. When the measurement is performed, if you observe the apparatus reporting state A, then the information about the part of the wavefunction where B is reported on the apparatus is lost to you forever.

And yes, it is a consequence of the equations. The only substantive criticism to this is that some have claimed it's not possible to recover probability from this description. But as David Deutch and others have recently shown, all you need to do is place observers within the wavefunction, and those observers will naturally experience what looks like a probabilistic system (with the correct probability rule).

 

[julcab12]

If you have an experiment which measures a two-state quantum system and reports either A or B with equal probability, then every time the measurement is reported there is a part of the wavefunction where the system is in state A and the measurement apparatus reports that it has measured state A. There is likewise also a part of the wavefunction where the system is in state B and the measurement apparatus reports that it has measured state B.
.

O.T
... Ok. That's how it appears in the detector but how will we specify which part of the wavefunction is a "world"?If the usual understanding is exactly analogous to whether an abstract formulation is satisfactory in the absence of clear link(which part/s) between observation and mathematical formalism. Then one can argue that it will only stay 'real' when the state of our branch and the measurement basis are assumed rather than derived And can only be resolved by postulating additional axioms or completely different formalism which is problematic IMO. I suspect the same derivation of MWI.

How does wave function branch when there is a preferred decomposition into the system and environment (Between system and measuring apparatus) especially when the system is not well defined? 'QM is based on the foundation of probability with problems that affect mostly on our perception of reality. Especially the time parameter which greatly affects visual outputs. Others include in response to Wallace argument:

http://arxiv.org/PS_cache/arxiv/pdf/0906/0906.2718v1.pdf

The Decision Problem

"The agent is choosing between different preparation-measurement-payment (or p-m-p) sequences (Wallace calls them acts, but this terminology is counter-intuitive, so I avoid it). In each sequence, some quantum state is prepared, then it is measured in some basis, and then rewards are doled out to the agent's future selves on the basis of the measurement outcomes in their respective branches. An example sequence: a state is prepared in the superposition 1/2 |up> + sqrt(3/4) |down>, a measurement is made in the up-down basis, then the future self of the agent in the |up> branch is given a reward and the future self in the |down> branch is not."

The Representation Theorem

"The preference ordering over sequences induces a preference ordering over rewards, because for any two rewards R1 and R2, there are p-m-p sequences which lead to R1 for all branches and R2 for all branches. If any sequence of the first kind is preferred over a sequence of the second kind, then reward R1 is preferred over reward R2."

I

I don't think this feature is captured by the idea "blurriness". One way that it is described that makes more sense is that information is lost to the environment. When the measurement is performed, if you observe the apparatus reporting state A, then the information about the part of the wavefunction where B is reported on the apparatus is lost to you forever.

(with the correct probability rule).

Ah. Ok. I don't get that loss information part. Can you please elaborate a bit or do you have a link? Thanks

 

[Torbjorn_L]

For metastable vacuum bubble collisions. Not for pocket universes that remain intact, growing, to constitute "multiverse"

What would be the difference?

"Many modern theories of fundamental physics predict that our universe is contained inside a bubble. In addition to our bubble, this `multiverse’ will contain others, each of which can be thought of as containing a universe. In the other 'pocket universes' the fundamental constants, and even the basic laws of nature, might be different.

Until now, nobody had been able to find a way to efficiently search for signs of bubble universe collisions - and therefore proof of the multiverse - in the CMB radiation, as the disc-like patterns in the radiation could be located anywhere in the sky. Additionally, physicists needed to be able to test whether any patterns they detected were the result of collisions or just random patterns in the noisy data.

A team of cosmologists based at University College London (UCL), Imperial College London and the Perimeter Institute for Theoretical Physics has now tackled this problem."

[ http://www.ucl.ac.uk/news/news-articles/1108/110802-first-test-of-multiverse ]

And nit or not, whether or not some smaller parameter volume is not testable, the initial claim that this wasn't testable at all. It is.

 

[Chronos]

That it is possible to propose a test of the multiverse hypothesis is unrelated to the possibility may, in fact, be untestable. Furthermore, no such 'test' has yielded observational support. The above referenced paper, http://arxiv.org/abs/1012.1995, is no exception. In the abstract the authors note "... We conclude that the WMAP 7-year data do not warrant augmenting LCDM with bubble collisions... ". You need not concede a negative result constitutes falsification, but, you must at least be willing to concede the hypothesis may not be testable. For further discussion, see Peter Woit's article at http://www.math.columbia.edu/~woit/wordpress/?p=6551.

 

[zoki85]

I wouldn't be so confident the laws of general relativity and quantuum mechanics remain intact as we get back close to the "Planck epoch"

 

[Chalnoth]

That it is possible to propose a test of the multiverse hypothesis is unrelated to the possibility may, in fact, be untestable. Furthermore, no such 'test' has yielded observational support. The above referenced paper, http://arxiv.org/abs/1012.1995, is no exception. In the abstract the authors note "... We conclude that the WMAP 7-year data do not warrant augmenting LCDM with bubble collisions... ". You need not concede a negative result constitutes falsification, but, you must at least be willing to concede the hypothesis may not be testable. For further discussion, see Peter Woit's article at http://www.math.columbia.edu/~woit/wordpress/?p=6551.

Your point?

Most of the proposed physics theories have a regime of parameters where they are either untestable or exceedingly difficult to test. It would be unfortunate if this were the case. But why do you think this is an especially important thing to note?

 

[Chronos]

My point is proposing a test does not confer the hypothesis testability. You can propose a test for the existence of refrigerator light fairies - which does not confer that hypothesis testability.

 

[Chalnoth]

O.T
... Ok. That's how it appears in the detector but how will we specify which part of the wavefunction is a "world"?

This goes under the heading of "einselection". You can find a decent amount of stuff with a search of the term.

 

[Chalnoth]

My point is proposing a test does not confer the hypothesis testability. You can propose a test for the existence of refrigerator light fairies - which does not confer that hypothesis testability.

Yes. So?

Popper's overly-simplistic view of falsifiability hasn't been a realistic part of science for a long time now. A better way of understanding this is "verification" instead of falsifiability. Consider a theory which has a free parameter that could be between 0 and 1. If this free parameter were in the range from 0-0.9, then there is no way to detect it. If the parameter is between 0.9-1, then it is possible to build an experiment that would detect it. This theory is not falsifiable in the strict Popperian sense, but that doesn't matter: we can still look for the theory in the parameter regime from 0.9-1. If that parameter regime happens to be accurate, we can then verify a detection through further checks down the road.

 

[julcab12]

This goes under the heading of "einselection". You can find a decent amount of stuff with a search of the term.

..That's the thing einselection is still a construction bec there are no experimental evidence to verify that each space is real. The formula assumes by installing a version of real space(hamiltonian space) to each probability space or pointer.

H=∑p|p⟩⟨p|⊗H(p), where H(p)-- Hamiltonian space and P is the "pointer".

 

[Chalnoth]

..That's the thing einselection is still a construction bec there are no experimental evidence to verify that each space is real. The formula assumes by installing a version of real space(hamiltonian space) to each probability space or pointer.

H=∑p|p⟩⟨p|⊗H(p), where H(p)-- Hamiltonian space and P is the "pointer".

Why do you think the default assumption should be that they aren't real?

 

[julcab12]

Why do you think the default assumption should be that they aren't real?

.. For me. I'm just cautious. My default is "i don't know or maybe". Nonetheless. I'm pretty sure that the extra parameter is indefinite/uncertain(for now)-- it can be a supplement to real space OR it can also be real(it is treated real anyways). Such a huge black swan..