Last March, Iyad Rahwan invited two members of his laboratory over to his house to spend an entire Saturday in front of their computers. His goal was to locate five people wandering the city streets of Stockholm, London, Bratislava, New York City and Washington DC within 12 hours, from his living room near the Arabian Gulf coast in Abu Dhabi. The people they sought were known only from mugshots released that morning.
The manhunt was part of a competition called the Tag Challenge, sponsored by the US State Department and the US Embassy in Prague, which tested the limits of cooperation over the web. The event was staged: the five people wandering city streets were actors pretending to be jewel thieves. Rahwan’s team, from the Masdar Institute in the UAE, won the competition.
Rahwan attributes their success to the directed manner in which the people they contacted early on worked their existing social networks. “As we got closer to the deadline, our people recruited specific friends,” Rahwan says.
The behaviour of cooperation, defined as engaging with others in a mutually beneficial activity, reaches unprecedented levels of complexity in humans. Humans frequently bond with unrelated individuals, unlike ants who defend their sisters in a colony.
Furthermore, humans don’t rely entirely on instinct to form groups, in contrast to wildebeest grazing in herds, or termites, who are genetically programmed to serve the insect chemically designated as their queen. Instead, the basis for human cliques fluctuates. Everything ranging from a common nationality to a shared favourite football team may distinguish who is on the “inside” from someone who is not.
The manner in which social groups interact online may be revolutionary, but the basic cooperative nature driving their actions is as old as humankind. Tools, technologies and cultures change, says Samuel Bowles, a behavioural scientist at the Santa Fe Institute in New Mexico, and author of A Cooperative Species: Human Reciprocity and its Evolution. “Statues and flags go up and come down,” he says. “What is a durable is both an ability to interact in peaceful ways with strangers and a propensity to help members of one’s own group.”
Anthropologists suggest that roughly one million years ago, cooperation gave the ancestors of humans, early members of the genus Homo, an advantage over a related hominid group, Paranthropus, who similarly walked on two legs. Perhaps, the larger brains of Homo lent our ancestors the cognitive capabilities required for cooperation, allowing them to hunt bigger bounty and mount stronger defences against predators. Indeed, archaeological digs show that Homo sapiens feasted on hippopotamus and other large game 75,000 to 90,000 years ago, in what is now South Africa.
To assess how widespread cooperation is among humanity, Gintis, Bowles and their colleagues conducted a game-theory experiment called the “ultimatum game” in communities ranging from those in Athens, to Hadza hunter-gatherer groups in Tanzania. Every person who played the game shared more goods than would be expected if they were completely selfish and people punished players who did not share to the degree they considered appropriate.
“I’m interested in this game because I care about the kinds of social behaviour that sustain morality,” Bowles says. “I’ve done this experiment around the world, and no society is completely self-interested,” he says. “There is a huge amount of evidence that people have strong feelings about helping others.”
Social groups were once tribes that lived within a specific locality. Now, with mail, telephones, and the internet, geographic boundaries matter far less. Statistics on Facebook speak for themselves. In December 2012, the social network boasted 1.06 billion active users. What’s more, each person can link to anyone else through an average of 4.7 hops through mutual friends, which means that groups can now grow to unprecedented sizes.
Rahwan’s team managed to locate three of the five “jewel thieves” from a living room in Abu Dhabi by relying on web-based networks. Their success demonstrates that concise tasks might indeed be possible through computer-mediated cooperation under the right conditions, but it will depend on how people use it, and its best use remains largely unknown.
“The mechanisms to cooperate are constantly changing, and so are people’s aims,” Rahwan says. “We understand the current forms of cooperation far, far less than the ones we had prior to the rise of the internet. Everything is new, and it changes very fast.”