Tons of experiments have shown that individuals violate basic axioms of choice (think transitivity, independence, etc.) Our question: is this a true preference? Or is it a “mistake?” It could be that people would PREFER to follow these axioms but they're hard to implement. 2/
-
-
Prikaži ovu nit
-
Ours is the first incentivized experiment testing between these two interpretations of violations. We directly elicit preferences over axioms and give subjects the opportunity to re-evaluate their lottery choices when they conflict with an axiom they’ve said to prefer. 3/
Prikaži ovu nit -
If the axiom is normative, individuals will change their lottery choices to follow it. If it's not, they might infer from their lottery choices that the axiom isn’t so great after all and abandon it. We allow for both of these, but don't force reconciliation. 4/
Prikaži ovu nit -
Individuals do say they want choices to satisfy the axioms. They reconcile ~50% of cases when choices violate the axiom, and almost always do so by changing lottery choices! This means that many classic violations in the literature might be better thought of as “mistakes.” 5/pic.twitter.com/Xgqii6GqOh
Prikaži ovu nit -
As always, Savage said it first. “In reversing my preference... I have corrected an error. There is, of course, an important sense in which preferences, being entirely subjective, cannot be in error; but in a different, more subtle sense they can be.” Leonard Savage (1954) 6/
Prikaži ovu nit -
Our methodology can be applied to many other domains. Eliciting social choice rules, fairness preferences, dynamic decision making principles, strategies in games, etc. We hope this will lead to more "procedural" experiments looking at how people make their decisions.
Prikaži ovu nit -
More results in the paper, and much more work to be done. Hopefully we’ll have some new stuff to share soon. Let us know what you think! Full WP can be found here
https://kirbyknielsen.com/wp-content/uploads/kirby/AreAxiomsNormative.pdf …
/nPrikaži ovu nit
Kraj razgovora
Novi razgovor -
-
-
Hi
@KirbyKNielsen really interesting paper! You mention that the order of options is pre-registered. I was wondering where you pre-registered it? (I poked around a few places and didn't find it) There's some overlap in an experiment for which I am working on a PAP. -
Whoops, sorry for the confusion. The experiment was not pre-registered. I meant that we randomized the order once ex-ante and used that order in all sessions. I need to change the wording; thanks for pointing it out! Would be happy to hear more about your project!
Kraj razgovora
Novi razgovor -
-
-
Sounds very interesting, I have this feeling that we gonna cite your paper in our own work :)
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.