Probabilistic game of life as a toy model for Everett's interpretation

[ShayanJ]

TL;DR Summary

Many world's interpretation of QM can be reinterpreted as the universe parallel processing of all possibilities at a fundamental level and Conway's game of life with probabilistic rules can be a good simple model for it

A while back I was watching an interview with Sean Carroll and for some reason his explanation of the Everett's interpretation at that particular interview clicked for me and I became a proponent of it.
But I always thought it made more sense to look at it at as the universe being a giant supercomputer that is processing all the fundamental possibilities in parallel.
To understand it better, you can think of Conway's game of life but instead of the usual deterministic rules, imagine that each rule only specifies a probability for a given cell to be dead or alive and then imagine that each time a a probabilistic rule needs to be applied, you spawn new processes in accordance with the probabilities associated with each possibility.
In such a world, things would appear random at first but after long and careful observations, those probabilistic rules can be discovered and it would look something like the current state of QM in our world.
I know this is just a philosophical alteration to an already philosophical idea but It seems to be a neat idea to me, so I thought I'd share it because I think this could help people have a better understanding of Everett's interpretation.

 

[.Scott]

Given your thoughts, your analogy between Physics and Conway's game (which decades ago, I wasted way too much time with) could be improved by running Conway's game in reverse. Would that be "CounterConway".

Once you run Conway's Game forward, the information left in that universe is not enough to determine its history. More generally, for any given pattern, there may be none, one, or many patterns that could have generated that pattern from the previous "tick".

So when running CounterConway, you would run into Everett's scenario - and more. Not only could there be more than one universe deduced at each moment (spawning more processes), but there also could be none - a last moment in time - process terminated.

But that is really not how actual Physics in our actual universe works.

When dealing with information at the quantum level, there are well-established theorems for no cloning, no deleting, and no hiding. So, at the quantum level, you can always have a past that is as well-determined as your future. And if you shuffle off one possible future by spawning a new process, that process would always continue to interact with all the other processes spawned for that event.

I am always willing to take any absurdity and extend it beyond its original bounds. So in that spirit, I have reexamined Schrödinger's cat. In my telling of the story, the cat may not simple come out dead or alive - but those two versions are allowed to interact and interfere with each other. So we may discover that the live cat has staved off starvation by snacking on the less-lucky version of himself.

 

[PeterDonis]

at the quantum level, you can always have a past that is as well-determined as your future

This is only true if (a) you know the exact quantum state of the entire system you are interested in, and (b) unitary evolution always holds. If we limit ourselves to the basic math of QM and don't adopt any particular interpretation, (a) is only true in highly idealized experiments and (b) is never true, since we always have to apply the collapse postulate to make further predictions once we know the outcome of an experiment, and that throws away information in our model.

In the MWI, if we assume an idealized experiment in which (a) is true, (b) is also true. You correctly point out that such a model is not the same as Conway's Game of Life, since in the latter, the analogue of (b) is not true: time evolution can destroy information.

 

[PeterDonis]

in that spirit, I have reexamined Schrödinger's cat. In my telling of the story, the cat may not simple come out dead or alive - but those two versions are allowed to interact and interfere with each other. So we may discover that the live cat has staved off starvation by snacking on the less-lucky version of himself.

This is personal speculation and is off limits here.

 

[PeterDonis]

Moderator's note: Thread moved to QM interpretations subforum.

 

[PeterDonis]

if you shuffle off one possible future by spawning a new process, that process would always continue to interact with all the other processes spawned for that event.

This is not correct as far as QM and the MWI is concerned. The "branches" corresponding to different measurement outcomes in the MWI are decohered and do not interact with each other. That fact is essential because it is the reason the MWI can claim to reproduce our actual experience, which is always of just one outcome happening: only if the branches corresponding to different outcomes never interact will that be the appearance in any single branch.

 

[PeterDonis]

I always thought it made more sense to look at it at as the universe being a giant supercomputer that is processing all the fundamental possibilities in parallel.

For the reasons given in my responses to @.Scott in previous posts, I don't think this is a good way to look at it.