#mind interferometry

We know this idea as a joke: "I'm not healthy, I just have a system of perfectly balanced disorders". But I think this is a useful model to try out, especially for mental stuff. What would happen if at least two disorders can coexist and mask each other?

It could hide or stop dysfunction or the differentiating features so much that the person doesn't fit either diagnosis - but internally there is still all the pain and struggle of the disorder. It's worth remembering that when trying to understand others, and ourselves.

It might help you be more durable, willing, or able in certain roles, in the face of certain pressures and risks - maybe sometimes that's even worth the personal cost?

When we say some meme like a phrase or symbol or whatever is a "dog whistle", you cannot just remember it, as if every time you see that thing it will have a clear, specific, or knowingly secret meaning. Dog whistling is not something you can boil down to superficial pattern matching.

The core of dog whistling comes down to mind interferometry. Dog whistles are born from the deep and huge information asymmetry inherent to different life experiences, not the shallow and tiny information asymmetry that you could compile into a dictionary of secret meanings.

Did I actually feel embarrassment pain before my first prediction tree from people laughing at me or judging me to something more directly negative?

I think there's a good chance that the answer is no. I think it's entirely possible, in fact logically necessary, that the reality is something like this:

Our genes cause our bodies to form brains that have certain neurological predispositions to detect certain patterns and react to them with greater activation - arousal of the sympathetic nervous system, readiness for action, maybe greater openness to forming more-trauma-like memories, etc.

This might include, for example, the pattern of multiple voices laughing in unison, or the pattern of multiple faces looking at us, or whatever.

But this activation is not inherently negative, at least not for everyone. It can be intense. Maybe for some brains it does contain an inherently negative component - sensations which are unpleasant or painful in some way - but probably not for all of us.

And I think it takes those early experiences, especially if they are intense enough to get saved the way trauma memories do, to create one or more prediction trees which cinch that into something more coherently like embarrassment or shame. Something more qualitatively negative and unpleasant and connected to other minds.

It might be notable here that when I was a child, I could often find myself feeling embarrassed when I was alone. My mind could try on the "what if this is observed?" cognition to set off the reaction, but hadn't yet trained up the "but there is no observer right now, so it's fine" and "if there is an observer, it's fine anyway" dismissals of that reaction.

To the best of my ability to find scraps of early childhood memory, I don't really know which came first. To the best of my ability to reverse engineer it with logic, based on everything I know about brains and minds, it only makes sense if some prediction trees towards negative experiences, and some mind interferometry and theory of mind, are grown first. Only then can "embarrassment" or "shame" be distinguished from similar physiological activations paired with uncertain or negative experience forecasts, like fear, startling, anxiety, and so on.

So then I think about how my root insecurity works, and like... my root insecurity, at its core, is a really thorough prediction tree/forest, made up of one or more prediction trees which start from various "bad cognition" patterns being matched strongly enough, and feed into more or less the same thick trunk of a myriad of bad possible outcomes.

So I say embarrassment was a big part of the emotional motivation/reaction/pain behind my root insecurity, but I don't think that the emotion of embarrassment is a raw negative experience in itself, independent of my root insecurity.

I'm starting to think more in terms of just cognition/experiencing and less in terms of minds, because:

Cognition/experiencing is more fundamental than minds. Mind interferometry starts not with other things having minds, but with other things having cognition/experiencing. Minds are only then inferred to exist in, or are used to model and predict, things which seem to persistently have some sort of cognition/experiencing that matters.

It does not take much to justify the idea that all cognition is potentially experiencing. At first the only cognition we know of is just our experience stream. The first reason we have to suspect minds in other things is when we realize their behavior is consistent with how we would act if we were experiencing their situation.

It is a lot harder to rigorously justify or even define any notion of minds. What mind interferometry evidence must a thing exhibit? What cognition/experiences must it have? How long? How often? How consistently? How coherently? In what combinations? At what speeds and time scales?

It is hard to draw the boundaries of minds. Mind interferometry tells us that cognition/experiencing happened seemingly within the thing. But where exactly is the mind? Is it even just one mind? Is it part of a bigger mind? If cutting off a part causes a reduction or change of cognition/experiencing, is some of the mind in that part?

Putting cognition/experiencing first kinda reveals that the whole idea of minds was a distraction anyway. Philosophically, legally, psychologically, and especially ethically. "Mind" is a heuristic, a summary, a pattern that things match more or less. To the extent that anything is best explained by having cognition/experiencing, it is mind-like.

TL;DR: Keep working to understand minds better, and care about them enough for that understanding to inform your decisions. Keep getting better at thinking and doing. Keep working to own and fix your flaws and problems. These efforts will gradually reveal more and more things as not ethical. What remains is the best approximation of ethics you can have, so far.

Long answer:

My philosophical foundation for ethics is summarized in my "From First Principles" post. A couple relevant elaborations and responses to typical objections to those ideas can be found at:

Some posts about the Experience Axiom

Some posts about the Empiric Definition of Logic

Sadly some other relevant clarifications and additions are scattered across my thousands of posts. They should mostly or entirely be tagged with the relevant tags though.

Of course, this only takes care of the more philosophically challenging part of justifying the existence of a likely external reality and other minds. I have not written up concise posts for all the steps beyond that. I think the broad strokes are intuitive enough from there, not enough to be rigorous but to take whatever seems obvious from there and run with it. I think I wrote a couple posts exploring what I call "the first value judgement", but basically:

On the one extreme, once mind interferometry lets you see that other minds seem to exist with roughly comparable experiences, you can decide that it is arbitrary to a priori value any mind over any other, and so you give all minds and their experiences equal ethical weight based on the empirical observation that they are all roughly the same kind of thing.

On the other extreme, you have unapologetic selfishness, based on the empirical observation that you only experience your experiences, so everyone else is a different kind of thing than you. Even when evidence is pretty strong that we are all having experiences, it kinda requires a motivated axiomatic leap to say "experiences which I never experience should be given as much weight as my experiences".

A third extreme is self-excluding altruism - just like the second and unlike the first, this is rooted in the empirical difference between the self and all others, but here the focus is flipped - the axiomatic choice is made the other way: the singular experiencer gives itself special insignificance, valuing its own experiences less or not at all.

Obviously you can make this initial axiomatic choice by putting the boundary somewhere other than just self/others, or even by putting down multiple boundaries or a gradient for how different minds are weighed.

What's cool is that you can see every one of these in different humans' moral intuitions, which is very validating of all my bullshit up to this point because it all fits - that's exactly what you'd expect if I'm right about how human brains empirically struggle their way through something like the progression from raw experiences which I outline in "From First Principles", but with distortions and errors added in by personal and circumstantial variation and the need to do it with naturally selected heuristics more efficiently within the constraints we have or had in our life or in our ancestral environment.

What's even cooler is that for several years now I keep realizing that each of these must converge onto the same ethics as you integrate the future and factor in the constraints that we are subject to and all the messy stuff in life, so I don't think which of the above choices you make changes the ultimate destination as much as it changes your initial naive ethically clumsy fumbling until you mature. Like, you'll make different mistakes and need different epiphanies along the way, or some of the same mistakes in different order, and you might have different dead-end branches of ethical development you might go into for a while, but I think there is something inevitable about being forced towards the direction of a certain natural ethics. The caveat is that many people don't seem to get nearly the large enough range of experiences to get very far with that progress, and it is progress which has a ceiling based on your mental abilities and how much you put in the work. Which also includes things like growing coping skill and healing your traumas and insecurities and so on (otherwise certain things are more unbearable than others, and this naturally distorts our ethics).

To be clear, I think you can grow all of the relevant abilities. I don't like the idea of there being some biological predestined cap on how far you can develop your ethics. It's just that it takes time and effort which is not very rewarded or supported or subsidized socially. I have the privilege of having the time and energy to think about this bullshit pretty much all the time, and thus to keep training the thinking "muscles" almost as a side-effect. I had the privilege of being raised by smart and fairly mature and compassionate people who kinda managed to cause or force me to develop certain mental skills. So even though I think wisdom and pragmaticism and uncertainty seem to combine to create pressures for ethics to look a certain way, because they tend to make lots of choices worse than other choices regardless of your base values, I think in practice at any given moment we're all capped in one way or another from seeing all those considerations all the way to the end, or even far enough to see if it's going to be a false or dead-end path.

This is actually very wrong in a subtle way. In truth, no choice of mind valuing is truly arbitrary - for any given life experience, one choice is the empirically justified one, but which choice that is depends on your exact life experience so far.

Consider a human infant, whose interactions so far have mostly been with its mother. Will this child experience the mind of a stranger the same way it experiences the mind of its mother?

Not at first. Early on in its life, the child has vastly more empirical mind interferometry evidence for the mother. When the child first begins to develop theory of mind, perhaps the evidence available to it so far suggests neither

"minds (of which I am one)", nor

"me, and other minds (which are a different kind of thing than me)",

Because you can work in the opposite direction: you can look at a behavior, interpretation, emotion, or other learned cognition and ask yourself: what prediction trees would cause it to make sense?

Which boils down to questions like "what combinations of experiences could have made this reaction logically correct, rational, or most compelling emotionally/motivationally?"

And you can ask something like that without prediction tree analysis - but without prediction tree analysis there are far fewer situations where you can reliably get an actual answer, and the answers you do get, or learn from others, are kinda like handwaves you take almost on faith.

Oh, sure, "because [earlier experiences], [person] now experiences [automatic reaction]". Anyone can have that thought shape. We can probably come up with some examples that might feel right and sensible and obvious to you. But why? How? What is the logically reproducible causative mechanism between [earlier experiences] and [automatic reaction]?

By which I mean:

In the one direction, if you only know [earlier experiences], what is the reasoning that would reliably figure out what [automatic reaction] is going to be?

In the opposite direction, if you only know [automatic reaction], and possibly other mind interferometry evidence about the person, what is the reasoning that would reliably come up with [earlier experiences] that must have happened? What is the reasoning that lets you figure out what else you might need to know to possibility collapse [earlier experiences] further?

Prediction trees.

Not that idea by itself. We still need breadth and depth of expertise with people, a rich mental library of correct-enough ideas about how minds work and cognition works, and other well-trained reasoning skills. The idea of prediction trees will not protect us from our own naivete, ignorance, bad assumptions, or reasoning errors.

I kinda always knew they had to actually be two Carls, because the mind interferometry just didn't add up - the statements quoted and ideas attributed didn't quite seem like they were consistent enough to come from one mind.