I wish they had not picked the price of paintings as their experimental test. Since the price of a painting is almost entirely subjective, i.e. if everyone els

I wish they had not picked the price of paintings as their experimental test. Since the price of a painting is almost entirely subjective, i.e. if everyone else thinks it's valuable, it's valuable, it could be argued that "herding" to the average of what others think is the rational behavior. You've got more hard data about the value you're supposed to estimate, which is more or less "what do others think this is worth"; it would be odd not to adjust your estimate accordingly.