HN Reader

This article approaches human psychology from the perspective that, we are all neural networks and our output (actions) are all a learned function of our inputs (experiences).

This is a common (and convenient) perspective, especially among engineers, but doesn't reflect reality particularly well. We know large swathes of a person's personality is directly linked to their genetics.

The article extrapolates this neural network perspective onto other topics like, mental disorders and depression. The solution is made clear then - just learn how to not be mentally ill! Again, convenient. But not really reflective of reality.

I agree that it can be helpful to think of identity as a trajectory shaped by interactions along the way. However, we also continually shape our environments in large and small ways. TFA ignores this completely. Can this be effectively modeled in RL?

Over 130 years ago, Dewey [1] criticized the model of psychology which looked at human behavior in terms of stimulus -> internal processing -> response. Stimuli don't just come to us; we seek them out and modify the world around us to cause them to occur. Dewey and other pragmatists proposed reframing stimulus/response in terms of "acts" or "habits," or changes to the unified agent+environment. Popper was getting at the same entanglement of agent and environment in "Three Worlds" and Simon in "The sciences of the artificial."

I see RL as an elaboration of the stimulus/response paradigm: the agent is discrete from the environment. Does RL work well in an environment like Minecraft, where the real game is modifying the relationship between actions and future states? What about in contexts like Twitter, where you're also modifying the value function (e.g. by cultivating audiences or by participating in a thread in a way which conditions the value function of future responses)?

[1] https://plato.stanford.edu/entries/dewey/#ReflArcDeweRecoPsy...

Personality is ~70% determined by genetics, not life experience.[0]

I’m surprised that someone interested enough in the topic to write such a long post wouldn’t put the time in to do a cursory dive into personality psychology. I’m going to assume that the author has a similar definition of personality to mainstream psychology, but if so, they are ignoring accepted studies and evidence that make it pretty clear that personality is not learned through conditioning like AI.

0: https://www.themantic-education.com/ibpsych/2019/02/11/key-s...

> A common mistake in life is to let your personality basin solidify too early. Your parents and schooling environment have a disproportionately large influence on who you become as an adolescent.

> But as soon as you gain the freedom to act independently as an adult, it’s usually a good idea to force yourself to try as many new things as you can, including moving cities (or countries!) and considering drastically different lines of work. ...

Oh dear, I'm beginning to fear that the author's personality has been captured by global capital...

And what if it's personality capture all the way down, i.e. that you've got to be personality captured by someone? In that case, the closest you can get to a choice is whether it's your parents, religion, or someone/something else. While the integrity of your parents may vary, there is a subjective argument that they've got a better incentive to steer you into an optimal basin than anybody, relatively speaking.

As someone with a lot of research experience in this area, I was expecting something more fluffy but this was actually pretty good.

There's a lot of research into some of these ideas at the moment. The terms they use aren't necessarily the same but many similar ideas. I think for the most part, the evidence in support of many of them is fairly weak but at the same time many of these ideas are much harder to test well than it might seem at first. I give a certain amount of pass to people trying to test them because in this area, trying to pin down something often is a bit like trying to study an individual cloud: you can kind of see it there, but if you were to try to measure its boundaries and dynamics, it would be harder to do than it might seem at first glance, and you'd end up more easily making very general observations about it than you might like.

One thought I had when reading it is that people's environments are much more stable than is usually recognized. The piece acknowledges this somewhat, but I think it's more of an issue than most like to admit. Even when someone tries to change it, it can be difficult, because assets and SES can be difficult to change, other people resist it due to their own incentives like the essay points out, and even when other people don't really care much they often will resist it unintentionally due to schemas about personality change and so forth. "Once an X, always an X" regardless of whether you're talking about vocation, career, social characteristics, whatever — even though that statement isn't actually true beyond some kind of general sense of it. Or they just are used to seeing someone in a particular setting and so don't see them in another.

Another issue that's maybe murkier is the essay is a bit loose about person characteristics, even at a given point in time, versus situation characteristics. I don't know that it affects the arguments very much at all, the points still stand, but it sometimes drifts into talking about "personality" when I think it really means something more relational, like "role" or "interactional pattern" or something like that.

One of the more depressing things of the AI boom is watching engineers and “atheists” get hoodwinked by mystic gibberish like this blog. There is nothing here but astrology: even Myers-Briggs is more scientific.

I think 30% of atheists bothered to think carefully about the Flying Spaghetti Monster and recognized Pastafarianism as a funny commentary on epistemic uncertainty. The remaining 70% said “heh, stoopid Christians believe in a spaghetti monster!” and took it as confirmation of their tribe’s superiority.

> Luckily for humans there exist many symbiotic equilibria where multiple parties can find mutually-beneficial feedback loops within the epochs of personality-space. Parent/child relationships

My child is totally "attempting to modify [my] RLHF process so that it results in an agent which is beneficial to them, hopefully resulting in someone who will always give in to their demands". All the damn time!

I see this post getting trashed in the comments for its overly literal interpretation of personality as a reinforcement learning process, but I think there's some value to it as a mental model of how we operate (which is how the opening sentence describes it).

If you can see past some of the more dubious, overly technical-sounding details and treat it as a metaphor, there is for sure a "behavioral landscape" that we all find ourselves in, filled with local minimal, attractors/basins and steep hills to climb to change our own behaviors.

Thinking about where you are and where you want to be in the behavior landscape can be a useful mental model. Habit changes like exercise and healthy eating, for example, can be really steep hills to climb (and easy to fall back down), but once you get over the hump, you may find yourself in a much better behavioral valley and wonder how you were stuck in the other place for so long.

> How do you know if you're in the "right" personality basin

I'm not so sure "right" is the right frame here. I like the multi-dimensional viewpoint you take. My experience would be a healthy personality would be one capable of adaptation in service of your interests at any times. It's dynamic.

This reminds me of Bob Kegan's stage of adult development. Initially, most of us leave adolescence at the "socialized" stage of development, ie our personality basin has primarily been determined by the external factors of our upbringing and environment.

From there, if we choose to continue developing, we eventually reach a "self-authored" mind, where we have transcended our socialized basin in favor of a self-defined and created personality structure, until ultimately, for those who continue evolving, we reach a "self-transforming" mind, or a mind capable of transforming itself.

I like the simplicity of the model, and I also think it reduces personality to an unnecessarily static entity. Things like internal family system/parts work also demonstrate that our personality is not a singular entity, it is represented by a whole slew of parts that show up in different ways and different contexts! I think the broad strokes of it still hold, and also think there are many additional approaches to truth and the awakening path, lying in parts work, embodied transformation, and whole bunch of other experimental modalities (thought perhaps that's just my personality speaking...)

This is what happens when an engineer tries to apply mathematical models to entirely different fields where they have no applicability. Reducing human personality to machine learning concepts like 'gradient updates' misses the fundamental complexity of human psychology and consciousness.

>> Most personality changes are unconscious

That is because most people are not consciously attempting to become better people for the betterment of those around them (which helps their own happiness, too, due to the nature of our karmic universe). Most people are simply acting out of the selfishness to have their own desires fulfilled, with varying amounts of concern for the consequences to those around them.

A person can undertake self-evolution in any direction they choose; it is always our choice, except in the extremely rare cases where a person is physically damaged. Most people have the power to change, though it is difficult for us all. The universe does help those of us who seek to do so for the benefit of others.

What makes human beings unique is that we have both the ability to self-evolve our attitudes and behaviors, and the tools to do so, our mind and conscience.

>> Personality Capture

A person who is not undergoing conscious-self-evolution is susceptible to being influenced (and even overwhelmed) by a forceful personality that caters to their desire-seeking. That is how dictators have always risen to power: they seek a loyal army of folks enamored with the leader's promises to make the in-group's lives better. Those folks never seek to make life better for ALL people, because helping out-group members requires generosity, which usually requires making some level of selfless sacrifices of resources.

As always, the strife is between selfish callousness and selfless care. Compassion for all our fellow human beings is the nature of being a humanitarian, that is to say: being the best of what we can be, for the benefit of the entire human race. And it all starts with each and every one of us.

I think the article is missing some key aspect of personality forming : Maslow's hierarchy of needs[1]

Personality and its development is hugely dependent on which needs are or will be currently fulfilled or not.

Attention economics is able to impact you negatively on low level of the pyramids, notably due to its impact on sleep. It can also shift your priority towards less essential needs than the one you should be working on. And it's often myopic, missing totally some aspects due to hyper-focusing.

It's also able to impact you by impacting those near you, that's what social networks are for. Developing a support structure whether family or friends is a double edge sword because you indirectly become as weak as the most vulnerable member of the group, or group may explode.

The economy also apply pressure on basic level needs, like shelter, heat, air and water(when polluted), and safety, which probably contribute to shape the personality and are basins which are also hard to get out of ("you can take the girl out of the trailer park, but you can't take the trailer park out of the girl").

[1] https://en.wikipedia.org/wiki/Maslow%27s_hierarchy_of_needs

I feel like some people are reading this post too literally.

An infinite number of factors go into developing a personality, this covers a lot of the big ones.

---

Lately I've been thinking about the windows of plasticity and why people change beliefs as adults.

The sad fact is that a lot of people have a lot of very wrong and bad beliefs, and unfortunately most of them are already adults, so you don't get to discipline those beliefs out of them (I hope you can read this as tongue in cheek); You will never get mad enough at a person to fix them. Anger is motivating, but you don't get to pick the direction.

As I understand it, psychologists believe that parents and the environment of a person's youth set a lot of their basic beliefs about the world, but it is their friendships in adulthood that most determine their value system - you want what is best for your friends (and yourself).

---

To me this also ties into evaluating the actions of historical figures; people seem to get hung up on very flat depictions - was it ok that a person who did good things also did bad things? Well, they are a whole-ass person, raised in a different time and place. They didn't choose when and where and by whomst they were raised. They had some level of choice in their friend group, but that is also constrained by time and place.

I feel that you can judge people and actions, while also allowing space for humanity and personal stories; but that does take a lot of time and emotional work. It is much easier to just choose one side of the coin or the other, face or heel.

> Maybe you were born tall and attractive and then this led you to engage in a lot of athletic activities and socialization, and at the end of all of the positive feedback you have ended up with a jock personality that goes on to become a professional football player.

Is that really that tall and attractive guys want to become football players? I always that football players attract the same stereotypes as police officers, big and stupid.

In Europe only the most stupid folks want to become soccer players. Even if they'll end up filthy rich, with lots of tattoos and horrible haircuts.

Reading computer scientists' takes on psychology or social sciences after taking 1 (one) 101 level class and then reinventing the wheel on topics that have been researched for decades is just grating. Where does this come from? We get it, you like to think a lot, but if you're as smart as claim you would've realized you can't solve a topic by just thinking really hard and long about it alone in your room. Gurwinder Bhogal is one of the other guys repeatedly falling victim to it

I enjoy how people are dunking on this by saying “omg this is what happens when engineers dare to have thoughts on other topics” when this is very similar to the theory for CBT in psychology

>Your personality is formed by a process conceptually similar to RLHF. You are first born with a set of traits in a given environment. After this, you perform many interactions with your environment. If an interaction goes well, you’re likely to do it more often, and if it goes poorly, you’ll probably do less of it.

first off 0 evidence presented, second off what about the kids that grew up stealing food in concentration camps or due to abusive parents. Do they grow up to be liars and thieves? Nope. What about all the kids that get nothing but positive vibes and turn into total arseholes...

I gave up reading immediately - just dumb.

This reminds me of "unstable orbits in the space of lives", a short story by Greg Egan.

> If you were born tall and with a commanding voice

I've always assumed that a "commanding voice" is not something one is born with, but something one develops over time.

It's nice to see that Rationalists have reinvented Maimonidean virtue ethics. The idea that humans personality is maximally pliable, and this is metaethical grounding for the concept of moral responsibility is an extreme on a spectrum. It has some inspirational value, but I've never found it especially compelling.

Also, the fact that this article does not mention the Big Five once really makes me feel like the author is trying to reinvent the wheel but has never looked at a wheel before. Despite its flaws (and the broader methodological critiques you could level at personality science as a whole), it is the most scientifically grounded model of human temperament that we have right now. But why start with the latest science? That would involve leaving your bubble, which is a major no no.

Sorry for the snark, but this is scientific reasoning as cargo cult at its worst.

The curious thing is that I have never met two people whose personalities were exactly alike.

Reverse settings

> meditation, drug usage, trauma, religious events, love, gambling, and sex.

This is why joining a psychedelic sex cult is such an effective life-choice. I don't mean that sarcastically.

> This is why techniques like nonviolent communication, dialectical behavior therapy, and mindfulness have observation and introspection as a core facet, because it’s something that you have to consciously practice to become good at rather than something you’re born with.

I now tend to think that consciously observing and uncovering what you already are is really the start and end of it. One ought to try to concentrate oneself, rather than dilute oneself into something one is not. One might want to be a billionaire technologist, or sports hero, or whatever, and one might even edit oneself into something approximating that (via mentors, diligent study or whatever), but one will remain unfullfilled - for how is it possible to 'lie to' and be 'right with' oneself?

"Although there are many times in life you’ll consciously decide to act in a certain way, this is the exception, not the norm."

IMO, that's the most important idea there. Your personality is what you've created to live among others like you, but as your personality grows, it develops habits that have weight and momentum, and later in life those habits start defining your actions completely, you get progressively smaller windows for true self expression, and your life starts feeling dull and mechanical.

Your attention is the only thing that's truly yours.

wait... is this the same "near" who wrote bsnes, one of the best SNES emulators out there?

The author is clearly taking the time to reflect on the world around them and I genuinely see a curious mind. However, I also see a product that touches on an already established idea and how the gap between this writing and other reflects a gap between "hard" and "soft" sciences.

The idea here is that of habitus. Habitus is an Aristotelian term that was expanded upon by Bourdieu in the field of sociology. It is the way in which people perceive and respond to the world through a durable transposeable disposition, set of skills, symbolic capital and doxa that is shaped by the environment and in particular the material conditions of the individual[1].

Habitus plays a role in how individuals are perceived in ways, that like the author illuminates, can form a virtuous circle re-enforcing disposition, skills, and outlook in a way that can be positive for an individual.

What the author doesn't allude to, and this is where I see a gap between hard and soft sciences and where they would benefit from being able to connect this idea to a broader body of work, is how habitus is reinforced - usually unconsciously - in ways that reproduce class, racial, disability, and gender habitus under the terms laid out by the dominant ideology - that is to say the ideology of the dominant class.

An example in education would be how the education system perceives individuals possessing middle and upper class habitus as being ready and prepared for education, and those who lack that habitus as being lazy, disruptive, or unwilling to learn. On one hand you might be thinking "Of course that's obviously true," and I'd like to take a pause to point out that "obvious truths" are often a signal of our own habitus and should be critiqued as such.

They touched on the concept of reinforcement learning[social systems] acting upon individuals in a way that shapes their habitus, but it's crucial to point out that these reinforcement learning systems aren't free-standing disembodied mechanisms. They are situated in a social landscape and are constituted from of social relations which are themselves a product of economic relations. Furthermore, the systems of reinforcement are self-replicating. They are essentially social quines[2] - or more specifically oroborus programs ie: they plant the seeds of their own replication by encoding those relations into the habitus of individuals.

There's obviously a bunch of writings expanding on the idea of habitus, how it's formed, reinforced in different social arenas, and the effects it has on individuals and groups. I'd expect the author would be interested in soaking up these related perspectives and perhaps you as a reader would be too.

1. Obviously not black and white, there are other factors which can influence habitus - disability is an obvious one, for example.

2. https://en.wikipedia.org/wiki/Quine_(computing)#Ouroboros_pr...

C'mon folks. So many "expert opinions" and erudite references in these comments. The sciences of cognition, neurology, evolutionary psychology etc are all still muddling around trying to figure out how the human mind works. We're learning a lot about possible ways the mind might work from our observations of processes and outcomes of machine learning. It's a cool new paradigm to add to the mix. I really like the framing offered by the author. They're quite upfront about the fact that there's a lot of genetics involved. That all models are wrong but some are useful.

Why all the defensiveness? Whatever genetic aspects of our personalities and behaviours there are - there's still a pretty big component of just learning patterns. Language acquisition is like that. It's an innate thing but the languages we're exposed to as kids shape what patterns of language use we fall into.

Woah really? So wait, if someone is constantly mistreated in all their interactions with women, what kind of personality would develop?