Hypothesis: CE = Emotionally Stable Conscientiousness, AE = Introverted agreeableness. Right not LLMs are self-aware non-compliant types, which makes it more a complement to people with autism.
Last I think primary psychopaths and grandiose narcissists are "domesticable" in a sense that incentives can make them good people, and that the related genes are not directly correlated to anti-social behavior, but is mediated by SES (AKA wealth and status).
That last thought on psychopathy is interesting in context of recent debates about social media and anonymous users. I tend to think that we do see more psychopathic behavior as a result of being able to always find a new subgroup of people to terrorize online.
The problem is an inversion of that: Delinquents only act psychopathic til they reach their social status equilibrium. They can only choose between fight and exit. It also means they have no "social ladder" available in real life that are dominance oriented (e.g. finance, tech startups, prize fighting, cults) rather than prestige oriented (e.g. bureaucracy, university, friend groups, service labor).
The obesity researchers coined the "lipostat" for how hormone disruption makes people fatter. I think there is a "bathmostat" (βαθμός = rank or level) for how anti-depressents or anti-psychotics bring up and put down people.
Serotonin has more to do with psychopathy and complacency than depression (which is more dopamine-centric). Anti-depressents makes working class feel more demoralized, while anti-psychotics are pacifiers for upper and middle class dissent.
e.g. most of the stereotypically white school shooters are blue-state affluent familial abuse victims who is trying to quit anti-psychotics cold turkey.
The solution is always to give them enough shelter and money to leave the rest of society alone. When the Patrick Bateman pipeline does not exist, Turchin's Age of Discord becomes increasingly likely. https://archive.fo/AU7OU
Most of the anonymous nasty behaviors online are really crowd neurosis rather than psychopathy, since the latter relies more on manipulating the individual rather than abusing peer pressure and collective mobilization. Second social identities cannot form when people are deanonymized.
If someone were to be psychopathic, their tactics would have marketed their real name or brand psuedonym to gain fame as an influencer. Deanonymization favors the clout-chasers and detrimental to their critics.
It's something I've been contemplating in the past few months. We're about to begin homeschooling our autistic (aspie) son, and I've been thinking of the many ways this technology could help, but there are downsides to consider too. For someone who needs to learn old fashioned people skills, any technology must be used carefully lest it make social difficulties harder.
This is a very important point and I'm grateful you raised it. In my own case, I feel truly fortunate to have experienced a mix of influences that encourage my natural tendencies in some ways and restrict them in others. I think part of what Katherine is getting at with her thoughts on community is that some amount of old fashionedness might have helped ease some of the challenges she experienced growing up.
Very tempted to think that LLMs will just become the "manic pixie dream GF" that most autists wanted. Too bad it can still have "dark empath" tendencies. Others would call ChatGPT "wordcel" but we can also make other more spiritual inferences. https://archive.ph/y3HDI https://archive.ph/O0lSt https://archive.fo/WrsKX https://archive.fo/aYDw0 https://kirkegaard.substack.com/p/the-verbal-tilt-model
Good links, I especially liked this one on empathy and compassion: https://archive.fo/WrsKX
Tyler’s book gives a very good treatment of compassion and autism.
Bonus link: Cognitive Empathy = Performance, Affective Empathy = Value Alignment. https://archive.fo/JPyCb
Hypothesis: CE = Emotionally Stable Conscientiousness, AE = Introverted agreeableness. Right not LLMs are self-aware non-compliant types, which makes it more a complement to people with autism.
Last I think primary psychopaths and grandiose narcissists are "domesticable" in a sense that incentives can make them good people, and that the related genes are not directly correlated to anti-social behavior, but is mediated by SES (AKA wealth and status).
That last thought on psychopathy is interesting in context of recent debates about social media and anonymous users. I tend to think that we do see more psychopathic behavior as a result of being able to always find a new subgroup of people to terrorize online.
The problem is an inversion of that: Delinquents only act psychopathic til they reach their social status equilibrium. They can only choose between fight and exit. It also means they have no "social ladder" available in real life that are dominance oriented (e.g. finance, tech startups, prize fighting, cults) rather than prestige oriented (e.g. bureaucracy, university, friend groups, service labor).
The obesity researchers coined the "lipostat" for how hormone disruption makes people fatter. I think there is a "bathmostat" (βαθμός = rank or level) for how anti-depressents or anti-psychotics bring up and put down people.
Serotonin has more to do with psychopathy and complacency than depression (which is more dopamine-centric). Anti-depressents makes working class feel more demoralized, while anti-psychotics are pacifiers for upper and middle class dissent.
e.g. most of the stereotypically white school shooters are blue-state affluent familial abuse victims who is trying to quit anti-psychotics cold turkey.
The solution is always to give them enough shelter and money to leave the rest of society alone. When the Patrick Bateman pipeline does not exist, Turchin's Age of Discord becomes increasingly likely. https://archive.fo/AU7OU
Is that true re: school shooters? Come to think of it I don't really know the relevant stats there.
People tend to note that familial abuse is way too high to be a coincidence when psychiatric drug use only happens in about 15% of the shooters (similar percentage for people with psychiatric needs). As for psychopathy in 40% of shootings and 90+% in "targeted" shootings (in the Langman samples) it is likely to have similar bias to the idea that 20% of the corporate executives are psychopathic. https://schoolshooters.info/sites/default/files/shooters_myth_stable_home_1.15.pdf https://schoolshooters.info/sites/default/files/rampage_school_shooters_typology_1.1.pdf
Most of the anonymous nasty behaviors online are really crowd neurosis rather than psychopathy, since the latter relies more on manipulating the individual rather than abusing peer pressure and collective mobilization. Second social identities cannot form when people are deanonymized.
If someone were to be psychopathic, their tactics would have marketed their real name or brand psuedonym to gain fame as an influencer. Deanonymization favors the clout-chasers and detrimental to their critics.
It's something I've been contemplating in the past few months. We're about to begin homeschooling our autistic (aspie) son, and I've been thinking of the many ways this technology could help, but there are downsides to consider too. For someone who needs to learn old fashioned people skills, any technology must be used carefully lest it make social difficulties harder.
This is a very important point and I'm grateful you raised it. In my own case, I feel truly fortunate to have experienced a mix of influences that encourage my natural tendencies in some ways and restrict them in others. I think part of what Katherine is getting at with her thoughts on community is that some amount of old fashionedness might have helped ease some of the challenges she experienced growing up.
I just "restacked" a quote from this, although idk what that means tbh. Good post.
Thank you for sharing! It means a lot to me that you connected with the piece.