is this the real life, is this just a complex cosmological simulation?

October 14, 2012 — 6 Comments

reaching out

Welcome back to yet another installment of the question of whether we’re all just products of an advanced simulation that created an entire universe, but this time, instead of plunging deep into the lore of the Matrix with Moore’s Law hijinks and philosophy, we’ll be hunting for physical proof that the universe is actually a simulation in the realm of quantum chromodynamics. What exactly is quantum chromodynamics? It’s the study of interactions between point particles that make up matter as we know it and its more exotic forms we sometimes glimpse when we smash atoms with enough force. How these particles interact basically defines what is and isn’t possible across the entire universe because without their fluctuations, the cosmos would still be a zoo of particles in no way, shape, or form resembling the planets, stars, and galaxies we know and love today. So the big question is whether those point particle interactions have a very telling limit and what this limit could tell us about the underlying nature of the universe.

One idea is that these limits should fit a three dimensional lattice around the interaction, which essentially means that interactions between point particles should fit into a predictable model on which other interactions can be neatly stacked. Since the authors of the idea in question aren’t computer scientists, they refer to this packet of quantum information as a cubical lattice. Being a computer person, I would refer to it as a voxel; it’s a three dimensional pixel which makes up the environment in which the simulation should take place. Think of Minecraft but with blocks on the smallest possible scale we know how to measure, a scale on which point particles would be as big as ants while atoms would be the size of buildings. This is essentially what the researchers are talking about when considering if our universe is a simulation; countless tiny voxels moving through a mind-bendingly complex simulation governed by exotic math of a computing device of unknown power, origin, purpose, and accuracy, defining the laws of physics we can detect.

But how does one prove that we live in a simulated environment and the limits of point particle interactions don’t simply happen to fall into a voxel on their own? Doesn’t the whole idea rest on circular logic? The voxels should have an energy limit of Ψ and if the quarks and gluons that we measure have an energy limit of Ψ they are voxels? Something just does not add up here. If we try to control the state of something virtual, we have to expend a lot of energy to do it. Today, it takes a supercomputer to simulate the behaviors seen in a cube of space barely big enough to fit a few simple atoms. If we want to do even a simple byte flip, we have to conduct a current that will be converted into 0s and 1s. Even on a quantum computer we’ll need to apply a good bit of energy to keep the qubits in a state we can manipulate. So if a universe is being simulated with some sort of a hypercomputer, it requires an immense amount of energy to run, even if all the supernovae and galactic collisions are just instructions on a stack.

Who would have such energy generation capabilities and why would anyone decide to simulate the universe in such detail? Simulations are best when they focus on the specific things to model at the appropriate level of abstraction. When researchers look at virtual galaxy collisions, they don’t spend the computing power and electricity to model the position of each star because they don’t really need to know where each star moves for the purposes of seeing how galaxies affect each other. They’re concerned about the overall shape of merging galactic arms so exact details of every solar system involved would only slow the simulation down. Likewise, a simulation of an entire universe down to the detail of a point particle doesn’t seem to make much sense unless the simulation’s goal is to create something like Laplace’s Demon, which we can do with enough computing grunt but will mean little in the real world. Beyond that, we get into philosophical and abstract questions like who designed the simulation and if their universe is a simulation too. And we’ll quickly arrive at the First Cause dilemma on rather shaky grounds. Not exactly the place a scientific proposal wants to end up when taken through its implied consequences…

See: Silas R. Beane, Zohreh Davoudi, & Martin J. Savage (2012) Constraints on the universe as a numerical simulation, arXiv: 1210.1847v1

  • Dave Bush

    You are assuming that a simulation always runs a constant scale. There’s nothing to stop a simulation dynamically switching scales as necessary. If you’re modeling galaxy collisions you can keep the approximate position of all the stars known, and zoom in on on stars that get close.

    With this scheme Quantum decoherence becomes the point at which the system switches scale.

  • Greg Fish

    In this case, the quantum voxel idea is obsolete because scale of quantum phenomena will vary with distance. And considering that quantum decoherence has to be induced for objects larger than a 60 atom buckyball, I have a lot of trouble seeing it as a scale switch.

  • Gregory Propf

    There might also be something a bit more devious going on with AI. Suppose that something in the simulation software is able to tell when a sentient being is around and specifically what that being is paying attention to. Then, suppose that complex life is fairly rare. Most of the universe could be simulated on a crude scale, then when some pesky scientist tries some of the experiments you suggest everything looks nice and real to him and seems to exist on a fully detailed basis. It would be maddening of course and make you suspect that the AI gods are evil or just have built a universe as an elaborate joke.

  • Greg Fish

    That’s possible, but now we’re making so many assumptions about whoever built the simulation that the discussion becomes about whether the universe is real rather than how do we prove whether it’s a simulation or not…

  • TheMechanicalAdv

    It’s just a metaphor! Many physics formulas are obtained by replacing infinities with finite regularizations. It works, but it’s mystical and must eventually be replaced with something naturalistic. Saying “what if there’s only a simulated infinity?” is a rhetorical gambit in the debate on how to make these changes.

  • Chris Warburton

    If the Universe were a simulation, there is no reason to suppose that its software would work in ways we might think, in particular what makes it reasonable to conjecture that Universes should be modelled numerically by making tiny little boxes of space?

    For my Masters thesis I wrote numerical simulations of little boxes of space, and quickly came to the conclusion that it is an ugly way to do things. Computers are exact*, discrete machines (excepting the odd cosmic ray) which run exact*, discrete software, but numerical simulations choose to ignore this; rather than encode the exact*, discrete symbolic laws of Physics into computers, they encode inexact representation of the Real numbers. The handle is turned a few times, then we spend years trying to explain the resulting numbers using exact*, discrete formulas!

    A far more elegant way to run a simulation would be to remain exact and discrete, performing symbolic algebra with the laws of Physics themselves, rather than plugging in numbers. Unfortunately we don’t know the laws of Physics well enough to do this yet; we have existential representations like differential equations, which *describe* solutions exactly but which are not the solutions themselves. In simple cases these can be manipulated symbolically, but most of the time we have to resort to methods like numerical simulation to find approximate solutions. If we could find a more direct representation of Physical laws (probably algorithms of some kind) then simulating Universes would be easy (though still very resource-intensive!).

    I think there’s a long road ahead before we discover such things, but as for the argument that we’re living in a simulation: I think it far more likely that our Universe is emergent from an exact, discrete symbol manipulating algorithm, compared to being a blurry approximation of something else which stops after some number of decimal places.

    * When I say that software or Physical formulas are “exact” I don’t mean to imply that they always gives ‘correct’ results. I mean that when I write “E = mc^2″, the forumula itself is perfectly represented using 8 characters. If instead I plugged some numbers into it, I’d get imperfect, inexact answers. This is because Physics generally work with Real numbers, which ironically can never be represented exactly.

    PS: A voxel is not a little cube