From Boyd and Richerson’s The Origin and Evolution of Cultures (references removed):
The proposal that human intelligence is at the root of human cooperation is difficult to evaluate because of the ambiguity in what we might mean by intelligence in a comparative context. As the Tasmanian Effect [the loss of their toolkit in a small population] illustrates, individual human intelligence is only a part, and perhaps only a small part, of being able to create complex adaptive behaviors. In fact, we think ‘‘intelligence’’ plays little role in the emergence of many human complex adaptations. Instead, humans seem to depend upon socially learned strategies to finesse the shortcomings of their cognitive capabilities. The details of human cognitive abilities apparently vary substantially across cultures because culturally transmitted cognitive styles differ. Although we share the common intuition that humans are individually more intelligent than even our very clever fellow apes, we are not aware of any experiments that sufficiently control for our cultural repertoires to be sure that it is correct. The concept of ‘‘intelligence’’ in individual humans perhaps makes little sense apart from their cultural repertoires: humans are smart in part because they can bring a variety of ‘‘cultural tools’’ (e.g., numbers, symbols, maps, various kinematic models) to bear on problems. A hunter-gatherer would seem an incredibly stupid college professor, but college professors would seem equally dense if forced to try to survive as hunter- gatherers (a few knowledgeable anthropologists aside). Even abilities as seemingly basic as those related directly to visual perception vary across cultures. Second, intelligence implies a means to an end, not an end in itself. Individual intelligence ought to serve the ends of both cooperation and defection. We suspect that actually defection, requiring trickery and deception, is better served by intelligence than cooperation. Game theorists assuming perfect, but selfish, rationality predict that humans should defect in the one-shot anonymous prisoner’s dilemma, just as evolutionary biologists predict that dumb beasts using evolved predispositions will.
I’ve sat on that passage for a while now, contemplating turning it into a larger blog post. But for the moment, the abstract for this paper from Garett Jones points to the crux of my response:
Are more intelligent groups better at cooperating? A meta-study of repeated prisoner’s dilemma experiments run at numerous universities suggests that students cooperate 5–8% more often for every 100-point increase in the school’s average SAT score. This result survives a variety of robustness tests. Axelrod [Axelrod, R., 1984. The Evolution of Cooperation. Basic Books, New York] recommends that the way to create cooperation is to encourage players to be patient and perceptive; experimental evidence suggests that more intelligent groups implicitly follow this advice.
There is some evidence that patience is enough when the game is not particularly cognitively demanding (noting that patience and cognitive ability are positively correlated in most studies). But beyond a certain point, intelligence and cooperation appear to go hand-in-hand.