The Case For The Experience Machine

It's like The Matrix, but better

The Experience Machine (from now on, TEM) is a famous thought experiment, created by Robert Nozick in his famous book Anarchy, State, and Utopia, often used to invoke the intuition that there is more value to life than hedonic pleasure. It goes something like this: imagine you are offered the opportunity to be strapped up to a chair and pumped with blissful-feeling chemicals that simulate the most wonderful experiences you can possibly imagine. There is a catch, however: once you go in, you cannot go out for the rest of your life. Also, during the entirety of your experience, you will think that the simulated experience is the real world.

While this thought experiment often pushes people away from ethical hedonism, it actually did (and still does) the opposite for me. In this post, I will give five arguments to show why, given the opportunity, everyone should go into the experience machine.

The Skeptical Argument:

A common response to The Experience Machine (TEM) is the idea that there might be something inherently bad about being detached from the 'real world,' such as failing to achieve moral obligations. However, there is a significant issue with this approach: our access to the world is always mediated by our brains and our own experiences.

From an individual's perspective, if they perceive something as real and it behaves as though it is real, it is the same as it being real from a motivational perspective. Consequently, the experiences within the experience machine should be as motivating as those in the 'real world.'

If morality as something objective fails to motivate us, then the Arbitrariness Problem—a concept typically applied to Divine Command Theory, which suggests there is no reason to act morally even if moral facts exist—also applies here. This gives us a good reason to question the alignment with moral obligations or other real-life duties in these experience machine situations.

The Arbitrariness Problem:

One may also reject going into TEM because they think it is bad to project superficial positive experiences into one’s brain. Perhaps they would argue that pleasureful experiences are only to be experienced in real life. On the other hand, this view doesn't seem at all consistent with peoples’ actual behavior at all, as many people watch movies that feel immersive for fun, take SSRIs (a widely used antidepressant), and much more. While one might respond with the following reasonable objection: the movie and SSRI counter-case doesn’t work because the objector would say the same thing if the movie or SSRIs that would make them blissfully happy for the rest of their life. However, this would then be an objection of the degree to which you can be made happy superficially.

While this objection seems fair, it still results in a problem. At what particular point does something go from a fun, ‘fake’ experience to the experience machine? At what degree of experience or duration does this no longer become okay? It doesn't seem like these questions have good answers, leading me to believe that the rejection is arbitrary.

Meaningfulness Objection and Rebuttal:

Another objection may be that mere pleasure cannot achieve the best value state because there are meaningful experiences that cannot be accounted in TEM (your child’s wedding, for example). However, this objection seems quite misguided.

If meaningful pleasure states can be reduced to neurotransmitters and chemicals in the brain, there is no reason to think that TEM can’t replicate that. Even if one thinks that the hedonic states can only be achieved by perceptual experiences, TEM can replicate what it would be like to have those experiences.

Meta-Preferences and Regret:

While one can argue that they would not want to go into TEM, by the very nature of the machine, one would not want to leave afterwards. Given that a rational decision is defined by making a decision with full information about all possible outcomes without regret afterwards, going into TEM is the most rational decision. On the other hand, in the case where one doesn’t go into TEM, one do not have the experiential information about the outcome when they do go into the TEM. This, of course, raises interesting questions about preferences and meta preferences, but I will not go through the implications of those here.

There is also a question about preferences in TEM because the experience machine would likely change your preferences by reengineering your brain. Given a scarce amount of resources, the machine would likely change your preferences to something that can be maximized for easily. This may end up with one’s neural reward structures favoring really basic things like mere existence or breathing as much as one previously enjoyed sex or spending time with their family.

The Reverse Experience Machine:

Adam Kolber makes an analogous thought experiment to show why our intuitions in TEM might be misguiding. The analogous situation goes like this: imagine you are going about your normal day, and a stranger walks up to you. He says that your entire life up until now has actually been in an experience machine that is far better than your life in the real world. The stranger then offers you a decision: you can either leave the simulation and go into the real world, or stay in the simulation, forgetting that this conversation ever happened.

This Reverse Experience Machine generally invoke intuitions that lean more towards staying in TEM than the first thought experiment, even though the cases seem to be approximately equivalent. The only difference seems to be whether you start in or out of the machine. Many argue that the reasoning behind the decision switch is Status Quo Bias, in which people’s preferences lean towards their current states of affairs. Conditional on the fact that one thinks that changing one’s preference because of Status Quo Bias is irrational, this analogous case seems to be a good reason to reject the initial intuition.

It should be noted that the point of status quo bias potentially goes both ways and so the argument fails because our changing intuitions don’t tell us which case more accurately reflect true preferences. This is still reason, however, to be skeptical of our intuitions in the first case — it just implies that the second case cannot be used to show in what way it may be irrational.

As always, tell me why I’m wrong!