#cogneme

[mentalisttraceur]

hey do u think that concepts can be harmful by existing

Practically, yes. Precisely and philosophically, no.

Like, obviously concepts only exist within minds, only have effects through how minds process them, and so the answer depends on if it makes minds do externally or internally harmful things with it.

And I think for basically all concepts, you could come up with a mind that can "safely" handle it: think it, fully understand and appreciate it, and yet be fully healthy, happy, and unharmed by it, and act ethically despite it, indefinitely, at least according to any definitions of those positives I would agree with.

I also think that, even within the much more narrow constraints of only all possible human minds who are also able to function in this reality, the range of possible variability and cognitive skills is large enough to keep that true.

On the other hand, obviously for some concepts we could come up with a realistic human mind which just needs that exact push to do something harmful, or have a harmful reaction. But I think it wouldn't be a very practical definition of "harmful by existing" if it included things just because they might harm anyone, ever.

So can there be concepts which are so liable to cause harm, so able to influence so many people, that we should treat them as hazardous? Which we should perhaps leave mostly sealed up in university philosophy department research labs, or at least not teach or explain until you are mature enough, until you have the right mental tools/skills/values/perspective/etc?

Well, if we think of cognition in terms of natural selection on cognemes, with enough familiarity with what happens with genes in nature to compare... I think that kinda reveals that it's almost inevitable that there would be harmful stuff like that in the cognetic ecosystem. Nature is full of parasites that harm the host. Nature is full of arms races that pile costs onto both sides, where both sides harm each other... but neither side can exterminate the other, so the only way to ripple into the future is to keep up in the race.

Of course, this also means that overt, severe, or fast-acting cognitive hazards would be very rare. Because nature is also full of immune systems. Because parasites that kill the host too fast or show symptoms too visibly have a harder time spreading. Because most possible ways to suddenly dominate in an arms race have already been found and played out by the time you're born onto the scene, leaving just the small incremental as-tiny-as-possible-because-it-costs-almost-as-much-as-it-gains escalations.

And sometimes it's like ... maybe the only reason a concept is not harmful is because we've got a lot of mitigating concepts that we pair with it, cultural practices/norms/safeguards, or cognition we had to be raised/taught/hurt to have ready in our minds, and maybe the cost of maintaining all that is itself a harm.

Of course, the trouble with this particular question is that the better of an example I think of, the more it would seem like I shouldn't share it.

[mentalisttraceur]

Cognetic Openings

We have to be mindful of how our cognition can interact with other cognition. We cannot just say "the result was okay because of what my mind did, so what my mind did was okay". We must also ask ourselves: if anything was different in our mind or the situation, what would this cognition cause?

Be careful before dismissively saying "I would notice and think differently" - cognition that is repeated starts to be faster, more automatic. Thinking habits eventually become thinking reflexes. Then our conscious mind is no longer consulted before our mind reacts. Unless the conditions are right, our conscious mind is not consulted after either. So that care to discern, those switches to different thought, must be part of our rehearsed thinking habits, or they will not always happen when it matters.

In general, a mental arrangement that normally works as desired, tolerated, or expected can often be liable to go wrong when one or more cognemes is added or removed. If the mind is not aware of that possibility or cannot internally prevent it, that is an opening - things can slip through with bad consequences.

[mentalisttraceur]

I had a realization just the other week, another possible consequence of the empiric definition of logic: a mind that is omniscient across time, or at least can always perceive infinitely into the future, might not even be able to have logic.

It would have no need or incentive for developing cognition that can predict correctly if it can already just know what will happen.

So if the mind's capabilities are sufficiently naturally selected in the right ways, then this suggests that the mind simply will not develop the ability to think logically, unless there's some other way for logical thinking to be useful that perfect knowledge of the future without logic doesn't already cover well-enough.

And I'm not sure that you can implement cognition that learns without something that amounts to the effect of natural selection on cognemes. That needs to be thought through more, but learning seems to entail having some sort of selective pressure inside the mind favoring cognition based on some measure of "correctness" or "truth", and efficiency or time boundedness thereof. I haven't yet thought through if that selective pressure is sufficient for this specific effect on mind capabilities though.

What's very interesting is that if I'm on the right track here, then this (combined with using the the anthropic principle for possibility elimination), implies that perhaps we are in a universe where perceiving indefinitely into the future must be impossible, or at least costlier than logic to predict just as far and just as well, at least in so far as it effects the selective pressures that shaped our brains.

[mentalisttraceur]

For my new followers, an overview of what I consider my most important topic areas:

Cognition habitualization: Most cognition can be trained by repetition to be more automatic, less effortful, and eventually less conscious; or untrained by finding ways to either avoid triggering that cognition or to redirect to other cognition, until it becomes less automatic and more effortful. This is well-established in psychology and neurology, but I figured out how to do this on my own as a little kid, so I've seen first-hand that you can take this much further than most people realize. Brains do what they train.

Cognemes: Cognition can be thought of as being made up of pieces that are subject to natural selection, and this is useful for understanding some things in the world and in human experience. There are at least three interacting dimensions of natural selection which are very relevant to us - in biology, in culture, and in the minds - each with different "selectable pieces" - genes, memes, and cognemes.

Cognetic openings: We can end up with arrangements of habitualized cognition that seem fine to us and normally work fine and as expected, but which can go wrong or motivate the wrong action with just a tiny change - such as a single cogneme added or missing at the wrong moment. A lot of problems in people's cognition and choices can be seen as cognetic openings to some cognemes causing bad consequences, and some problems can only be seen with that lens.

The empiric definition of logic: Philosophically, this elegantly frees logic from depending on any arbitrary axioms, and enables what I think is the most sound and justifiable "from first principles" build up to any other right idea. Psychologically, it reveals that our sense of logic is also at least in large part empirically developed. For me, this idea was key to even conceiving of countless other ideas.

Prediction trees: Our brains - and in particular almost all of our emotions, motivations, and reactions - can be explained almost entirely by the brain learning to treat some things as predictive of other things, and then reacting to those predictions preemptively and automatically, before or entirely without any critical review of whether those predictions are accurate in the current situation.

Decision shares: If we iteratively improve on democratic governance and try to solve problems with it, or optimize a benevolent non-democracy that cares about its people, or take money further as a mechanism of coordination, they all converge on a way that we can do thoroughly better as a society in almost every way imaginable.

Idea-fitting, possibility collapse, and "thinking in probabilities": Good thinking involves trying on many different possible interpretations, and ideally striving to hold as many conceivable truths or versions of events in mind, giving them more or less weight based on how much contrivance and unfounded assumptions it would take to make it consistent with your experienced universe so far. Most people miss many, many possibilities in their idea-fitting, and there are lots of thinking techniques and improvements that can make thinking of so many possibilities more practical.

[mentalisttraceur]

I keep having this thought... The latest generations of people are raising themselves far more than any other generation in human history, and I don't mean in a good way.

Ever since the internet became a mainstream thing that all the kids are on, it has become a lot easier for communities to form where most people are kids, teenagers, or in their early twenties, and this creates echo chambers of inexperience.

There is just so much that experience - both personal and encoded in cultural habits - teaches you or provides solutions for. Problems you wouldn't even know needed solving until you've had years of suffering them to identify and word and learn how to explain. Solutions and ideas that would take years of dedicated work if starting from scratch to figure out.

And in one sharp transition, in the span of essentially just a handful of years, humanity took this huge step - with all the foresight of an infant's first steps - that effectively chopped off a significant chunk of the latest generations from any wisdom that the prior generations could have offered. Not literally chopped off - if you take proactive steps your access isn't blocked - but you have to know to take them.

Because people from significantly different age ranges rarely hang out in the same digital spaces nowadays. How cross-generational is TikTok, for example? How many of us just stick to the social networks that we got into during our teenage years? How many people deliberately check out whatever the kids these days are into to cross-pollinate memes and cognemes across generational boundaries?

And of course, once a space is already dominated by a single age group, or more generally by a group of ideas and norms, it's harder for members of any other group to convey any wisdom that their ways or experiences might have to offer. Because the more a space becomes homogenous like that, the easier it is to end up with ever more polarizing narratives about how everyone who disagrees is evil, actually.

Cultures had a less severe version of this happen much more gradually, when humanity transitioned to increasingly segregating kids and teenagers into their own world - schools and activities where kids interact mostly with age peers, unlike earlier societies where kids would be mixing with people of all ages, as apprentices, helpers, or full fledged participants in adult work.

That was no where near as sudden, but maybe looking closely at that can teach us something about what's happening now.

Anyway, sometimes nowadays I look at what I see in social media and popular culture, particularly when it comes to ethical and political discourse, and it uncomfortably resembles children confidently teaching other children some naive simplistic model of the world or morality, except on a massive scale and real lives and consequences on the line. And this worries me.

Other times it encouragingly resembles young independent thinkers realizing things on their own, figuring things out that even a lot of adults never realize. But at the same time it's a little worrying because I don't see those epiphanies in the discourse at large as much as I see them in the spaces like Tumblr which have had a decade to "grow up", have had a "popularity die-off" in the meantime, and now have an accordingly older and less mainstream user base.

I hope the future looks like the internet expediting how quickly these epiphanies ripple, instead of just many separate generational bubbles having to relearn the same basic lessons over and over again, except slower because of the reduced cross-generational socializing.

[mentalisttraceur]

Epiphanous Habitualization

One phenomenon that might seem counterintuitive to the idea of cognition habitualization, is that sometimes, it only takes one epiphany to lastingly and thoroughly habitualize some cognition flow.

The way this manifests for me is that often, if I just realize something new, like a flaw in my reasoning or interpretations or a new angle of compassion or empathy activation, that is enough to make an immediate and lasting change in how I react or feel or think.

I think what is really happening is that an overwhelming majority of the cognition making up or relevant to the new change has already been habitualized, and only one final cogneme is missing or just on the verge of habitualization. Once a mind is in that state, the moment that cogneme is activated in it, it bridges what is already there into one path that can now be traversed automatically as a whole. The result can look like sudden and even very deep habitualization, even if it actually took a long time to habitualize behind the scenes.

So I suspect this is likelier in minds who are already skilled at mind work or have significant plasticity.

[mentalisttraceur]

Someone replied to the language holes post:

what’s an example of a big language hole?

How big you looking for?

I mean mostly I'm just giving a name to something very mundane: all the ideas that any given language can't express except without going past some threshold of convoluted wording to invoke the right idea.

The size of the hole is roughly measured by how much you have to say to explain it to people, or perhaps more accurately, how much cognition that is not invokable by the language is needed to get there. How much language-free thought do you need.

To be clear, a bigger language hole is not just one thing, it is many many things - the bigger the hole, the more stuff there is in it that might be usefully given words. (I mean, the space of possible cognition isn't quite neatly countable, trying to partition it into discrete things is like trying to think of flowing water as discrete things, but there is more of it in bigger language holes).

So, as you probably realize, it is inherently harder to give examples of bigger language holes. The bigger the language hole the more struggle and time and words it would take for me to convey it (or even remember it or reconstruct it, because I have no words to index into them or to hold onto while there), and the less certainty I would have that I even conveyed it.

A manageably small one that's still bigger than the "food : feed :: drink : ______" example is the unity of experiencer experiencing experiences, which took a small post, and like most of my small posts, it maybe doesn't actually explain the idea it's trying to word to most people, so that language hole is maybe bigger than I think it is.

I obviously also think the preference-pleasure root is also naming an idea in a language hole, of probably approximately similar or slightly bigger size, and that's still a struggle for me to actually reliably conjure in my mind, let alone convey to others.

There are language holes in the space covered by the "empiric definition of logic" umbrella. Not sure how big they are, because there's more entangled here than just a lack of common language.

You can probably witness a big gaping language hole, or several, if you take a lot of LSD at once. Just beware, if you haven't done it before, or don't have experience with similar hallucinogens: LSD at high doses is powerful, harsh, and it will temporarily disassemble your mind in ways you probably didn't know possible, while giving you brand new qualia that you have never had before.

I guess the most important really big language hole I can think of is in mind work. A huge part of my blog for the last several years, and my life for much longer than that, has been trying to give names to experiences, mechanisms, patterns, techniques, or models of cognition. As just one example, it took me most of my life so far and many failed attempts to figure out how to create a tolerably decent wording and cognition funnel for the idea of "cognetic openings" in a few paragraphs, and I still need to rely on a redefinition of cognition and the idea of "cognemes" to do it.

[mentalisttraceur]

The whole "why did humanity luck out and find so many conveniently domesticatable animals in Eurasia but not in Africa?" is making a lot more sense to me now that I just looked at it from the lens of humans as the typical invasive species that upon moving out of its tougher ecological niche just demolishes the unprepared competition (the same way cats and rats do).

The stuff in Africa evolved in sync with us. I would not be surprised if, for example, zebras are more aggressive because early proto humanity tried to pursuit predate them, and that would select for the zebras that would viciously attack instead of fleeing.

Once you have the right tools and coordination, suddenly an aggressive prey animal can be brought down too, but not early on, and not for a long time. Everything around where humanity started had the evolutionary advantage of evolving with us, compensating for every tiny gain we made with their own adaptations.

But the life over in Eurasia? It didn't get that. Most of it was there long before we were. It didn't have pressures for the same adaptations. When humans spread into that area, that life was unprepared to resist us as predators and threats, and as a lucky consequence, it was more receptive to the interactions needed for domestication.

We can also look at it in terms of exploitable complexity: there is a constantly building reservoir of exploitable complexity in the form of these animals and their minds and their ability to do or produce things. (See my generalization of @togglesbloggle's "Toggle's Law" for more on that idea.)

So to domesticate an animal is in a way to predate or parasite on it at that higher level, to consume its outputs for our gain. It is technically a symbiotic relationship, but hardly a symmetrical one. The fact that where humanity started had so few animals receptive to domestication is evidence of this: life literally evolved to not let it happen.

(You can make an argument from suboptimal equilibria, that the domesticated animal is truly better off but that it takes a transition initially uphill on the benefit gradient to get to that state. But that's not the level evolution cares about. Animals and minds are not selectable pieces. Genes and memes and cognemes are, and a typical wild animal consists of an overwhelming amount of genes and memes and cognemes whose persistence and duplication would not be benefitted by domestication.)

So your local ecosystem evolves to resist domestication like any other predation or parasitation, and then suddenly you, humanity, the species forged and sharpened in the crucible of the ecosystem that has been adapting to match you, spill out into a space with animals who were never evolved to defend against that. Now domestication itself is possible, and evolves to fill this empty niche that opens up, mutating from the genes and memes and cognemes that are closest to causing the traits and behaviors and cognition that is conducive to domestication.

So that seems plausible to me. The other continents have other compounding factors, so I'm not even trying to tackle those for now. Just sorta thinking out loud because I just thought of this idea.