Sharing is caring. But how much sharing seems fair? In a classical economics test of fairness called the Ultimatum Game, two individuals must agree on how to divide a pot of money. If I propose a split and you accept, we both get the cash. If you refuse my offer, we both get nothing.
When people play this game, proposers usually offer about 40% of the total, and responders typically refuse anything less than 20%. A recent study explores how humans’ closest living relatives play the game. Chimpanzees and bonobos will give away a bigger share if they see the other animal has access to an alternative offer, the study shows. This suggests, say the researchers, that the great apes understand the concept of social leverage, a finding that could help shed light on the evolution of prosocial behavior.
“The key is whether giving the responders leverage could somehow influence the proposers—to see whether the proposers would start to switch a little bit towards making the more prosocial offers,” says Alejandro Sánchez-Amaro, a psychologist at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany, who led the study, which is published in Proceedings of the Royal Society B.
With his colleague Federico Rossano, a cognitive scientist at the University of California, San Diego, Sánchez-Amaro worked with groups of chimpanzees and bonobos at Leipzig Zoo. They installed a series of compartments, doors, and sliding panels in the zoo enclosures that allowed the animals to play the Ultimatum Game for food rewards. The scientists taught the animals to play as both proposer and responder.
The proposer could choose between two trays, each containing six food rewards. On the first tray, the rewards were split equally. But if the proposer selected the second tray, it could keep five rewards and the responder ape—if it accepted—got just one. (As in the human game, if the responder refused the offer, neither animal got the food from the tray.)
Previous attempts to get apes to play the Ultimatum Game have been controversial, because the responding animals tend to accept any offer of food, however small. To counter this effect, responder animals in the new study could choose between the offer from the proposer and food placed in a separate compartment by the researchers.
If the responder selected this alternative reward, the proposer animal got nothing. If the responder chose the proposer’s offer, both animals got the food. To vary the responder’s leverage, in different rounds of the game, the researchers baited the alternative compartment with zero, two, or four rewards.
Having an alternative choice seemed to give the responders more leverage in the scenario. The larger the alternative offer, the more likely the proposers were to select the tray that gave each animal three rewards. When the alternative tray was empty, proposers more often kept five treats for themselves. Economists call this self-maximizing behavior.
“The finding is not that chimpanzees are prosocial. The finding is that they are strategic,” says Sánchez-Amaro. “They understand somehow that when the partner can refuse, it’s better to sometimes give up and don’t go for the self-maximizing [choice].”
“What they’re saying makes sense, and I think their approach is reasonable,” says Keith Jensen, a psychologist at the University of Manchester, UK who was not involved in the study. “I think the strength of the paper was the opt-out options.”
Like much research with great apes, the study suffers from small sample sizes, adds Jensen, who has also studied the Leipzig chimpanzees. Only six chimpanzees and three bonobos played the full game, and to boost the effective sample size of proposers and responders, the researchers were forced to swap animals between roles across different tests. “But that’s a pretty generic critique,” he says. “Primate researchers don’t have several hundred undergraduates in the classroom that we can just test at the drop of a hat.”
But zoologist Manfred Milinski at the Max Planck Institute for Evolutionary Biology in Plön, Germany sounds a note of caution about the authors’ interpretation. “I am skeptical,” Milinski says. “To test whether chimps play the Ultimatum Game as humans do, you need to train them to understand the logic. This is extremely laborious and needs a lot of experience. How can we be sure that the animals understood exactly what the alternative offers meant?”
Sánchez-Amaro stresses that the animals had extensive training and testing that was designed to make sure they seemed to grasp the concepts involved. One reason so few animals were used to play the experimental games is because the researchers screened out those that did not pass what he calls “quite thorough training.”
Next, the researchers want to expand tests of prosocial behaviors to study chimpanzees and bonobos in games that involve more than two animals. “We think these will be more ecologically relevant,” Sánchez-Amaro says.