Major World Religions and Meditation Explained in Terms of a Common Neuroscientific Model

All in seven only-slightly-oversized paragraphs

Consciousness is a workspace where different subsystems in the brain communicate. When different systems in your brain conflict with each other, consciousness is where those conflicts get resolved. The feeling of “having a self” is a proxy for both the exertion of energy, and whatever drive or emotion is loudest in the brain at that time. A belief that says “I ought to choose A over B” is amplifying one voice over another in the workspace of consciousness. That amplification has an associated energy cost, which feels like ‘exerting your will’. The energy cost of amplifying one signal is why sometimes making choices feels difficult. The “same” choice becomes far easier when the brain is quiet. When multiple drives in the brain are in conflict, it’s harder for us to make the right choice.

Willpower experiments fail to replicate probably due to problems with their setup; making the same choice to pass up a cookie 500 times in rapid succession is far easier than making the same choice 10 times at different moments over the course of a month. After the 400th rejection, “I will not have this cookie” is the loudest signal in your brain, so it becomes trivial to pass it up again. If researchers offered subjects a cookie at ten random times over a month-long interval, whatever was going on in the subjects brains would be wildly different in those 10 times, increasing theirs odds of cookie consumption. To truly test the depletion hypothesis, experimenters should try creating additional emotional difficulty, like, say, murdering the friends of people who have rejected the cookie 300 times, to see what happens, under the hypothesis that murdering a person’s friend will increase the noise in their brain and therefore make it harder for them to continue avoiding eating the cookie.

When we feel bad, we generally perform worse, i.e. make choices that lower our net long-term valence. Often we make those choices because they increase short term valence. There is a feedback loop in play here: bad choices make us feel worse over the long term, and thus we are more likely to make even more bad choices. “Do more of what makes you feel good and less of what makes you feel bad” is obvious advice that’s hard to implement because many things feel good in the moment but, on net, make us feel worse. How can we satisfy all the drives that pull us in various ways?

We can’t. Nothing we do will ever satisfy all the drives (the first noble truth) but we can subsume those drives under a higher level goal, by becoming mesa-optimizers, and consciously selecting an abstract concept to optimize for. If optimizing for an abstract objective generates behaviors that map roughly onto effective strategies for satisfying the various drives, your desires will stop screaming at you in all caps. Your emotions will begin to only use lowercase letters. The conflicts between your drives will dissipate. You will feel calmer. Your drives will settle down enough that good choices become easier for you. When your drives aren’t all inflamed, long-term positive-valence choices will be the loudest internal signals more often, and when they aren’t, it will take less internal signal-boosting (i.e. willpower) to make them so. With the right mesa-objective, good choices in the routine situations we encounter become solved problems. When we consciously optimize for a carefully selected concept, we feel less anxious and afraid. Trapped priors get released, and we increasingly anticipate positive-valence experiences. Because we act more coherently in alignment with our long term expected valence, we feel better. Translated into language that made sense a thousand years ago, “mesa-optimizing for a concept that reasonably subsumes innate drives” starts to look a lot like many different things that people from Christianized nations would call ‘religions’. In the words of David Foster Wallace, “Everyone is a mesa-optimizer, the only choice you get is what to mesa-optimize for.”

So what should you mesa-optimize for?

Greek philosophers said “there is only one thing that is the most important, which is goodness itself. Mesa-optimize for goodness itself,” which is either a delicious tautology, or a claim that somehow there’s a reality to valence. Monotheisms personify this notion of goodness (i.e. positive valence) and say “This thing called good is actually a person, responsible for creating the entire operating environment. Mesa-optimize for having the right attitude towards that person.” Eastern religions go the opposite route and try to encourage you to cultivate (i.e. mesa-optimize for) presence and patient awareness directly.

In other words, monotheisms claim the best top-down predictive model is this:

Whereas eastern religions, and the meditative frameworks inspired by them, claim it’s this:

Either way, these seemingly different approaches both point in the direction of reducing conflicts in the content of consciousness, either by directly saying, “hey, meta-optimize for this impossible-to-compute function!” or else saying “hey, don’t try to do anything, there isn’t even a you, there is only calm”.

Cultivating peace and presence directly has a similar end effect as trying to do good in the world - it moderates imbalances between drives, quieting conflicting drives so that you can consistently do what makes you feel good. When you feel calm and happy, what makes you feel good generally will be doing things that are constructive and pro-social. A key difference is that eastern religions mesa-optimize only for an internal state (peace) rather than for changes in the external environment. I find this technique, of having only internal goals, to be remarkably effective, even though I still want to make changes in the outside world. I find that aggressively restricting my locus of attempted control to the space between my ears, and, in particular, to “ensuring what am I adding to consciousness at this moment is positive”, tends to make me more effective in the world as a whole.

So are all these religions nothing more than tricks for getting the brain to settle down? I don’t know. What I've been doing is trying to make all this stuff fit together into a coherent package, and one key question for me was, “Is there a best thing to mesa-optimize for? Is there an external reality to valence, or is it just one strategy among many for managing my emotions? ” For a long time I suspected the latter, but now I lean towards the former, for reasons which are TOO LONG to fit in this post.