You walk into a store that sells two identical candies. Would you buy the candy that ordinarily costs $2 at 50% off (for one dollar), or the candy that ordinarily costs 10$ for 80% off (for one dollar)? 

In an isomorphic situation, sparrows preferred the latter deal.

http://users.ox.ac.uk/~kgroup/publications/pdf/kacelnik_marsh_2002_animbehav_costs.pdf

New to LessWrong?

New Comment
10 comments, sorted by Click to highlight new comments since: Today at 9:32 AM

the candy that ordinarily costs 10$ for 80% off (for one dollar)?

This seems, um, off. Should it be "90% off", or alternatively, "(for two dollars)"?

Anyway, if the price is the same and I believe the candies are identical, I don't see why not to go for the ordinarily-more-expensive one; presumably the store owner knows more about the quality of his goods than I do.

That's not really a sunk cost fallacy. It's an implicit judgement of value based on cost. I'd definitely rather get the candy that is usually 10 dollars even if I thought they were pretty much the same because I would assume the person costing the candies knows why one is worth 8 dollars more normally than the other.

Agreed that it's not quite the same - that's why I had to say "sunk-cost-like".

I'd definitely rather get the candy that is usually 10 dollars even if I thought they were pretty much the same because I would assume the person costing the candies knows why one is worth 8 dollars more normally than the other.

Perhaps, but the sparrows are (presumably) unaware that any sort of transaction is being made. Although I suppose it's possible that they dimly perceive that the apparatus which they are interacting with is somewhat agent-like, the same way we might view a vending machine as more agent=like than a plant even if we didn't know that someone had fashioned it with the intention of trading with us.

But even so - they've already sampled the rewards that each choice provides during the training tasks. So if you've tasted the candies and know what you are getting, why should it matter how much the other person thinks they should cost?

I that case it's probably a difference in being able to know how much effort you've put in versus knowing how much reward you got, which is a problem.

Bird geek mode on: the paper cited is about starlings, specifically the European Starling, Sturnus vulgaris, not sparrows. They are not the same thing. Starlings (members of the Sturnidae family) are probably one of the more intelligent passerine families, though likely not as intelligent as the Corvids (Jays and Crows). Sparrows, both Old World (Passeridae) and New World (Emberizidae), are not generally known for their intelligence.

Oh...thanks! I seem to have made a rather large number of tiny factual errors in posting this link, for some reason.

Of course, it doesn't actually matter which one is chosen in this scenario. They could easily have gone one step further to see if they could use this effect to make the sparrows prefer the worse of the two deals - that would be much more interesting.

It's not sunk cost, it's anchoring.

This isn't quite sunk cost fallacy. This is more common to the behavior (which unfortunately I don't know the name) where people make decisions to save money based more on percentage than absolute total, e.g.being willing to drive from one gas station to another to save a small amount on gas cost but not driving from one store to another to buy an expensive piece of electronics for less when the percentage difference is small even when the total amount difference is high. Does anyone know the name of this bias?

I don't know any name for this, but I like to just refer to it as "thinking logarithmically".