A lot of relatively weighty decisions wind up being made for trivial reasons simply because all of the non-trivial factors cancel each other out—for instance, if I were trying to decide whether to go into ethics or metaphysics (a choice with long-term career impact, assuming I get to be a professor one day) and I didn’t find myself strongly preferring one over the other, I could see myself picking one for a silly reason. If my name were Ethel, which it is not, and I liked the sound of “Ethel the ethicist”, that might tip the balance. Either that, or contemplating that choice would throw into sharp relief something I’d been overlooking in favor of metaphysics. But if there is no such factor, then why not choose on the basis of “Ethel the ethicist” sounding nice? It arguably makes slightly more sense than flipping a coin.
It seems like utility calculations should cancel out to within such a small margin as U(good name) - U(bad name) only very very rarely, especially when there are many different independent considerations. I worry that people may have overactive indifference detectors, and could improve their decision algorithms by learning tricks to distinguish the real slight preferences/estimates underlying what feels to them like very precise indifference but isn’t.
I agree that it may plausibly be argued that the difference should rarely fall into the small margin: U(good name) - U(bad name) (up to varying priors, utility functions, …). However, should people calculate to the point that they can resolve differences of that order of magnitude? A fast and dirty heuristic may be the way to go practically speaking; the difference in utility would be less than the utility lost in calculating it.
If you haven’t already, read the part of the paper where they talk about hardware and roofing stores. They ran some clever analyses to see whether the effect was caused by a love of alliteration (for example someone named Herman decides to go into hardware so he can call his store Herman’s Hardware) and the results suggested this wasn’t the explanation.
Right. Given the prior knowledge, silly reasons may be independent of the serious consequences, even if in the end, when you learn more, they become dependent. High-impact decisions can be indistinguishable from each other at the stage where they are made, so that silly reasons don’t (anti)correlate with serious reasons. Silly reasons are not silly because they are anti-intelligent, they are silly because they are irrelevant.
A lot of relatively weighty decisions wind up being made for trivial reasons simply because all of the non-trivial factors cancel each other out—for instance, if I were trying to decide whether to go into ethics or metaphysics (a choice with long-term career impact, assuming I get to be a professor one day) and I didn’t find myself strongly preferring one over the other, I could see myself picking one for a silly reason. If my name were Ethel, which it is not, and I liked the sound of “Ethel the ethicist”, that might tip the balance. Either that, or contemplating that choice would throw into sharp relief something I’d been overlooking in favor of metaphysics. But if there is no such factor, then why not choose on the basis of “Ethel the ethicist” sounding nice? It arguably makes slightly more sense than flipping a coin.
It seems like utility calculations should cancel out to within such a small margin as U(good name) - U(bad name) only very very rarely, especially when there are many different independent considerations. I worry that people may have overactive indifference detectors, and could improve their decision algorithms by learning tricks to distinguish the real slight preferences/estimates underlying what feels to them like very precise indifference but isn’t.
I agree that it may plausibly be argued that the difference should rarely fall into the small margin: U(good name) - U(bad name) (up to varying priors, utility functions, …). However, should people calculate to the point that they can resolve differences of that order of magnitude? A fast and dirty heuristic may be the way to go practically speaking; the difference in utility would be less than the utility lost in calculating it.
If you haven’t already, read the part of the paper where they talk about hardware and roofing stores. They ran some clever analyses to see whether the effect was caused by a love of alliteration (for example someone named Herman decides to go into hardware so he can call his store Herman’s Hardware) and the results suggested this wasn’t the explanation.
Right. Given the prior knowledge, silly reasons may be independent of the serious consequences, even if in the end, when you learn more, they become dependent. High-impact decisions can be indistinguishable from each other at the stage where they are made, so that silly reasons don’t (anti)correlate with serious reasons. Silly reasons are not silly because they are anti-intelligent, they are silly because they are irrelevant.