[benrawk]

I know this paper well (and the larger point being made in the paper; I have seen Dr. Gal present on it in seminars). I think the point he is trying to make is that what we really need for understanding of human behavior is good psychological theory of how humans operate. In many cases (loss aversion being one of them, per Dr. Gal), behavioral economics describes the data, but does not provide a deeper analysis of why the data are the way they are (that is, why the humans being studied acted as they did). Of course not true of all behavioral economics, but some of it.

[sibmike]

I'd like to add that the data comes from experiments, which are simplified versions of real life problems. Oftentimes, "irrational" behavior depends on the way simplification is made and have no chances to generalise outside the laboratory, because in real life signal is more complex and informative. Yet a paper will be published stating "people behave irrationally". (in the lab with given peculiar instructions)