The simulation hypothesi.., p.24
The Simulation Hypothesis, page 24
It’s not just the mystics but many scientists who also believe these techniques have value. Carl Jung, for one, believed that there were psychic entities that existed independently of our physical consciousness and that communication with them was possible. Thomas Campbell, the physicist whose physical experiment we will explore in just a moment, also participated in experiments at the Monroe Institute, which pioneered OBE research, and his belief in the simulation hypothesis is supported by his research into these areas.
As for skeptics who object on the basis of consciousness (whether it’s the universe being conscious or individual players in the universe), this can also be dealt with in the simulation hypothesis. If you’ve diligently read through this book, you’ll notice that the video game metaphor is used precisely to account for this idea: that we are players outside of the game (and thus conscious entities) playing characters inside the video game.
So, technically, I don’t believe this is an argument against the simulation hypothesis but an objection to the idea that everyone is an AI and not conscious beings. If we extend the idea of downloading consciousness as something that happens at the moment of birth (which does require the soul be codified into some type of information) and then uploading is what happens upon death, then this objection falls away. In fact, as we’ll see in some of the experiments, the requirement for consciousness actually may support the simulation hypothesis, since that consciousness needs to exist independent of the physical entities we see around us.
The Skeptics: The Resource Argument
While even advocates of the simulation hypothesis admit there is some probability (less than 100 percent) that we are in a simulation, there are those who argue that certain experiments and arguments can show us that there is evidence we are not in a simulation.
It turns out that these are more arguments than evidence. These skeptics’ basic arguments are based on the computing power required to simulate the universe and usually based on our own existing ideas of computing. These skeptics purport to show, either through experimentation or estimation, that it would be impossible to simulate a universe as complex as ours using any kind of computing that we know of. The resources required would be the same as the universe, meaning that a full-scale simulation of the universe is, at least practically speaking, impossible.
For example, a team from Oxford University, Zohar Ringel and Dmitry Kovrizhin, were running simulations on the Hall effect, which deals with particles that have magnetic fields. (The important point here is not so much about the Hall effect itself but on the computing power required to simulate it—or, by extension, to simulate other quantum particles or processes). In a paper published in the journal Science Advances, they reported finding that to simulate these particles, the amount of memory and information grows exponentially with the number of particles, making it impractical, at least for today’s computers, to really simulate a large number of particles. Andrew Masterson, editor of Cosmos magazine, reported that even simulating a few hundred particles would require more atoms than were available in the universe today, using today’s understanding of computer science, and thus “fears that we might be unknowingly living in some vast version of The Matrix can now be put to rest.”70
Rudy Rucker, a computer science professor and well-known science fiction writer, makes a similar point—that to simulate a virtual Earth, it would require the same number of particles that are used to make up the real Earth. If this is the case, why bother? So, he, too, believes that the simulation argument can be put to rest.
But can it really?
The argument that “it would take more atoms to simulate the universe” than exist today is reminiscent (though a more sophisticated version) of the argument made by royal astronomers in the early 20th century: that we could never send a ship to the moon because the amount of fuel that would be required would make the rocket ship too heavy to leave Earth’s orbit. Obviously, this statement was made before much of our current knowledge of rocketry and space travel was developed (and proved false), though the Apollo program did require the building of the largest rocket to date, the Saturn V.
Others made similar arguments about the impossibility of 3D worlds inside video games before 3D modeling, textures, and conditional rendering techniques were refined. Proponents of this argument are assuming, of course, that there aren’t optimization algorithms that make computing more efficient. But the whole point of putting computer science front and center in this book is that video games are optimized to render only that which is needed. The rest of the “world” exists in some informational form, so computing resources aren’t expended unless and until they are needed.
In his summary of the Oxford team’s findings in Science Advances, Zohar Ringel, the lead author of the paper himself says: “Who knows what are the computing capabilities of whatever simulates us?”71
In fact, there may be evidence that computation models used in simulating our universe utilize a quantum computing paradigm, which adds a whole new dimension to information science and how much information can be stored at a single time.
Evidence of Conditional Rendering
Clearly, computational optimization—which means finding ways to use fewer resources, whether memory or processing power—is a key part of any mystery that could unlock both the veracity of the simulation hypothesis and how we might implement a Matrix-like simulation ourselves.
A number of physicists have proposed that finding evidence of conditional rendering—i.e., drawing only what the player sees—is a key way to prove that we are inside a video game. In short, these are more advanced versions of the delayed-choice experiment proposed by Wheeler, which build on the topic of quantum indeterminacy referenced in Chapters 6 and 7. If we can show that it requires someone or something to observe the results in order to “render the results,” this by itself can be evidence of computation. In fact, quantum indeterminacy may only make sense if we are part of a video game–like simulation and there is someone or something playing the game and watching the results!
In fact, several experiments that rely on proving that observation is the key to determining the collapse of the quantum probability wave have been conducted, and they suggest that the universe is somehow optimizing resources similarly to the way video games optimize computing resources by rendering only what is necessary at that point in time.
One of the most interesting of these experiments was conducted by the Italian Space Agency’s Matera Laser Ranging Observatory (MLRO). The MLRO actually conducted a version of the Wheeler’s delayed-choice experiment (described in more detailed in Chapter 7), in which a photon had to travel thousands of miles to a satellite and back after it had gone through the initial slits. The result was consistent with Wheeler’s original conclusion that the observation of a particle, even if it happened clearly in the future (in this case, the time it took the particle to travel thousands of miles), actually influenced the choice of what the particle did in the past.
Meanwhile, NASA physicist Tom Campbell and Caltech physicists Houman Owhadi, Joe Sauvageau, along with David Watkinson raised money via a Kickstarter crowdfunding campaign to conduct and document experiments to test the simulation hypothesis. In 2017, they published a paper in which they stated that the simulation hypothesis could certainly be tested. They speculated that they could show that a simulated universe was a system that would “… as in a video game, render content (reality) only at the moment the information becomes available for observation by a player (and not at the moment of detection by a machine).” 72
Campbell and his colleagues proposed experiments that are related to the particle-wave duality—involving a quantum “eraser” or a quantum “inserter” after the particle has gone through the double slits. These experiments share similarities with other experiments based on Wheeler’s original delayed-choice experiment. The point of Campbell’s experiments is to show that consciousness is fundamental to quantum theory, and by doing so, also show evidence of his version of the simulation hypothesis—that we are all players in a type of computation-based world. At the time of this writing the experiments had not yet been concluded.
Still, multiple experiments have confirmed the need for an observer as a fundamental part of quantum mechanics. While we can debate whether the observer has consciousness or not, the findings of quantum physics thus far are pretty conclusive. This by itself is evidence of conditional rendering—the video game equivalent of quantum indeterminacy. As we discussed in Part II, this kind of unexplainable behavior seems to make sense only if the simulation hypothesis is true.
Experiments for Evidence of Pixels
There is another class of experiments that relies less on the observation and conditional rendering argument of quantum physics. In this class, the idea is to detect directly the artifacts of a computer simulation in a 3D world around us. If we are, in fact, in a world consisting of pixels of information, then perhaps we can look at some of the characteristics of pixels (at least as we know them) in our understanding of computing and computer graphics, and see if some of those characteristics are exhibited in the so-called physical world around us. I call this class of experiments “looking for artifacts” of the simulation.
In their 2012 paper Constraints on the Universe as a Numerical Simulation, Silas Beane and his colleagues at the University of Bonn in Germany, Zohreh Davoudi and Martin J. Savage, argue that the structure to look for is a “lattice” that would represent the 3D pixels in a 3D simulated reality and watch how these lattices change over time. In a simulation, Beane argues, time would go in discrete steps (the idea of clock speed, which we discussed in Part II), while in most physics today it is still based on continuous equations.73
It turns out that there are certain types of cosmic rays that display this characteristic of a small lattice that has a “cutoff,” where nothing can be smaller than the cutoff. This is called the Greisen–Zatsepin–Kuzmin or GZK limit. While the GZK limit is well known to high-energy particle physicists, the source of the cutoff is not. Beane and company theorize that it must be revealing the “geometry,” if you will, of the simulation, the constraints imposed by the “lattice spacing.”
To put it even more simply, lattice spacing would reveal the arrangement of the pixels in the real world. Moreover, by studying the angular distribution of the smallest lattices, they found that energy flowed from one lattice to the next in certain directions, a kind of geometry that is reminiscent of adjacent pixels in a rendered image/screen.
Evidence of Computation: Error-Correcting Codes
Discovering the pixels of the simulation would of course be direct evidence, but it may not be the only type of evidence. Evidence of computation in the physical universe might be enough to prove the simulation hypothesis.
James Gates, a professor of physics at the University of Maryland, has studied string theory and supersymmetry. He claims that he has found what would be the equivalent of “error-checking codes” in the physical universe, suggesting strongly that the universe may be generated by some sort of computer.
In 2016, Gates released a video describing his findings of checksums in physical properties of superstrings and supersymmetry. “How could we discover whether we live inside a Matrix? asks a synopsis of his video. One answer might be ‘Try to detect the presence of codes in the laws that describe physics.’ And this is precisely what [Gates] has done.”
The synopsis goes on to explain that, “Specifically, within the equations of supersymmetry, he has found, quite unexpectedly, what are called ‘doubly-even self-dual linear binary error-correcting block codes.’ That’s a long-winded label for codes that are commonly used to remove errors in computer transmissions, for example to correct errors in a sequence of bits representing text that has been sent across a wire.” 74
In the video, Gates himself explains: “This unsuspected connection suggests that these codes may be ubiquitous in nature and could even be embedded in the essence of reality. If this is the case, we might have something in common with The Matrix science fiction films, which depict a world where everything human beings experience is the product of a virtual-reality-generating computer network.”75
Gates doesn’t detail the actual codes he has found. But this approach, of finding evidence of computation may be a very fruitful path for looking for evidence of the simulation hypothesis.
Quantum Computers, Error Codes, and Quantum Entanglement
Thus far, we have focused primarily on standard computing techniques, which use digital bits to represent information, namely computer models and pixels within scenes.
One of the arguments made by skeptics is that to simulate quantum processes would require more computational power than we have available in the universe. This is of course, assuming traditional computer bits.
However, if quantum indeterminacy is a fundamental part of nature, then it would make sense that whatever algorithms or processing units are being used to process or render the world would take this into account. Just as we created GPUs in order to help us render more sophisticated video games, we can assume that any civilization that moves toward the simulation point and wanted to run ancestor simulations would create processors that were optimized for the task at hand.
We stand today on the threshold of a new type of computing—quantum computing, which represents perhaps the first really radical shift in actual computing since the dawn of computer science in the 1950s and 1960s. Quantum computers, first proposed by physicist Richard Feynman, who suggested that if we were going to simulate the physical universe, we should use quantum particles and not classical computers. While the idea seemed like science fiction initially, the technology is rapidly progressing, with several early working models available even now.
At the heart of quantum computers are the principles of quantum physics—in particular, quantum indeterminacy and superposition. Traditional computers work on digital technology, expressed as bits, a single piece of information that is designated either a 0 or a 1. If you look under the hood of a digital computer, there are transistors and Boolean logic gates—AND, NOT, OR, gates—which are the basic building blocks of computer processors. These gates are set up to take in a single value of 0 or 1, a single bit, and to return their corresponding operator.
Quantum computers, on the other hand, rely on superposition, a concept we explored in Part II, which states that a particle can be in one of several states, or all of the states simultaneously, until and unless we observe the particle.
Similarly, qubits, or quantum bits, are superpositions of bits—they can have both values—0 and 1 (just like Schrödinger’s cat is both alive and dead)—until the bit is observed. Qubits are superposed bits, and a string of qubits can take on every single value of each qubit, giving it all possible values simultaneously. This opens up quantum computers to solving problems more quickly, faster than the serial processing of many processors in traditional computers.
For quantum computers to implement qubits, they need a way to represent the superposition in a physical way. It turns out that we already have particles in nature that exhibit quantum properties: photons, electrons, etc. It is a matter of constructing a machine or device that can rely on and measure these quantum properties of existing particles.
Quantum computers bring us full circle to the idea of a sophisticated simulation that mimics reality. Since quantum indeterminacy is a feature of our physical reality, if we are inside a simulation, then the simulation must have a mechanism to process both uncertainty, probability waves, and a way to force the collapse based on observation.
Quantum computing actually shows us that what we think of as fundamental particles are actually information—and the physical universe is thus a quantum computer! Even if the simulation required exactly the same number of particles that are in the physical world, the fact that quantum computers now exist gives us a clue that the physical universe is most likely a super-sophisticated quantum computer.
Quantum Error Correction and Quantum Entanglement
One of the mysteries of physics over the last century has been how to square Einstein’s theory of general relativity, which gives an explanation for gravity, and quantum mechanics, which tells us about the behavior of subatomic particles. No evidence had been found to unite the two—until recently. The overlap may come in the field of quantum computing.
One of the issues that early experimenters with quantum computers found was that qubits are notorious for “flipping” their value. This is a physical property of the substances used to represent the qubits.
One of the innovations in quantum computing that has made quantum computers possible has been error-correcting code. It turns out that the simplest way to do error correction on sets of qubits is to rely on another mysterious, unexplainable property of quantum physics—quantum entanglement.
Quantum entanglement—which we mentioned in Chapter 8—describes how two or more particles correspond to one another no matter where they are located in the physical world. Just as it’s possible to entangle particles, it is also possible to entangle qubits.
Knowing that several particles (or qubits) are entangled, it’s possible to come up with error-correction code such that if one qubit flips unexpectedly, this can be found out and reversed. Without going too far into the computer science or the physics, if evidence for error-correcting codes is found in models of the universe, it becomes ever more likely that the universe is some kind of simulation running on a computer.
