I wonder if its less a matter of neurodivergent brains being literally unable to do this kind of reasoning, but more a difficulty of being able to notice patterns in reactions and behavior to build a useful enough predictive model of people at all.
Zyan
Karma: 8
Looking back on this post having learned about the Free Energy Principle, I wonder if instead of saying “rationality is improving accuracy of beliefs and achieving your values”, one could instead just say “rationality is the minimization of free energy”. The free energy principle, or more specifically the theory of active inference, encodes both goals of “having good models” and “achieving your values” as simply minimizing the difference between observation and prediction, either by making the prediction more accurate through more comprehensive models, or by finding the best policy (action) upon the world that most likely to produce observations more in line with an existing model.
zyansheep’s Shortform
It seems possible to generalize the Free Energy / Predictive Coding Principle from individual brains to markets as a whole, where there is one universal metric that the market attempts to minimize.
https://gist.github.com/zyansheep/1b9dd1b6e4d4186ab4a67250a48e5c3a
This framework may enable finding parallels between neuroscience/AI and economics and potentially even enable ontology transfer between them: Bubbles/Trapped Priors, Market Crash/Crisis of Faith, etc.
I think I agree with Feynman being a straight talker, but I just want to caution on inferences on Feynman from books about Feynman. See:
TLDW: Feynman didn’t actually write any of the books that use his name, and his influence over them is pretty tenuous. (e.g. Surely Your Joking was written by a young friend of Feynman’s, and the book wasn’t written until like at least 10 years after the stories were originally told, allegedly)
I would expect that if five years in a row the “variable income” happens to always be around X, people will start to expect it, will make this expectation a part of their personal finances, and if the next year the income is only X/2, there will be riots in the streets.
Doesn’t this kind of already happen with the stock market? I can imagine rioting in the case where people loose half of their investments overnight, but I think it’d be different for a cut in dividends since any economic downturn that reduces land values (and thus land taxes) would correspondingly reduce rents and thus people probably wouldn’t be left immediately destitute if dividends cut in half or something.
“Indeed,” snapped the other boy. His stern face lightened somewhat. “Tell me, what House do you think you might be sorted into? I’m bound for Slytherin House, of course, like my father Lucius before me. And for you, I’d guess House Hufflepuff, or possibly House Elf.”
Draco Malfoy mentioned it previously :)
Modern multimodal LLMs are trained not just on text data, but also on images and video. In the finetuning stage for reasoning models, they are trained not to predict human reasoning, but to do reasoning themselves to accomplish a binary goal. Not sure about exact percentages but in terms of capabilities, they can recognize images very well and generate very realistic images and videos, to the point where many people can’t tell them apart.