Joshua Greene appears to create a false dichotomy: we can either trust our intuitions, or we can "shut up and multiply" (it is interesting that Yudkowsky has already (http://lesswrong.com/lw/if/your_strength_as_a_rationalist/ ) shown the problem with mistrusting our intuitions and taking our model too seriously by "shutting up and multiplying". There is a third answer, which is actually the most traditional answer: find the best model (whether by researching models that more closely approximate the hypothetical, or by asking someone who has developed such a model by working in an appropriate field). So, when we are thinking about donations to charity, we know that helping the wounded is a dissimilar task. Using a model designed to work in nearby emergencies is likely to produce poor results when contemplating distant charities tackling ongoing problems. Instead, we are better served using a model that best approximates the actual situation. If we look at smart, decent people who work for overseas charities, we discover that their models emphasize striking a balance between aid for others, our loved ones, and ourselves. This is different from our behavior in emergencies (the bleeding stranger on the road)- and it's not that they're wrong. They just have a more appropriate model for that kind of situation.