Honestly, this is one of the greatest things I've ever read. (But then again, as I've admitted before, I'm not that well read.) There is a pretty compelling blend of cute monkey imagery with behavioral economics analysis.
Some choice excerpts:
On the test subjects: "'You can feed them marshmallows all day, they'll throw up and then come back for more.'"
On altruism: "The selfish jerk, meanwhile, was punished even worse. . .'They'd throw their feces at the wall, walk into the corner and sit on their hands, kind of sulk.'"
But can the data be generalized to humans?
"The data generated by the capuchin monkeys, Chen says, 'make them statistically indistinguishable from most stock-market investors.'"