This is a linkpost for https://radimentary.wordpress.com/2018/01/26/magic-brain-juice/
We are agents who cannot simply act because every action is accompanied by self-modification.
Just want to say; this is one hell of a line.
I'd like it if more people understood this deeply. A lot of confusion seems to arise because people hold on to the (possibly unintentionally) dualistic notion that there is a core self or soul that is unchanging and independent of the rest of the process.
I circumambulated the idea of meta-processes with the wonderfully inscrutable SquirrelInHell
I realize the word "circumambulate" isn't doing zero work in this sentence, but is the value of "circumambulate" over a more common word like "explore" high enough to have it start making it's way into common jargon?
(I liked the initial circumambulate post, but then you and Squirrel started using it fairly regularly and I'm not sure it's pulling it's weight as new-jargon-to-keep-track-of. I am pro "give nuanced concepts nuanced words that prevent you from thinking you understand them", but still think each new term that get's commonly used on LW is a barrier-to-entry-cost that should be weighed against that)
Yeah, I hear you. My current strategy is just innovate and let the market decide if things are important enough to catch on, I'll probably drop it in a few weeks if it's just the two of us still using it. Curious if there's a clearly better plan?
See, this is exactly the sort of thing I was talking about. If Raemon hadn’t posted this comment, I would have no idea that this even is a specific piece of jargon. I read it, and assumed that ‘circumambulated’ was just a rather quirky way of saying ‘discussed’.
I don’t even really have an opinion on whether this is a good (i.e., worthwhile) piece of jargon or not! But it would be quite helpful if, at least, there were a norm of linking to something at least somewhat relevant, when using a piece of jargon that isn’t definitely very widely known among your readers—if for no other reason than to indicate that it is a piece of jargon (i.e., a word which you are using in a domain-specific way, or that has a domain-specific meaning).
Ok, fair enough. So why are diet and exercise programs so hard to stick to medium to long term? Maybe it's just me, but I've done all that you say. I've worked out the holy Meal Plan that will bring my weight down far enough to meet my goals. I've worked out exactly which sets to do and which lifts are optimal.
And I do them. For a week or three. And something happens and I stop doing them. It feels like the failure point is not the first time going to the gym or the second, it's somewhere around the tenth time.
People vary considerably in which habits stick, which habits don't, and how much work any specific habit takes. To the extent that there is a "self", I'd say it involves not how you are at any given moment, but the range of modifications accessible to you over time.
Yea, there are a bunch of other things that happen, including realizing you don't actually want the thing and having a decaying initial momentum to do the thing. This strategy is mostly to help get over the fear of having to put in the same amount of effort every time and actually get started on doing the thing.
The Ten Percent Shift is a thought experiment I’ve successfully pushed to System 1 that helps build long-term habits like blogging every day. It makes the assumption that each time you make a choice, it gets 10% easier.
I'm wary of baking the number into this term, since the number is almost certainly wrong (but the basic idea seems roughly true). I'd contrast it, though, with a counter-problem (see "Superhuman by Habit" or "Sunset at Noon") that not going the gym also count as a habit that makes not-going-to-the-gym 10% easier. (But, formalizing this precisely gets pretty hairy)
At grave peril of strawmanning, a first order-approximation to SquirrelInHell’s meta-process (what I think of as the Self) is the only process in the brain with write access, the power of self-modification. All other brain processes are to treat the brain as a static algorithm and solve the world from there.
Let me clarify: I consider it the meta level when I think something like "what trajectory do I expect to have as a result of my whole brain continuing to function as it already tends to do, assuming I do nothing special with the output of the thought process which I'm using right now to simulate myself?". This simulation obviously includes everyday self-modification which happens as a side-effect of acting (like the 10% shift you describe). The key question is, am I reflectively consistent? Do I endorse the most accurate simulation of myself that I can run?
The meta process is what happens when I want to make sure that I always remain reflectively consistent. Then I conjure up a special kind of self-modification which desires to remember to do itself, and to continue to hold on to enough power to always win. I aspire to make this meta process an automatic part of myself, so that my most accurate simulation of any of my future trajectories already automatically includes self-consistency.
Also: enjoy your CFAR workshop! :)
An aside, some might be interested to know that the story of the two wolves was invented by a white man and runs against most tribes' notion of goodness and badness https://crossingenres.com/you-know-that-charming-story-about-the-two-wolves-its-a-lie-d0d93ea4ebff. You don't present it as an Indian story, but it's hard to hear the story and imagine that that grandpa is an old timey european, talking about good wolves. Europeans historically have not tended to believe in good wolves, or, in the least, they always call those "dogs". This extended to indian dogs, which were visually indistinguishable from, and probably genetically indistinguishable from wolves, but when europeans encountered them they could not bring themselves to call them wolves, for clearly they were sociable and well behaved, and wolves could be no such thing.
but it's hard to hear the story and imagine that that grandpa is an old timey european, talking about good wolves.
That is what I thought. Not "old timey" per se, but modern grandparent age. The story feels five to ten years old, to me.
I think there should be another side. Surely, under this model, falling out of a habit should be hard?
I very easily fell out of habit of doing 2-hours-long yoga-style exercises every two days after doing it for about seven years or more (parents' idea). What is more, I wasn't doing it alone, so sometimes it was easier to begin because someone else already had. I remember just being fed up with it.
Shorter and less Pruned due to CFAR.
I circumambulated the idea of meta-processes with the wonderfully inscrutable SquirrelInHell recently, and a seed of doubt has been circling in my head like a menacing sharkfin ever since.
At grave peril of strawmanning, a first order-approximation to SquirrelInHell’s meta-process (what I think of as the Self) is the only process in the brain with write access, the power of self-modification. All other brain processes are to treat the brain as a static algorithm and solve the world from there.
It seems to me that due to the biology of the brain there is a very serious issue with isolating the power of self-modification to the meta-process. After all, every single thought and experience causes self-modification at the neural level.
This post is another step towards a decision theory for human beings.
Unintentional Self-Modification
There is a central theme buried in my post The Solitaire Principle about building habits across time: human beings are not rational agents. We are not even “bounded-rationality agents,” whatever that means. We are agents who cannot simply act because every action is accompanied by self-modification.
Every time you take an action, the associated neural pathways are bathed in the magic brain juice [citation needed]. When you go to the gym, it becomes easier to decide to go to the gym next time. The activation energy for the second blog post you write is lower than that of the first. Acquired tastes are a real thing. After repeating a habit for a month, it is practically free.
Due to magic brain juice, every action is accompanied by an unintentional self-modification.
Let that sink in.
Every action you take is accompanied by an unintentional self-modification.
In an iterated game involving human beings, the choices made in each round influence not only their scores but their utility functions in perpetuity. Even disregarding sunk costs and irrational attachments etc etc, it literally becomes easier for Brain 1 to press Button A the second time around.
The Ten Percent Shift
The Ten Percent Shift is a thought experiment I’ve successfully pushed to System 1 that helps build long-term habits like blogging every day. It makes the assumption that each time you make a choice, it gets 10% easier.
Suppose there is a habit you want to build such as going to the gym. You’ve drawn the pentagrams, sprinkled the pixie dust, and done the proper rituals to decide that the benefits clearly outweigh the costs and there’s no superior alternatives. Nevertheless, the effort to make yourself go every day seems insurmountable.
You spend 100 units of willpower dragging yourself there on Day 1. Now, notice that you have magic brain juice on your side. On Day 2, it gets a little bit easier. You spend 90 units. On Day 3, it only costs 80.
A bit of math and a lot of magic brain juice later, you spend 500 units of willpower in the first 10 days, and the habit is free for the rest of time.
So the thought experiment is this: feel how difficult going to the gym is once. Call that x units of effort. Now, imagine you get to trade 5x that total effort for the results of going the gym for a year. Decide whether you go today based on your reaction to the Ten Percent Shift.
The exact details of the Ten Percent Shift don’t matter - the goal of the Ten Percent Shift is to convince System 1 that a single act of installing individual daily activities has far-reaching consequences. Thus: