But I gave him another quick scenario to see how he would handle it: imagine that you are at war, just you and five other fellow soldiers, all standing around in a circle. A grenade gets launched into the middle of the circle. If someone jumps on the grenade, only one person dies. If no one jumps on the grenade, there's a 20% chance someone might die and everyone will suffer moderate to critical injuries. Everyone is equidistant from the grenade and has an equal opportunity to jump on the grenade. Before I tell you what he said, I want the sociopaths who are reading this to think what they would do.
So, I asked my economist friend what he would do and he immediately replied, "I would jump on the grenade." Of course he would. He's rational and cares about efficiency. He would be the type of person in the trolley problem to throw the switch and kill the one to save the five, and apparently that answer doesn't change even when he is the one who needs to die. I think his answer surprised even him, though I'm not sure why. Perhaps because he had convinced himself that economists are soulless or at the very least selfish (i.e., rationally self-interested). But there's nothing remotely selfish or even self-interested about jumping on the grenade.
The reason I knew that this example would "work" on him is that he and I think similarly and it's something that I think I might do too. I like efficiency, and it would be efficient to fall on the grenade. Also I like winning, and it would be "winning" to thwart the enemy. It would be powerful, to smother the force of such a powerful device with just my body. Also I'm impulsive and not particularly attached to life. I actually think that a lot of sociopaths would do the same for one or more of those reasons. In fact, and I wish there was some way to accurately test this, I predict that a higher percentage of sociopaths would jump on the grenade than non-sociopaths, if for nothing else than the indecision or paralyzing fear that a lot of non-sociopaths might experience -- by the time they got around to making the decision, it might be too late. These are just guesses, but I don't think it's crazy to think that sociopaths might be braver and more pro-social in certain situations than normal people, just like economists might be more selfless than the average person in certain situations.
Whether or not my prediction is correct, I think this example also illustrates how dangerous it is to perform a couple experiments in controlled situations and extrapolate the data far beyond those particular situations. Sloppy science writers (and even serious researchers) make this mistake all of the time, e.g. if sociopaths seem to not show empathy in one situation, it's easy to make the (apparently incorrect) presumption that they never feel empathy. The truth is context matters immensely and we only know a sliver of all there is to know about ourselves and others.