The response I would make is that you’re not buying a lottery ticket; you’re buying a social interaction ticket, and that can be rational. Given the “waste of hope” message, though, hypothetical you should think long and hard about whether that social interaction is positive.
However, we assume that the social interaction itself isn’t enough to justify my ticket. Let’s say it’s just a warm greeting and a short small talk with someone I find sympathetic, but probably won’t play a decisive role in my future. And I’m buying the ticket, because I rationally know that I might win the lottery (despite that I find it so unlikely that I don’t actually expect it). I have only included the social interaction to offset the wasted time.
The first basically says “let’s come up with an imaginary score that we give to potential futures, such that that we can do expected value calculations over probabilistic gambles between those potential futures, and the EV calculation will be correct by definition.” This is generally recommended as a good way to make decisions (at least, it clarifies what the difficult parts are- but beyond focusing your attention may not make them easier).
The second asks how various descriptive biases coexist, and comes up with a model that has three deviations from risk-neutral VNM but fits how many humans actually behave.
The response I would make is that you’re not buying a lottery ticket; you’re buying a social interaction ticket, and that can be rational. Given the “waste of hope” message, though, hypothetical you should think long and hard about whether that social interaction is positive.
However, we assume that the social interaction itself isn’t enough to justify my ticket. Let’s say it’s just a warm greeting and a short small talk with someone I find sympathetic, but probably won’t play a decisive role in my future. And I’m buying the ticket, because I rationally know that I might win the lottery (despite that I find it so unlikely that I don’t actually expect it). I have only included the social interaction to offset the wasted time.
Ah, now I see what you’re trying to get at.
You might be interested in reading about VNM utility maximization and prospect theory.
The first basically says “let’s come up with an imaginary score that we give to potential futures, such that that we can do expected value calculations over probabilistic gambles between those potential futures, and the EV calculation will be correct by definition.” This is generally recommended as a good way to make decisions (at least, it clarifies what the difficult parts are- but beyond focusing your attention may not make them easier).
The second asks how various descriptive biases coexist, and comes up with a model that has three deviations from risk-neutral VNM but fits how many humans actually behave.