I am unable to reconcile the fact that SBF’s funds collapsed and he is sentenced to a decades long prison term and the idea that he somehow was “careful” and “smart” about how he modeled people as PDFs. At first blush this seems to be an argument to avoid using this model to interact with people, since in purely pragmatic terms this one example doesn’t work.
That being said, yes, broadly speaking when we build a mental model of another person it is a probabilistic assessment of that person. The research of Robin Dunbar is particularly interesting in this regard: larger cortices in social animals correlates to social network size. But are these “Bayseian” or “rule-based” is another argument all together. When I say: “Bob never drinks (alcohol)” is that a hard and fast rule or a probabilistic statement?
How do you think, from a purely pragmatic level, SBF could have modeled more accurately and effectively so that he would be better positioned today to have his funds still operating and not be in prison? How does this directly relate to the model of “people as PDFs” and why is it a more effective way of predicting people’s behaviors than, say, Myer-Briggs or even star signs?
I am unable to reconcile the fact that SBF’s funds collapsed and he is sentenced to a decades long prison term and the idea that he somehow was “careful” and “smart” about how he modeled people as PDFs. At first blush this seems to be an argument to avoid using this model to interact with people, since in purely pragmatic terms this one example doesn’t work.
That being said, yes, broadly speaking when we build a mental model of another person it is a probabilistic assessment of that person. The research of Robin Dunbar is particularly interesting in this regard: larger cortices in social animals correlates to social network size. But are these “Bayseian” or “rule-based” is another argument all together. When I say: “Bob never drinks (alcohol)” is that a hard and fast rule or a probabilistic statement?
How do you think, from a purely pragmatic level, SBF could have modeled more accurately and effectively so that he would be better positioned today to have his funds still operating and not be in prison? How does this directly relate to the model of “people as PDFs” and why is it a more effective way of predicting people’s behaviors than, say, Myer-Briggs or even star signs?