Simulation theory

From Virtual Reality, Augmented Reality Wiki
Jump to: navigation, search


A number of different writers, technologists, and futurologists have predicted that the available computer power in the future will be enormous. Future civilizations may have the capability to create simulations with a high level of detail. These would simulate the universe and its laws, allowing for the emergence of self-conscious entities that could communicate with one another. They could be simulations of that civilization's forebears, and since they would run on very powerful computers, they could run a great many of them [1] [2]. The idea that our universe is a software process running on some deeper computational substrate is known as the Simulation Argument (or Simulation Hypothesis). Nick Bostrom has provided an argument for this thesis, and while other philosophers are taking this idea seriously, physicist suggest that there might be practical ways to find evidence that confirms it [3] [4] [5].

Simulation argument

The philosopher Nick Bostrom (University of Oxford) explored, with rigor, the issue of the simulation argument, for the first time, in a 2003 article. The type of simulations he explored are not akin to the ones in the movie Matrix, for example. In the film, the world was simulated but the conscious minds were not. On the contrary, the simulations explored by Bostrom do not have a biological component, being run on a deeper level hardware or in virtual machines inside other simulations. The argument takes into account the assumption of substrate-independence and the technological limits of computation, and a bland indifference principle [1] [4].

Assumption of substrate-independence

Substrate-independence is a common assumption in the philosophy of mind studies. It asserts that human consciousness is not dependent on a biological substrate. It could, theoretically, be replicated in silicon-based processors, provided the system has the right sort of computational structures and processes, with fine-grained detail like on the level of the individual synapses. The replication would not, necessarily, have to be perfect; just good enough that a human-like subjective experience could be generated. This assumption, although not universally, is widely accepted [1] [4] [6].

Currently, there is not enough computer power to run the computational processes required to replicate the human brain for the emergence of consciousness. Even if they were available, there is still a lack of knowledge in how to program such a thing. Besides this, there would also need to be sophisticated ways of making a very detailed scan of a human brain. These are only technical difficulties, and not physical law or material constrains. A sufficiently advanced civilization with enough computing power to create conscious minds in computer hardware would be classified as posthuman [4] [6].

Technological limits of computation

Although we are still a bit far off in the creation of conscious minds in computers, due to the limits in technological development, some have argued that if technological progress continues unhindered, those problems will be surpassed. For the simulation argument, it does not matter the timescale in which humankind will reach a posthuman stage where such a capability will be available [1].

It is not possible, presently, to set an upper bound limit to the computing power that will be available to future posthuman civilizations. Indeed, such a civilization could even have the capability to convert planets and other astronomical resources into computers with power beyond imagination. A lower bound for computation in a posthuman future is easier to establish, assuming only mechanisms that are already understood. Some authors have suggested a computer system roughly the size of a sugar cube that would perform 10^21 instructions per second, and another with the mass of a planet that could perform 10^42 operations per second [1] [6].

There are also some estimates regarding the amount of computer power needed to emulate a human mind. One gives a figure of ~10^14 operation per second for the entire brain, and another that is based on the number of synapses and their firing frequency gives a number between ~10^16 to 10^17 operations per second. Beyond these figures, the addition of an environment in a simulation will increase the computing power required. The value of the increase will depend on the scope and granularity of the simulation. To obtain a realistic simulation of human experience it is not required to simulate the universe down to the quantum level (something that could be infeasible unless a radically new physics is discovered). What is necessary is that the simulated humans interact in normal human ways with their simulated environment and do not notice any irregularities. Indeed, according to Bostrom (2003), “a posthuman simulator would have enough computing power to keep track of the detailed belief‐states in all human brains at all times. Therefore, when it saw that a human was about to make an observation of the microscopic world, it could fill in sufficient detail in the simulation in the appropriate domain on an as needed basis. Should any error occur, the director could easily edit the states of any brains that have become aware of an anomaly before it spoils the simulation. Alternatively, the director could skip back a few seconds and rerun the simulation in a way that avoids the problem.” [1]

In conclusion, the main computational cost to create completely realistic simulations seems to be in simulating organic brains down to the neuronal or sub-neuronal level. Since a posthuman civilization may build a great number of very powerful computers capable of running simulations indistinguishable from reality, the computing power available to them is enough to run a huge number of ancestor simulations [1].

The core argument

The core of the simulation argument does not try do demonstrate that reality is, in fact, a simulation. It merely shows that one of three propositions should be accepted as true [1] [3]. The general idea can be understood without mathematics, although a formal version of the argument uses probability theory [7]. Although each of the propositions may seem implausible, if the simulation argument is correct, at least one is true [6]. According to Bostrom (2003, 2006), the three propositions presented are:

1. Almost all civilizations at our level of development become extinct before becoming technologically mature.

2. The fraction of technologically mature civilizations that are interested in creating ancestor simulations is almost zero.

3. You are almost certainly living in a computer simulation. [1] [6] [7]

If the first proposition is false then a major portion of all species at our level of development will achieve the posthuman stage (technological maturity). On the contrary, if true, it does not mean that humanity will face extinction soon but that it is unlikely to reach a posthuman stage. It is also possible that a civilization could remain arrested for a long time at the current level of technological development before going extinct [1] [6]. If the second proposition is also false then it follows that a portion of these species that are technologically mature will use some of their computer resources to run ancestor simulations, recreating minds like ours and doing so in vast numbers. In order for it to be true, there needs to be a convergence among the development of advanced civilizations, in which almost none of them are interested in running computer simulation, or have wealthy individuals interested in doing that. Another possibility is that they have enforced laws that prevent individuals from acting on their desires to run simulations. Advanced posthuman civilizations could also perceive the ethical problem of running ancestor-simulations, due to the suffering that is inflicted on the inhabitants of the simulation. This could lead to the prohibition of such simulations being created. Finally, if the third proposition is true, then reality is almost certainly a computer simulation created by an advanced civilization [1] [6] [7] [8].

If the proposition that we are living in a computer simulation is true, then the universe that we are observing is a small piece of the totality of physical existence, and the physics of the universe in which the computer is running may not resemble the physics of our observable universe. Therefore, our world would not be located at the fundamental level of reality. The veracity of that proposition could also mean that it could be possible for simulated civilization to achieve the posthuman stage and run their own ancestor-simulations, on computers built in the simulated universe. These would be like “virtual machines”, which is a common concept in computer science. Also, virtual machines can be stacked, being possible to achieve several levels of simulations inside simulations. If our civilization develops ancestor-simulations, this would support the third proposition, leading to the conclusion that we would most likely be living in a simulations [1].

Living in a simulation

The veracity of the simulation argument would not necessarily entail a loss of rationality or changes in human behavior. Even in a simulation, a person should continue life the same way as before. To predict what would happen in the simulation, ordinary methods like extrapolation of past trends, scientific modeling or common sense would be used. Also, if the third proposition is correct, that would decrease the probability of the first one, avoiding extinction before technological maturity. However, computational power constraints could make it possible that the simulators would finish the simulation before our civilization reached a posthuman level [1] [6] [8].

There have been some suggestions for behavior or to predict what would happen if reality was a simulation in order to increase the probability of continuing existence or of being resimulated in the future [2] [6]. For example, if the simulator happened to be a believer of some Christian fundamentalist church, the simulation might happen to reproduce those beliefs, and the simulated beings would be rewarded or punished according to Christian moral criteria. In this way, an afterlife would be a possibility for a simulated being, in which his existence would continue in a different simulation after death or be uploaded into the simulator’s universe and given an artificial body [6]. Indeed, if the simulators intervene in their simulations and are not only passive observers, they can take aspects of gods, having power over life and death, determining and changing the laws of the simulation, and engineering anthropic fine tunings. They can end the simulation at any time or watch upon the development of the beings in the simulation [2] [8].

Living in a simulation could also mean that a person could not be sure if all the people possessed a conscious experience. While some people would be simulated with enough fine-grain detail to possess the property of consciousness, others could be simulated at a cruder level, allowing only the appearance of real people, but without a subjective experience. However, according to Bostrom (2005), “some people have argued that it is necessarily true that anybody who acts sufficiently like a normal human being must also have conscious experience.” [8]

The fact that the universe could be a simulation does not mean that the world around us does not really exist. Instead, one should perceive it has having a different nature than previously thought. Ultimately, the computer in which the simulation runs, and its electrical activity, would be physical at the basic level of reality [8].

Implications of living in a simulation

It is possible that if we are living in a simulation there would be no way to identify it. The virtual reality would seem completely real. Even so, we could never be certain that we would not be living in a virtual reality [6]. However, some researchers have suggested that simulations may have limits – that even posthuman simulators with advanced knowledge of the laws of nature would still not have a complete knowledge of them. These flaws would be subtle but could result in glitches in the simulation. Another possibility is that the simulators would try to fix these flaws by patching the virtual reality. These updates could result in changes to the laws of nature, over time. Living in a simulated reality would mean that occasional glitches would occur, along with small drifts in the constants and laws of Nature [2] [4].

A study published in the journal ArXiv, in 2012, suggests that there is always the possibility for the simulated to discover the simulators, and offers the prediction that there might be limitations on cosmic ray energy levels if reality is indeed a simulation. Furthermore, it predicts that the reason for the posthuman civilization to run simulations is to test out string theory. Detailed simulations could allow for future researchers to test hypotheses about the universe and disprove a number of possible different versions of string theory [4] [5]. This would give credence to the suggestion that the posthumans have an incomplete knowledge about the laws of physics, and therefore it is expected that there would be gaps and flaws in their simulations [2] [4].

It is expectable for the simulators to be economical and practical in their simulated realities if they were for entertainment, for example. They could avoid the complexity of using a consistent set of laws of Nature, patching instead “realistic” effects. These could cause some problems and be identifiable from within the simulation. Another cause for sudden glitches in the simulated realities could be the use of error-correcting codes. This is a technique that has been effective in the simulation of complex systems. These codes would correct mistakes in the simulation much like the error correcting system that exists in DNA. If the genetic system did not have a correcting mechanism, it would eventually be corrupted by the build-up of mutations. The computer equivalent of this system also guards against error accumulation. The use of error-correction codes could lead, once in a while, for a correction to take place, leading to sudden changes that would appear to contravene the laws of nature present in the simulation. Finally, the simulations would have a similar level of computational complexity, in which “the simulated creatures should have a similar complexity to the most complex simulated non-living structures.” [2]

There is also the question of how can we trust the observations made about reality if it is a simulation. The simulation argument relies on the assumption of the technological capabilities of a posthuman civilization, and the evidence for that assumption is empirical, based on the current best theories about the physical limits of computation. Ultimately, these observations of the world around us could be misleading, providing data about the simulated reality and not about the underlying reality in which the simulation is running. However, according to Bostrom, “the claim that we cannot have any information about the underlying reality if we are in a simulation is false.” He provides two conditional claims that can be know if reality is a simulation. These are adapted from the Frequently Asked Questions (FAQ) page available in the website [9]:

1. If we are in a simulation, then the underlying reality is such as to permit simulations, it contains at least one such simulation, and (3) (the third proposition of the argument) is true.

2. If we are not in a simulation, then the empirical evidence noted in the simulation argument is veridical taken at face value, suggesting that a technologically mature civilization would have the ability to create vast number of simulations; and consequently, by the simulation argument, there is a very high probability at least one of the disjuncts in (1)-(3) is true [9].

A technological mature civilization could run a vast number of simulations since it is expected that it would have enough computer power available. If our reality is a simulation then there are probably a myriad of other simulations, differing in some detail or overall design. These may run sequentially or simultaneously. A simulation can have a civilization that reaches the posthuman stage and proceeds to build its own virtual realities. In this way, reality could have many levels, with simulations within simulations running on virtual computers. The number of layers of simulation would be dependent on the computer power available at the base-level computer (which is not simulated) [1] [6] [8].

The posthuman simulators can, in some ways, be considered to be like gods to the beings inhabiting the simulation: 1) they created the virtual world; 2) they have a superior intellect; 3) they can interfere in the everyday world, violating its physical laws (omnipotent); and they can monitor everything that happens (omniscient) [1]. Indeed, the simulation argument has theological implications, and some have considered it as an interesting argument for the existence of God [3].

If it is plausible that there can be simulations within simulation then in each deeper level the physical power of the computers increase, as well as the intelligences of their respective civilizations. If there is an infinitely deep computer then God could be viewed as functionally equivalent to that infinitely self-programming computer, being pure hardware and not software running in another simulation. The designer of the universe would not be God, in this case, but the deeper level civilization. God - being the base hardware in which all other simulation ran - could be considered the ground of being. It would support an infinite hierarchy of simulators that would produce their own virtual realities. An afterlife would also be viable in this hierarchical system of simulations. Simulated beings could be resurrected many times, moving to deeper level simulations and getting closer to the base-level world in which the base computer runs (something analogous to moving closer to God) [3].

Finally, other thinkers have explored the question of natural evil in a world created by a benevolent creator god, in view of the simulation argument. According to Johnson (2011), “if we grant the theist the knowledge of God’s existence that they claim, the problem of natural evil forces the theist to choose between rejecting “God designed our universe” and “natural disasters are evil.” We have much more reason to reject the former than the latter and, if Bostrom is right, our being in a computer simulation is the best non-divine explanation for our universe’s design.” [10]


  1. 1.00 1.01 1.02 1.03 1.04 1.05 1.06 1.07 1.08 1.09 1.10 1.11 1.12 1.13 1.14 Bostrom, N. (2003). Are you living in a computer simulation? Philosophical Quarterly, 53(211): 243-255
  2. 2.0 2.1 2.2 2.3 2.4 2.5 Barrow, J. D. (2007). Living in a simulated universe. In Universe or Multiverse? Cambridge University Press, pp. 481-486
  3. 3.0 3.1 3.2 3.3 Steinhart, E. (2010). Theological Implications of the Simulation Argument. Ars Disputandi, 10(1): 23-37
  4. 4.0 4.1 4.2 4.3 4.4 4.5 4.6 Jones, A. Z. (2015). Are we living in a computer simulation? Retrieved from
  5. 5.0 5.1 Beane, S. R., Davoudi, Z. and Savage, M. J. (2012). Constraints on the universe as a numerical simulation. arXiv:1210.1847v2 [hep-ph]
  6. 6.00 6.01 6.02 6.03 6.04 6.05 6.06 6.07 6.08 6.09 6.10 6.11 Bostrom, N. (2003). The simulation argument: why the probability that you are living in a matrix is quite high. Retrieved from
  7. 7.0 7.1 7.2 Bostrom, N. (2006). Do we live in a computer simulation? New Scientist, 192(2579): 38-39
  8. 8.0 8.1 8.2 8.3 8.4 8.5 Bostrom, N. (2005). Why make a matrix? And why you might be in one. Retrieved from
  9. 9.0 9.1 Bostrom, N. (2008). The simulation argument FAQ. Retrieved from
  10. Johnson, D. K. (2011). Natural evil and the simulation hypothesis. Philo, Vol. 14(2)