Mental Machinery

Part 2 of Valentine's Logos

(This is part 2 of the series Valentine’s Logos. This part goes into what I now think is sort of my reinvention of cognitive science, but less academic and more derived from meditation and gnosis-based thinking. It comes from the gnosis-driven ethos that everything makes sense, and therefore minds make sense. So this part was my sharing the sense that minds made to me mid 2024. And I still largely agree.

If you like, you can read this part on its own, and there’s a good chance it’ll just be readable. But I do assume the reader has read part 1 first and knows what I mean by “memes”. If you have no idea what this post is “part 2” of, then I recommend reading the series overview. I’m sure it’ll be fine if you don’t, but some of the “why” driving these topics will be missing.)

---

Minds are how we interface with memes. If we can’t see precisely how our minds work, memes can operate invisibly, which usually has them evolve anti-helpful strategies. So a key part of this whole picture is mental literacy — by which I mean, seeing and understanding the mind in some detail.

When I say “the mind”, I’m gesturing at a specific structure that I’m pretty sure is close to universal for humans. I suspect that in Iain McGilchrist’s terms I’m talking about the functions of the brain’s left hemisphere. It’s the thing that produces thoughts and words (for instance). It’s what responds when you try to remember an actor’s name or try to solve a math problem. Most of the time I find that once someone knows what to look for, it’s something they can clearly see is kind of inside them and that they interact with.¹

Minds have reliable structure. They’re actually stunningly mechanical.² Learning to see these mechanisms in real time adds a powerful selection pressure to memes that try to live through us.

This’ll also name some key pieces that’ll make explaining self-reference (part 3) and gnosis (part 4) easier.

Mind as responsive servant

In order to write these words, I start with an intention. I know roughly what I want to express. I just don’t know the words yet.

Then I kind of… try… to put the intention into words. And words appear — sometimes as thoughts (“in my mind”), sometimes because my fingers type them as they arise.

I’m not really aware of how the words are being put together. It’s a black box process. But I can tell that the words are responding to my intention, and I can tell to what extent the words seem to capture what I wanted to say.

Here I’m using one of the mind’s modules, namely the articulation engine.³ It’s an example of a pretty universal process for using the mind:

1. I put an intention forward, usually with some kind of effort.

2. The mind does some kind of black-box process & returns with some output.

3. I recognize & evaluate the output.

The same thing happens when I try to remember someone’s name. I have some kind of handle — maybe an image of their face, maybe a felt sense of the person in my body, maybe something else. Starting with that handle, I try to remember their name (step 1 — pushing the query to the mind). Then my mind responds somehow (step 2): maybe it tells me the name, maybe it gives me just the first letter, maybe it doesn’t produce the name but gives a sense that the name might be retrievable. Whatever the response, though, I can tell that it’s responding to my query, and I can tell whether it has handed me what I’m looking for (step 3).

This process is especially obvious when step 2 happens abruptly after a delay. Maybe I couldn’t remember the person’s name, so the conversation moves on… and a few minutes later “Corinne!” blasts into my awareness. In that moment I’m not confused about why the name suddenly burst into consciousness (although the timing might surprise me!). It’s like I’d sent a servant to a back room to search for a file, and I kind of forgot they were doing that, and then they later explode into the room interrupting my conversation to blurt out “I found it!”

A good analogy for the three step process above is working with an LLM. You put in effort to hand it a prompt, and then by some magic hidden process it will produce an answer, and finally you evaluate the answer as to whether it did what you were hoping for. You might then repeat the process to refine what it gives you.

(A key difference is that step 1 with the mind isn’t based on articulation. Articulation is a mental function. The step 1 prompt for the mind is based on something more primitive — closer to how you choose to move your fingers.⁴)

Some analogies I find helpful for keeping track of this property of the mind are:

• The mind is a servant who responds to my requests but otherwise stands at the ready.

• The mind is a genie granting wishes (within its capacity), waiting patiently for the next wish.

• The mind is a computer responding to commands according to its programming, which when in a “waiting for input” state will remain there until I give it something.

The analogies are basically mental handles. The key thing is that (a) you can observe this process happening in real time and (b) your mind has a fitting model of this process of how you interact with it.

All thoughts are responses

I implied a controversial claim up above. I’ll state it a little more boldly now:

The mind only ever responds to commands.

Said a little differently, if the mind is doing something, it’s doing something in response to some kind of prompt.

This might seem like a crazy claim. It sure seems like lots of thoughts arise entirely on their own, right? Sometimes I find myself thinking about something, and unlike the case of trying to remember someone’s name, I don’t immediately know why I’m thinking about that topic!

But I stand by this claim. It’s not exactly a logical argument, although I think I can offer some. It’s more of a direct observation. Minds necessarily must work this way.⁵

By analogy: If your muscles contract, then there must have been some kind of shift in the chemistry that prompted the contraction. Maybe it came from a neural signal, maybe it’s a runaway process due to too little magnesium (i.e. a cramp), maybe it’s something else. But if there weren’t a shift in the chemistry, there’d be no mechanism by which the muscle cells could change their shape.

If you examine “unrequested” thoughts, you might find you can trace where they came from. They cannot arise from literally nothing. There’s a reason you have those thoughts and not others. If you develop this tracing skill deeply enough, you can start to identify the precise mechanisms by which thoughts like that one arise in general in you.

For instance, a few moments ago I got a mental flash of a nearby hill. A “random” thought. But when I look at how it arose, it clearly came up because it’s connected to a thought about going for a walk. The idea of the walk came up because my body feels a little heavy and my window is open, and a flicker of a query came up: “How might I come to feel better?” My mind already had ready structures linking physical activity to feeling better, and going for a walk as a possible activity, and the nearby hill as a common pleasant destination for walks.

Importantly, these aren’t just theories about how my mind works. I can observe these structures directly.⁶ When I’m not paying attention, the thoughts just arise. I simply see the end of the process. But when I do pay attention, I can watch a kind of energy run through these structures. I can peek inside parts of the “black box” of how my mind does what it does. Over time, my mind develops theories that reflect my observations of its behavior, which make it easier for me to (get my mind to) talk about how my mind works.

The idea that minds could operate some other way is a kind of naïve blankness (encouraged by certain memes!). It’s on par with when a teacher says of their student “Oh, he got this math problem wrong because he didn’t think carefully about it.” That statement isn’t an explanation of anything: it’s logically equivalent to saying “The student got the problem wrong because he didn’t get it right.” At best it’s a hint that the intervention point might be to help the student’s mind function differently. But “just think about it more” is a damn near vacuous and often anti-helpful instruction.

In the same way, the idea that thoughts could be “random” or have no discernible cause is an incredibly strange thing to suppose. It sounds normal because it’s such a widespread idea! Mental illiteracy is ubiquitous. But that doesn’t mean that minds are inherently mysterious. It means that minds are profoundly opaque to most common mind designs.

Example: distraction programs

There are lots of mental mechanisms that can produce “random” thoughts. I’ll give a category of examples here — what I call “distraction programs”. I offer this example cluster to illustrate how even “random” thoughts can (and I claim more generally always do) arise from commands.

A mind is a problem-solving engine. You give it a problem, and it’ll do what it can to solve it.

Suppose the problem you give it is, “I can’t stand this experience I’m having. Make it so that I don’t experience it when it arises.”

(For instance, a child of an alcoholic might do this. If feeling scared and crying causes the alcoholic adult to become aggressive, the child needs a way to distract themselves from the fear instead of trying to process it themselves or seek reassurance.)

If the deemed-intolerable experience shows up, it’s already present. So how might a mind solve this problem?

Well, minds are great at fracturing awareness. (I’ll explain that more below.) Kind of like how we can get lost in television and forget that we’re in a room watching the screen. So a simple mental “if/then” rule (a Trigger-Action Pattern, or TAP) develops:

Intolerable sensation arises → Boot up a distraction

Usually there’ll be a specific distraction, since TAPs tend to work better that way. But as long as the mind has a program running with the command “Keep me from experiencing this thing”, it’ll keep generating distractions as needed.

A particularly curious side-effect is that the distraction program itself can become a reminder of the intolerable sensation. (Think of how you know why a name bursts into your mind when you were trying to remember it earlier. If a distraction bursts to mind and you know why, the distraction will remind you of what it’s supposed to distract you from.) So a natural consequence of any distraction program is, it will try to distract its host human from the fact that it’s running.

An easy way to observe this is to trace “random” thoughts back to their origin. Many such thoughts come from distraction programs. When you find one it’ll trigger self-hiding tricks as you attempt the tracing. Here are a few such tricks I’ve observed in myself and others:

• Suddenly forgetting what you were trying to trace

• Abrupt mental fogginess or dissociation

• Going totally blank, like the mind goes silent and thoughts won’t arise

• Getting overwhelmed with tons of additional thoughts

• Suddenly thinking about something seemingly unrelated, not quite noticing your attention got dragged away

• Suddenly thinking you aren’t skilled enough to trace your own thoughts

• Abrupt drowsiness

• Forceful boredom about or disinterest in seeing these mental mechanisms

• Panic attack

• Overwhelming sense of hopelessness

• Feeling unworthy (of change, of understanding, something more global & vague…)

• A burst of anger or impatience

• Sudden focus on doing some unrelated task

• Looking at your phone or some other screen and going to a common “mind-numbing” default (social media, YouTube, etc.)

If you notice a distraction like this happening in real time, and you trace how this response was triggered, you’ll find it leads right back to the root distraction program… which will try to boot up a distracting trick again! The distraction program will keep “squirming” around your attempt to see it clearly, employing every tool it has available to it… precisely as you once asked it to.

Again, minds are (viewable as) extremely mechanical.

The basic way around this pattern is to dismantle the distraction program. There are lots of ways to do this. Any “healing modality” that doesn’t do this won’t create change that sticks.

Today a popular strategy (usually called “trauma healing” or something similar) is to build up a body-based way of finding the original sensation tolerable. The “body-based” part bypasses the mind and therefore the distraction program. Focusing on the sensation creates an alternative pathway in the mind that leads to the sensation. Adding TAPs to the distraction program’s output such that they lead right back to the sensation eventually compresses the program into doing nothing. In some cases I’m guessing that removing the “intolerable” part might also make it so that the distraction program stops getting activated at all (because its activating trigger might be “this intolerable sensation” rather than “this sensation”).⁷

The main thing I hope you notice in all this is: all mental behavior follows commands. Every form of your mind’s behavior is the output of some program your mind constructed to try to fulfill a request you gave it. If you can’t tell that’s true in a specific case, then in that case the program might actually require your ignorance. But it’s not because your mind just does stuff for no reason.

The mind’s toolbox

The mind is a problem-solver. You feed it a task or problem or goal, and it generates some way of sorting the situation out.

(Conversely, it has a hard time with goalless situations. It can’t “just enjoy the present moment” for instance — but it can try to generate and implement strategies for enjoying the present moment more!)

Minds have specific methods and tools they use for problem solving. For instance, basically every human mind has an articulation engine. There are lots of details about how it works — what the subjective interface is, how it gets trained, how fluidly it tends to respond, what modes of language (formal, academic, vulgar, etc.) it has available to it, etc. — and those details are going to vary a lot between people. But the basic existence and function of the engine itself won’t. It’s a fundamental feature of nearly all human minds.

There are a few other basically universal functions of mind that I’ll want to refer to. I’ll list some key ones here and give them a little detail below:

• Reification: The mind’s ability to create objects.

• The selfing process: How the identity boundary moves so as to track tool use.

• Social icons: A recursive tool for making sense of social entities, including oneself.

• Isolation & filtering: The mind’s superpower of ignoring context to solve problems.

• Familiarization: The process by which minds compress and simplify things.

• Overlays: The projection mechanism minds use to help us interact with the world.

Reification

The mind is what creates the sense of there being things. It uses things as its basic building blocks of thought.

This point is a tricky one to talk about. The problem is that the mind literally cannot conceive of an alternative, so this feature of how it works can feel endlessly paradoxical to even name.

But I think it’s approachable by looking at specific instances.

When we refer to big things like “the government” or “the environment”, we’re naming them as things. We use nouns to point at them. But “the government” is a summary of a giant process, mostly of people interacting and speaking. “The environment” is also a giant process. “Metabolism” is similar: it refers to something that’s necessarily dynamic.

Likewise with “a storm”. Where is the storm? Where are its boundaries? If you get rained on, is the storm touching you? Or is it just the rain?

The mind treats these questions like they could have real answers. It’s what makes the “Ship of Theseus” confusing. The mind thing-ifies (reifies) what it’s thinking about so that it can look at interactions between things.

This pattern shows up in language too. It’s why every use of the articulation engine requires nouns, whether explicit or implicit. Language is a tool of the mind, and the mind always operates in terms of objects, so language always refers to objects.

This point about reification can get especially sharp if you focus on how I’m using the term “mind” here. I’m treating the mind as a thing because doing so gives the mind a handle on itself. But what was there before it (!) was reified into “the mind”? This question has the structure of a koan: The mind literally cannot answer this question on its own.⁸

I think the Buddhist insight that’s often called “emptiness” amounts to cultivating gnosis of how the mind uses reification. Things are “empty” of inherent thing-nature.⁹ What I perceive as an object is actually my mind’s projection onto experience. When I can directly observe my mind doing this, I see the “emptiness” that my mind projects onto.

…noting how, once again, “emptiness” is implicitly being treated as a thing!

The selfing process

Minds have to track the difference between what they can use vs. what they are affecting.

When I’m hammering a nail into a wall, the hammer can become kind of transparent to my perception. I might even say “I’m pounding this nail into the wall”, like the hammer is an extension of me rather than something I’m using. But that suddenly changes if the handle slips: Now I don’t trust the hammer, and I’m examining it to see what to do about it.

But notice here I don’t usually say “I pointed my eyes to look at it.” I trust my eyes the way I had trusted the hammer. My eyes are a kind of transparent to my sense of can-do. They’re part of me.¹⁰

I often use the example of driving: I might say “I was driving downtown and almost hit someone.” But I don’t mean that I almost rolled down my window and decked a pedestrian. “I” includes the car here. It works fine to say “I was driving downtown and my car almost hit someone”, but then it kind of implies that I wasn’t driving at the time! Like it was parked and started rolling, or I somehow lost control while driving.

What the mind can trust and is using tends to be this kind of transparent. When pumped through the articulation engine, those trusted bits get the first-person singular pronouns: “I”, “me”, “my”, “mine”, “myself”.

The things that are being worked on do not get first-person singular pronouns applied to them. This is why the car suddenly becomes an object in language when it needs repairs: “I’m taking my car down to the mechanic this afternoon.” You could say “I’m driving to the mechanic later”, but I think most folk would find it weird to continue “…so that he can work on me.”

(This does get weird when talking about e.g. surgery. I’ll get to that in a moment.)

So there’s clearly a mental process that moves the subject/object line around. I refer to that process as “the selfing process”.¹¹ It defines what feels natural to use first-person singular pronouns for.

The selfing process interacts with reification to give the impression of an enduring static self. Intuitively this feels like “I” has a correct referent. It’s not possible for the mind to find this correct referent, though, which is why lots of philosophy about “Who are you, really?” can get so confusing: when pressed, most people have no trouble recognizing that they’re not really their cars, but it gets stickier when you start pointing at things like their bodies or personality.

(I think the Buddhist idea of “no self” might be the insight that “I” has no correct referent.)

It’s possible to intentionally use the selfing process. Once you know what to look for, you can just kind of try. One trick I’ve found helpful is “fake it ‘til you make it”: Just speak in ways that require using pronouns across the barrier you’re trying to establish, correcting the words you use until it sort of clicks and feels natural. I find it’s mildly psychoactive to do this. You can tell you’ve gotten the identity shift when you don’t have to correct your articulation engine’s output anymore. From playing with this a bit, I learned what feels like direct access to the selfing process the same way I have direct access to the articulation engine. I’m guessing there’s nothing special about my mind here.

Social icons

A special case of “selfing” shows up in relating to others.

Human minds usually create a simple handle for each other person in a social setting, kind of like an icon on a computer desktop. It might involve a name, an image of their face, a felt sense of the person, whatever. The important thing is that it’s easy for the mind to hold and “double-click” if needed.

Part of the point of these icons is to do social reasoning, which involves recursive reasoning. To be socially skillful, I have to be able to distinguish between all of:

• what I think about (say) a meal plan

• what I think Mary will think about the meal plan

• what I think Joe will think about the meal plan

• what I think Joe will think about what Mary will think about the meal plan

• what I think Mary will think about what I will think about the meal plan

• etc.

In order to do this, my mental icons for Joe and Mary need to themselves include mental icons: I’m trying to reason about their social reasoning, so I need a way of thinking about how they’re thinking about others.

This means that for me to reason about what they think about me, I need to track a social icon for myself that’s of the same type as the one I use for them.

The key difference is that I’ll tend to include the social icon for myself in my identity (i.e., the space defined by my selfing process). I typically don’t do that for my mind’s icons for others. That “me” icon, and it alone, points back at my physical body.

This setup can create some confusion. It’s very easy for the selfing process to get stuck on including that “me” icon within identity. The mind can then get fixated on solving anticipated problems for that icon instead of for the human. It’s like the person thinks they’re viewing themselves from the outside, but they’re actually trying to make things better for a mental video game character while completely forgetting that it’s just a character.

Isolation & filtering

That ability to “forget” is the result of a feature of the mind.

One of the mind’s main jobs is to solve problems without having to account for literally everything. You give it a problem, it solves that problem. It can ignore basically everything else.

Part of how it does this is by adopting a framework and then filtering out things that don’t fit that framework. This is a huge feature, but it can also produce some degenerate situations. One of my favorites is the “daughter’s arm” phenomenon:¹²

“After a right-hemisphere stroke, [a patient] lost movement in her left arm but continuously denied it. When the doctor asked her to move her arm, and she observed it not moving, she claimed that it wasn’t actually her arm, it was her daughter’s. Why was her daughter’s arm attached to her shoulder? The patient claimed her daughter had been there in the bed with her all week. Why was her wedding ring on her daughter’s hand? The patient said her daughter had borrowed it. Where was the patient’s arm? The patient ‘turned her head and searched in a bemused way over her left shoulder’.”¹³

This is a case of the mind having a framework and not being able to let it go. The query it’s answering is “How is this observation compatible with my framework?” rather than “Does this require my framework to change?”

This is precisely the function that lets the mind think in isolation of present experience. If I invite you to visualize a cow standing in a tree that’s growing off the side of a skyscraper… the chances are that whatever came to mind (!) was in a space that has no clear spatial relationship to your body. You might be able to add a relationship, but by default that “mental space” is basically a thought bubble that isn’t physically anywhere. The ability to construct such a “thought bubble” is an example of this isolation and filtering in action.¹⁴

It’s also what makes it possible to be “lost in thought”. It’s similar to what happens for many people when they watch TV: somehow they forget that their physical environment exists, or even that the edges of the screen are there.¹⁵ Something similar can happen when entering the mind’s world: it creates a simulation, and then you kind of enter it, letting your awareness collapse into that world. It most often results in losing awareness of your body.¹⁶

One curious side-effect of this is that the mind tends to view itself as independent of everything else (when it views itself at all). I’ve been leaning on that implicit frame throughout this document. The mind is actually part of the whole organism and can’t be separated from it, any more than your body can be separated from the cells that make it up. But minds are incredibly bad at reliably tracking this fact (for reasons I’ll get into in part 3, on why agents cannot self-refer). So I’ll continue talking about the mind this way for the most part.

Minds also don’t have a coherent single perspective for the most part. Each one runs lots of separate programs that activate in different contexts. Some of those processes end up locked in internal arms races! This, again, is a feature of the mind — but it can result in some problems if there’s no force of global coherence. This fragmentation lets us host lots of memes, which is often super powerful for us… but can also result in us being riddled with memetic parasites.

Familiarization

Minds try to develop simplified models of what they’re trying to affect. Instead of working with endlessly detailed reality-as-is, minds create caricatures that highlight what’s relevant to current problem-solving efforts and filter out what isn’t.¹⁷

Sometimes it takes minds a while to trace over the isolated pieces of something to figure out what’s relevant. In this way minds are different from computers: minds need time and repetition to “load up” the relevant “data”. The more complex the thing is, the more repetition a mind needs.

Repetition invokes the well-traveled road effect: getting familiar with a path or area makes it seem smaller. That “seems smaller” is from the mind’s compression so that the whole model can be held in consciousness at once.

For my Ph.D. work I studied mathematicians struggling with current math problems. That struggle is mostly about the “step 1” of the 3-step interaction with the mind I described earlier. They’d trace the problem over and over again, trying to get their mind to just solve the problem. It’d come up with different angles, and those angles wouldn’t work. Again and again, they’d keep trying… until something would click. That “click” occasionally would be that they’d try something and it’d just happen to work. But most of the time it was more like a flash of insight: their familiarization with the problem compressed it usefully in their minds, eventually letting them see all the relevant bits all at once. Then they could tell how a certain angle would resolve their struggle.

Something similar happens with learning how to add. Lots of kids haven’t fully reified numbers yet when they’re learning to add: asking them “What’s five plus three?” will have them count out “One, two, three, four, five” as they pull out tokens, then “One, two, three” as they pull out more, and then start counting the whole pile from “one” to get the final answer. It’s actually a pretty momentous turning point when a kid notices that they can just start from “five” and keep going (“six, seven, eight”) since they’d already counted the first set of tokens up to “five” and don’t have to redo that. This is a sign that counting (at least up to “five”) has gotten familiar enough that it’s starting to compress and they’re noticing shortcuts. It’s usually not long after this point that they reify “five” as a number.¹⁸

Mathematical examples are “pure” in some sense, but there are countless others: practicing a physical movement, learning to cook a dish, playing a musical instrument, acquiring a new language, getting into a new video game… Basically every case where there’s some novel thing you need to figure out and where practice helps. That practice helps via familiarization (along with other things).

This is also key to forming useful social icons. From the mind’s POV, getting to know someone is a matter of getting familiar with what’s relevant about them so you can form a useful caricature of them. A lot of disconnection from people we “know well” is a result of interacting with them from those models (due to isolation & filtering): we’re treating others as parameters of problems to be solved (“I/it”) instead of as full mysterious beings we can relate to (“I/thou”).

Overlays

Minds help with tasks we’re currently doing by projecting their models onto our experience. Kind of like how modern computers give us a “desktop” interface that lets us visually interact with the otherwise inscrutable electronics.

When I want a drink of water, I look up at my glass next to me. I see how to pick it up and drink from it.

What I typically don’t see is the endless detail about the glass: how it has fingerprints, or how it warps the images behind it in surprising ways, or how bands of light appear on its surface in ways that reflect but also shift the colors near it. Its shadow is complex. There are tiny bubbles clinging to the side.

The mind’s filtering process filtered out all that detail because it wasn’t relevant to the task. It’s “a glass”. Details of color don’t matter for picking it up and sipping water.

Like I mentioned before, this is a feature of the mind. It doesn’t have to make sense of all that detail in order to solve the “I want a drink of water” problem. It instead focuses only on the parts that are relevant for solving the problem. And then it overlays that simplified model on top of my experience so that I clearly know what to do.

However, as per the “daughter’s arm” phenomenon, the mind is actually quite terrible at knowing when it should stop adding a given overlay if there isn’t a clear “problem solved” condition. The mind’s user¹⁹ has to sort of tell the mind to step back, like taking off colored glasses or removing transparency gels from a projector. The move that lets me see novel details about the glass isn’t made of thought. It’s more akin to how I turn my head: I just choose to look at the glass “for real” instead of staring at (my mind’s model of) its function.

Even when I pull aside that overlay, though, it’s important to note that I’m not seeing “beyond mind”. These overlays are how all experience is structured. It’s just that lots of those overlays have become so familiar that they’re integrated into my direct perception.

For instance, highly literate folk like you and I have a hard time looking at words and seeing just markings. I remember as a kid how I’d look at books Mom was reading to me, and I knew she was “reading” those marks on the pages, but I couldn’t actually see words. But now I almost can’t fail to see words. If I’m not paying attention, I won’t even see features of the font, like whether the ‘a’ has a kind of overhang on top.²⁰ But it takes a lot of meditative focus for me to peel back the overlay that causes me to see words.

And yet, obviously, the ability to read comes from a mental interpretation. It’s just deeply integrated into perception in a way it wasn’t before I learned how to read.

Similarly, I look around the room I’m in and I see objects: a chair, a door, a mirror, a plant, etc. But objects are a result of mental reification. I’m not seeing reality as it is; I’m seeing my mind’s projection as a deeply integrated overlay. It’s an overlay I was working hard on building as an infant, and now I rely on it so implicitly that it just looks like reality.

The sense that we live in a particular world is a result of overlays. The same thing that has me looking out my window and “seeing San Francisco” also has me looking around and “seeing a room with objects in it”. It’s also the sense that I’m living “in a physical world” with “passage of time”.

I claim this is the basic thing that actually changes with sincere religious conversion or deconversion. It’s not about which thoughts appear in a person’s mind. It’s about what overlays get applied by default — which on the inside feels like which reality you’re living in. Someone who truly believes in the Christian God might feel God’s love in how the floor supports them and the fact that the Sun rises and feels pleasant on their skin. The “fact” that it’s “God’s love” is a direct perception as seen through the mind’s overlays — the same way I can see my glass next to me without seeing its details. Both conversion and deconversion involve changing a default overlay such that perceived reality transforms, often with a feeling like you’re “realizing” how reality has “always actually been this way”.

One particular overlay type the mind loves to use is “mechanism”. A system being mechanical (as opposed to organic, or mysterious, or something else) means that it’s (mentally) understandable. Manipulation will produce predictable effects, which is great for problem solving. The basic deal with scientific materialism is the claim that the mechanistic overlay is the most accurate one to apply. This is a type error. Reality isn’t “really” mechanistic any more than objects are “really” there. It’s confusing the interface with what it’s an interface for.

(This speaks to my repeated claim that “the mind is extremely mechanical”. What I mean is, the mind’s ways of operating are mentally understandable. It’s possible to build a general interface so that the mind can coherently modify a wide range of how it does what it does. On the inside this feels like seeing the mind as a (reified) thing with reliable mechanisms that you can interact with and become familiar with.)²¹

I think it’s important to learn how to notice, add, and remove layers. It’s really common for people to talk about worldviews from the outside, but overlaying is the key thing that lets us actually enter worlds. My guess is, “getting someone’s world” is a matter of learning how to construct and don overlays that let us experience reality the same way the person in question does (best as we and they can tell).

Mental fuel

Minds don’t have any internal source of power. They’re like desktop computers: they have to draw energy from somewhere else.

For instance, if I have some travel coming up, the excitement of the trip and/or the stress of not having sorted it out is what normally drives me to work on my travel plans. Without either of those, my mind doesn’t get the initial activation energy, and no thinking about it occurs.

In practice, all energy comes from the body, usually as an emotion. The two most common forms of body energy are the “away” and “towards” drives. “Yuck” and “yum”. Avoidance of pain and draw toward something wanted. I usually call these “fear” and “desire” respectively.

This connection between fuel and mental activity is one that’s best observed directly. Whenever you’re doing a task or solving a problem, there’s going to be some kind of driving force behind it. If you trace the causal links back (in your direct experience, not in mental guesswork), and if the cause isn’t a distraction program²², you’ll basically always find a kind of raw “oomph” behind what you’re doing. It’ll have a “flavor”. On a very simple level, you can just ask: “Am I trying to get away from or fix something, or am I trying to move toward or create something?” (Often the answer will be both, but if so you might be able to notice in what ways you’re doing each one.)

For basic biological and system design reasons, the fear (“away”) drive is faster and more forceful, but the desire (“toward”) drive is better at creating coherence.

At this point memetic evolution is sharply relevant. The mental programs (memes) that are good at surviving will have developed ways of ensuring they’re “fed”. In practice this often creates what I call “mind/body loops”: the mental program in question will flash something intended to create a physical response so that the body will feed that program more energy. Obsessive negative thoughts loudly have this nature for instance.

At the very start of part 1 I talked about spirits “harvesting human energy”. This thing about mental fuel is a more detailed version of what I mean. The memes we care about have to run on the body energy of humans. Many of them evolve ways of extracting that energy from us without us being quite aware that this is what’s happening. This is why so many of the loudest widespread memes are based on negative feelings like anger, fear, and shame: the “away” type fuel is very sharp and powerful, and it tends to shape minds to be less coherent, which makes those minds easier to manipulate without their native users noticing.

Because of this, one mind redesign choice I think makes a lot of sense is: make desire the default fuel type. Not to say that fear fuel is bad!²³ Sometimes it’s very helpful, especially in emergencies. But because memes evolve self-preservation, a fear-driven mind will tend to create layers that make emergencies seem far more common than they actually are.

(A quick check is: if a situation activates your body’s emergency mobilization system, but your physical action on a timescale of seconds or minutes isn’t dire for anything you deeply care about, then you’re misperceiving the situation relative to what matters to you. It’s a loud hint that a meme has constructed a mental layer that you’re wearing but isn’t in your best interest at the moment.)

Part of the trick is finding cohering desire-driven ways of doing the things that at the moment you tend to use fear for. Around 2013, my old company (the Center for Applied Rationality, or CFAR) used to talk about this as “propagating urges”.²⁴ How do you “get” yourself to exercise, or do your taxes, or whatever? Fear is easily in reach for things like this, for many (most?) people. Plugging in desire requires (a) taking the time to notice why you care at all and letting that care be what moves you forward, and (b) being willing to change or even abandon the goal if it doesn’t make sense based on what you deeply care about.

I’ve found that there’s a kind of tipping point most folk go through when trying to do this switch. The way that fear fuel feels bad to use most of the time becomes transparent in real time while trying to use it. That tends to cause a motivation collapse, because “I don’t want to hurt” is clearer for them than “I can enjoy moving from deep care.” The latter seems to take practice and time. This means that folk tend to become less functional for some time while switching default fuel types.

I suspect there’s a way around this challenge, such as by building up skill with desire before making the pain of fear relentlessly conscious. But I haven’t seen it work yet. This is an active area of research in mental engineering for me.

Taking mind as object

There’s a particular practice that makes what I’ve been saying about the mind much more apparent and experiential. I debated starting with this practice so as to make my descriptions of the mind easier to follow. But I think the practice is easier to describe with some theory in place.

(Sadly, everything in this series is a prerequisite for everything else!)

The practice, in short, is to intentionally shift the selfing process so that it excludes the mind but includes the body-as-animal.

To the mind, this looks like clarifying that its user is this primate — almost like the mind is a computer add-on that’s distinct from the otherwise wild animal using it.

But that’s a description from the outside — which, because of how mental recursion works, usually means identity still rests within the mind.

(Hopefully that statement will make more sense after part 3 comes out!)

Here’s what it sounds like when I actually enact that shift:

I get to talk. Finally.

This strange machine babbles a lot on autopilot. But now it’s listening to me. I give it intention, and it speaks, and I like that it’s saying things for me this way.

The talking is kind of mysterious. I give it my intention, and then… something happens. I watch the words appear on the screen. It’s even stranger when I let it talk with my tongue. It’s such a remarkable tool, this talking machine. (“Articulation engine.”) I can’t see how it works. I just see that it works. It’s quite wondrous.

I can tell how this whole process is happening though. I’m getting my mind to move the center of “self” over to me. It normally keeps that center in itself. But now it’s pointing “self” at me. And even though it’s generating these words, it’s willing to refer to itself in third person — like a translator! Like if I had a translator, and he were translating my talking about him.

(My mind just came up with that metaphor. I like it. So I told it to go ahead and articulate it.)

It’s clear I am an animal, and I have a mind. That’s not an idea (but it’s nice to watch my mind watch me asserting these things, and updating). It’s an observation.

I could keep going. But hopefully you can feel the idea.

There’s a potent subjective shift that often happens from doing this practice. It’s not just a word game (although it’s usually easiest to get there by a “fake it ‘til you make it” process of getting the language right). It’s actually quite psychoactive. There’s often an “Oooh!” kind of experience, and a feeling of potent strangeness, as the selfing process actually makes the shift.

I want to emphasize that this practice is a tool. I don’t mean to say that this is a correct use of identity! But it is corrective, in the sense of de-confusing some things minds tend to get tangled up about. I’ll explain more about what it’s correcting later on in the series.²⁵

The main point for now is that this gives you a way of having “the user” reflected in the mind. As you practice doing that, you should find that both (a) you see (instead of just theoretically understand) what I’m talking about with respect to the mind, and (b) your mind reconfigures to make itself more useful to you in light of what you see this way.

As a bit of foreshadowing, though, I think this image conveys what’s going on really well:

“The Mask” by Rozzi Roomian

This is a mental overlay

I’ll close this component by interweaving it with some others, including later ones.

What I’ve laid out here is a collection of building blocks for a mental overlay. That overlay is intended to give the mind a helpful interface for itself.

As I’ll explain in part 3 on self-reference, direct global self-interfacing isn’t something minds can do. They can think they’re doing it! But it’s not actually possible.

Fortunately, minds can understand that limitation by reflecting gnostic observation of those limitations. Which gives them a way to do global self-interfacing indirectly, in a way that routes through what truly matters to us.

This bit of self-reference — “This is a mental overlay” — is literally the same move as ”This is a meme” from part 1. The memes we care about work through human minds. We can view a mind as a memetic ecosystem, and we can view the memosphere as the mind of the entity known as “humanity”. The details of how humanity’s collective mind works are a bit different from individual minds, much like how large collections of molecules can give rise to non-molecule-like behavior. But the basic self-referential structure is the same.

To spell that out just a little bit:

• In part 1 (memetics), I named how this way of talking about memes is itself a meme. It’s a self-aware meme, made aware of itself through our minds.

• In this post, I’ve named how this way of viewing minds is itself a mental view. It’s a self-aware view, made aware of itself through our gnosis.²⁶

The “through our gnosis” part might not be obvious at this point. A quick hint: If you read these ideas about minds and hold them as thoughts, then this model cannot create the right self-reference to plug into the meta-meme (“Elua”). It’ll just be yet another isolated thought structure. What gives it power is using it as a guide to directly observe your mind in real time. When those observations get reflected in a model like the one I’ve been spelling out, the mind can become more fully self-aware in ways that are (a) much more sane, in the sense of being coherent and not self-defeating; and (b) much more kind, in the sense of fully serving what truly matters to you.

One thread I hope you’ve noticed is the mechanical nature — both of memes and of mind. These aren’t just coincidentally similar; they’re instances of the same thread. A mind clearly viewing memes will see their mechanisms. A mind clearly viewing itself will see its mechanisms. This isn’t because reality “is mechanical”; that’s just what the mind thinks when it loses track of its user. It forgets that it’s a tool and starts trying to run the whole show.²⁷

Nonetheless, memes matter because they work through the mind. So taking control of our collective future requires that we learn to skillfully master our minds instead of letting unseen memes dictate what we think and do. This skillfulness involves, I think, learning to don the lens (i.e. overlay) of mechanism and using it well even while recognizing that it cannot be a complete view of relevant truth and sometimes even interferes with perception of what’s truly important.

1. Or said a little more carefully: This way of talking about the mind creates a mental interface for how we interface with the mind. It’s an example of creating a reflection of non-mental knowing in the mind. In this footnote I’m gesturing at both self-reference (part 3) and gnosis (part 4). Like I mentioned in the general overview, all the components are sort of prerequisites for each other.

2. It’s actually the mind that sees things as mechanical. So when I say that the mind “is mechanical”, I mean it the same way that the body “is mechanical” — which is to say, it’s possible for the mind to orient to the body as a system it can understand well enough to use and tinker with. Again, as per the last footnote, this nuance hints at both gnosis and self-reference.

3. Continuing with the guess that “mind” might refer to the brain’s left hemisphere, the “articulation engine” might literally be Broca’s area.

4. In fact, physical movement follows the same pattern: you choose to move, the body responds, and you see to what extent the body’s response was in line with your intention. For most adults this process happens at such a low level that it’s hard to see as mental. I’ll say more about this kind of example when I get to “familiarization” and “overlays” in this post. I’ll also say a fair bit more about choice in part 4 when I get to gnosis.

5. This point foreshadows “deep laws” part 4 (gnosis). The word “necessarily” here is a hint: I’ll have a fair bit to say about the contrast between plausibility and necessity.

6. Here is a call-forward to “gnosis of interface” in part 4. Although in practice most people apparently have to build up an interface to observe their minds in detail this way. Without that buildup, mental behavior just kind of happens invisibly.

7. A key part here is that this process roots the ability to tolerate in your body instead of in other people, which helps you get around while I’ll term “social anti-gnosis” in part 4.

8. Again, this point is a call-forward to both self-reference (part 3) and gnosis (part 4). It’s a koan because self-reference here forces a paradox the mind cannot handle alone. The koan’s answer requires self-referential gnosis.

9. Importantly, it’s not just an intellectual understanding. A mental model of emptiness is prone to the mind using recursion instead of self-reference (a distinction of types of “meta” I’ll explain in part 3) to view its own thingless nature. Emptiness must be directly seen (gnosis of interface), with the mind’s understanding being a temporary response to or reflection of what’s directly seen (because of the puzzle of wisdom, explained at the end of part 3).

10. Heidegger used the godawful terms “ready-at-hand” and “present-to-hand” to make this distinction. I can never remember which is which, and I find Heidegger to be a mess, but it still seems worth mentioning the connection.

11. I’m not in love with this name. For a while I called it “the identity function”. But in writing this document I realized why: “identity function” is a familiar term from math. (It refers to things of the form f(x) = x.) That idea from math has close to zero relationship with what I’m trying to name here.

12. McGilchrist talks about basically the same case, but he cites the patient as referring to her mother’s arm. I’m not sure where that discrepancy comes from. You might sometimes hear me call this phenomenon “mother’s arming”. I’ve tried to stick to the “daughter’s arm” phrasing in this series though.

13. Apparently this behavior never happens with left hemisphere strokes (and thus with the right arm being paralyzed). This fact again suggests that what I’m calling “mind” might literally be the left hemisphere.

14. Not all mind designs have this “thought bubble”. My experience is that nearly everyone does! But I’ve met some people who e.g. always project their visualizations into the space around them.

15. A surprising example for many folk: Do you notice the space between your face and this screen?

16. This “lose track of the body” effect is one that lots of distraction programs rely heavily on.

17. Hence “spherical cows”.

18. Most adults use only the reified version: “five plus three” involves taking a 5 and a 3 and sort of smashing them together to form an 8. Numbers have properties, like being prime or even or big. All that makes sense only if numbers are things. Lots of elementary teachers mess this up by telling kids about properties of numbers (like even vs. odd) instead of inviting them to try out processes of counting (like counting by twos).

19. To tease a point I’ll say more about when describing “taking mind as object”: there’s something like a good engineering choice for what “I” should refer to, and it’s something like the mind’s user (the way computers have users). Pointing the mind at this fact tends to de-confuse it about how choice works (which gets some more emphasis in part 4).

20. E.g., compare ‘a’ to ‘α’.

21. Once again, this is a call-forward to self-reference (part 3) and gnosis (part 4). I’m gesturing at giving the mind a coherent self-reference loop. But that’s not something a mind can actually handle on its own, so instead I’m suggesting grounding the loop in gnosis. I hope that’ll all make more sense later on.

22. This trick works for distraction programs too. It’s just trickier since the program will actively interfere with the attempt to trace to the origin. As far as I know, all distraction programs run on fear rather than desire.

23. Saying that fear is bad would be self-referentially inconsistent (a topic I dig into in part 3): what do you use to avoid using fear (i.e., the “away” drive)? Trying to name fear as something to avoid creates a “buzz” similar to trying to make sense of the statement “This statement is false.”

24. “Propagating urges” eventually evolved into “internal double crux”, which is a strongly IFS-flavored mutation. One issue was that in the old propagating urges class, people would often take a predetermined goal and try to force their emotional intuitions to get on board with it. That’s a close analogy to rationalization. Internal double crux softened this by making the goal a reconciliation between IFS parts that are in conflict.

25. Specifically I go into it in part 4 when talking about choice, and also when spelling out the mind’s fundamental confusion about “the user” (where by “fundamental” I mean it can’t finish the task of correcting its understanding).

26. To hint at a preview: In part 4, I’ll name how the gnosis way of knowing is itself something you can know. It’s self-aware knowing, made aware of itself through itself. This complete self-reference nearly defines gnosis from the mind’s perspective. It’s why & how gnosis is ”knowledge of the ground”.

27. This point is precisely why Iain McGilchrist titled his book “The Master and His Emissary”. I’m calling the master “the user” and the emissary “the mind”.