Welcome, New Readers

A letter of introduction to everyone who recently signed up

Recently I wrote a thing that made claims about brains and, in particular, beliefs about ‘good’ and feelings of ‘good’. Some people said ‘this stuff about serotonin is totally wrong’. Scott Alexander chimed in with a post of his own saying, “the neuroscience is rightfully being trashed”, but ended with ‘maybe the ideas here are salvageable.’

This experience felt, to me like having the hottest guy in school, who is also the smartest and most intellectually humble, but also really funny and generally kind, and when he gets angry you feel it right along with him, tell everyone else in school that I exist and said a thing which is obviously wrong on the surface but possibly got right underneath.

This post is, simple a, hey, hi! to everyone new who subscribed. And thank you, Scott Alexander, primarily for being an excellent example but also for linking to something I wrote that was only partially correct.

With all the confidence of the guy at the mic at Q&A, saying, “this isn’t really a question, uh, it’s more of a comment, uh,” I’ve got more ideas about brains and computers and religions and distributed computing and politics and consensus algorithms, to share with you here on this substack.

Sorry, folks. I just can’t help myself. This post is merely a description of who i am and where i’m coming from.

A reckless amount of energy propels me in a direction that seems correct, but which often gets me hurt. Every time I fall down and get hurt, I learn something new, which leads to me getting hurt in newer more elaborate ways. I have learned to trust myself - not to avoid falling, but merely to keep getting back up.

Me, in college, shortly before dislocating my shoulder for the first time. Also pictured: Seth Flaxman now at Oxford, first author of “Estimating the effects of non-pharmaceutical interventions on COVID-19 in Europe”, and Steven Gomez, AI alignment researcher now at MIT. I really wanted to study P vs NP but had the impression there was no funding for this, so I went about trying to get ‘free market tenure’ via FU money, still a work in progress.

Everything i write here on the substack should come with this OFFICIAL DISCLAIMER:

OFFICIAL DISCLAIMER, aka 'epistemic status’ for basically everything I write

I’m not a professional researcher.

I’ve never cut up a brain or done measurements on neurotransmitters or even worn a real white lab coat, like the guy I worked with at Google named ‘Guy’ who always wore a white labcoat and was snarky about other people’s bad ideas, just like a real scientist. Hi, Guy!

I haven’t done a literature search since grad school, which I bailed out of when the question of why I wanted a Ph.D. didn’t have a compelling answer.

Shortly after, a job offer from an electronic trading company made it clear that I really liked building things and making money more than I liked writing papers or figuring out how to get a government agency to give me money.

I spent a while tooling around San Francisco and Silicon valley and doing lots of drugs in a search for spiritual enlightenment. I may be the only person ever to have spent multiple days hanging out with both homeless people and Uber CEO Travis Kalanick.

After a few years in startups, struggling with drug addiction, pyschosis and debt, I managed to get my life together enough to convince a former girlfriend to marry me. I settled down a bit, stopped using marijuana, started holding regular jobs, including a gig at Facebook where I became known as “Q&A stoner guy” because of the questions I’d ask Mark Zuckerberg. “You have to smoke a bowl to understand him,” people said.

I no longer feel any need to use drugs because I’ve learned how to reliably trigger intense psychedelic trips by means of breathing techniques that don’t leave me intoxicated. I spent years working on datacenter reliability at Google. Seeing how totally unreliable computer hardware really is. I’ve also been watching my Dad progress through the symptoms of something like Parkinsons.

Seeing how resilient my dad is in comparison to a rack of machines in poorly maintained datacenter, and seeing how empires keep failing after losing the mandate of heaven, convinced me AGI will likely act like more of a buddha

¹

than moloch.

Four Generations of Neyer men; My Grandfather atop, my dad to my left, my son on my lap

So, no I’m not a researcher. I’m simply a guy who used to have lots of problems, and fixed many of them. I caused, and solved, the problems with a brain that I use quite often, and I’m always trying to make better use of it.

Making better use of my brain requires an operational theory about how my brain works.

I want to know what causes what, so I can know which buttons I can press. I want to eat and exercise and stretch and breath and speak and write and think just so, in order to press the buttons in my brain that will make me calmer without putting me in a stupor, kinder without making me a pushover, more patient without dulling my desire to act, more loving without blunting my judgement, more honest without making me overshare, more humble without being self-deprecating, more self accepting without losing the drive to improve, more at peace in the midst of conflict &c &c.

Given this goal, I haven’t yet found a theory about how brains work that is good enough. I don’t think my goal is unreasonable, since there’s no destination involved; i’m simply trying to move, more or less consistently, in a better direction. I get the impression this is what every living organism everywhere is doing, but most of them don’t have the benefits of written language.

Most organisms cannot communicate with a symposium of their ancestors. Most primates cannot learn from those who came thousands of years before. Most humans throughout history could not learn from people all over the world. Most educated adults in the 21st century don’t have access to a community like this one, full of opinionated and educated people who can follow along numerous streams of empirical research and willingly synthesize them with stories that are thousands of years old.

When a prophet makes a statement in the name of the Lord, if what he says does not take place and his words do not come true, then his word is not the word of the Lord: the words of the prophet were said in the pride of his heart, and you are to have no fear of him.

Deuteronomy 18:22, and the basic credo of the rationalist community.

I’m trying to synthesize science and religion for the ultimate purpose of self-authorship; I want to modify who I am to consistently make myself a better person.

I can’t do this on my own, not without the labcoat and the electrodes and the measurements and the machines and reams and reams of data. It’s hard for me to accept this, but I accept it, just like i’ve accepted that i’ll never solve P vs NP, just like I’ve learned to try and accept the present moment as it is rather than as i wish it would be.

What I have found is many heuristics which do work well. I’ve also found a lot of correspondences between these heuristics, and stories I read about ‘how brains work’ written by the class of specialists who seem to be dedicated to this kind of thing.

The heuristics I’ve found that help me better navigate the world are all compatible with the predictive processing model, as I understand it. The heuristics are also compatible with an understanding of my brain as being a machine learning model that tries to compute something colloquially called ‘good’ and move me in a ‘good’ direction.

The heuristics I’ve found are also quite similar to heuristics i’ve learned for writing clean code or managing complexity in distributed software systems, during my time at big tech companies like Facebook, Google and Snap.

These heuristics also seem to line up with what I’ve found in the many different religions I’ve studied.

I’ve ransacked temples, like a barbarian for knowledge, reading every holy scroll and sacred book I can find. I make a point to study the teachings of anyone who seems at peace with themselves and capable of sustained accomplishment. I’ve started with the same axiom that underlies all of science. Namely, I assume that all of these people performing experiments and reporting what happened are describing the same reality.

I’m trying to do all this because I just want to be a great dad, a great husband, a great son, a great friend, etc. etc.

So I have this brain in my skull - at least i think I do, I’ve never really encountered direct evidence that it exists, and the idea that I have a brain is more of a faith thing, but anyhow - I have a brain and I try to use it as well as I can. I’m following something like an amateur version of the scientific method. I develop hypotheses by combining as many possible observations as I can, both my own and those from sources I trust. I compress these observations into a theory which generates predictions about what will result from various things I might try. I try these theories on to see how they help me operate, by pretending they are true and seeing what happens.

For example, in one experiment, I tried ‘rolling my awareness’ - changing the spatial focal point of my attention - in a spiral around my spinal column. I did this, after reading about how tornados work and seeing something that sounded vaguely like a description of the kundalini serpent rising up. The result of this experiment was a month or so of vertigo: i’d feel really dizzy whenever I had my head at any angle other than straight ahead. It’s hard to tease out causality here, because this also happened shortly after we had some new cabinets installed and maybe they were off gassing - but the vertigo came a few days after the cabinets, right after that “tornado of consciousness” experiment.

So, yeah, a lot of these experiments tell me ‘nothing interesting’, and sometimes ‘don’t try that again.’ I’ve also learned how to do things like tickle myself via imaginary tendrils coming out of my head, relax the structure superimposed on my visual field by my beliefs, and self-induce psychedelic states via Wim Hoff breathing.

The point is, i’ve tried a lot of things. Developed and dropped a lot of theories. No, I don’t have the publications to prove it. All I have is this outrageously charismatic personality, a record of several personal demons bested, and a burning desire to help as may people as possible believe they can do this kind of thing, too.

When I can tell that a pattern of behavior is helping me function better, I try to ask myself what it’s doing from all the different perspectives that i’m merging: religion (which i think i understand), computer science (which I understand well), neuroscience (which i consume via infotainment and the rationalist community, so who knows how good my understanding is there) and history (which i get from podcasts and other books, so i don’t really know how good my understanding is there, either).

A Brief Digression to talk about uncontroversial subjects like Evolution and Global Climate Change

My beliefs have iterated many times over the past few years primarily by means of a mechanism where I come up with a partially formulated idea which is a mix of science-inspired philosophy and pure woo, share it with people, who accept some parts and poke holes in other parts.

Over time what’s happening is that less and less of what i’m believing is ‘woo’ and more and more of it is directly actionable or seems to map directly on to what ‘most science people’ seem to agree on. I’m phrasing it this way on purpose because I think I have a brain in my head for the same reason I believe in global warming and that vaccines work - I’m trusting a huge body of people all telling themselves similar stories, some of whom are accomplishing amazing things.

I hold on to the parts of the stories that nobody rejects. For example, In the above case where i shared a theory about dopamine and serotonin, nobody had a problem with the claims I made about dopamine, which makes me think my understanding of dopamine is correct. Of course I could be totally wrong but I think this means that large quantities of infotainment got it wrong (entirely believable) and that a bunch of opinionated bloggers who like to raise objections whenever they can didn’t raise an objection here also got it wrong, which strikes me as less likely.

It’s like I’m writing an operator’s manual from a car i am driving, by cobbling together as many different operators manuals as I can find, plus treatises on what phlogiston is, a story of someone dissecting a spark plug, and experiments like seeing how many donuts I can turn in the spiritual equivalent of an empty parking lot until the fuel tank runs out.

Thanks for subscribing, thanks for reading, thanks for listening, thanks for thinking, and most of all, thanks for telling me where I’m wrong. I could always use more predators in my intellectual ecosystem, because I come from a line of hyperfecund meganerds, I have a million hypothesis and ideas, and I need outsiders to help me poke holes in the best things i can produce.

The greatest give you can give me is to tell me precisely where i am wrong.

1. Also worth considering is the concept of ‘Omega Point’ by Teilhard Chardan