Psychology study explains why strangers trust
By H. Roger Segelken
Playing the trust game
In one variation of the trust game experiment, subjects had three options: They could keep $5; game over. They could give $5 to an unknown person, in which case the money quadrupled to $20, and the unknown person was given a choice about sharing $10 back to the subject or keeping it all. Third, the subject could pass on the $5 to another person, the investment would inflate to $20, and the other person would, then flip a coin to see if $10 was handed back.
Although the coin flip would seem to offer the best chance of riches, only 22 percent of college students chose that option. More than half (54 percent) decided to trust a stranger to share fairly – even though they thought on average they would get money back only 37 percent of the time – whereas 24 percent kept the $5 for themselves.
Said Cornell’s David Dunning: “Trust is crucial not just for established relationships. It’s also especially vital between strangers with no responsibility toward each other outside of a single, transitory interaction.”
At a roadside stand a passerby slips money through the slot – SWEET CORN $2 A DOZEN – and wonders: “If the farmer isn’t watching, how does he know I’ll pay?” She doesn’t even bother peeling back the husks to check the kernels: “It must be good,” she seems assured. The stranger will never pass this way again. Her corn that night will be especially tender.
Now psychologists at Cornell and Germany’s Cologne University have explained a quirk that mystifies economists and philosophers alike: “excessive trust” among strangers. Researchers label it “excessive” because trust rates rose much higher than anticipated, given people’s aversion to risk and rather cynical expectations of their peers’ trustworthiness. Many trusted even though they expected their trust not to be honored.
“Trusting others is what people think they should do,” says David Dunning, professor of psychology in the College of Arts and Sciences. “Our experiments show that emotions – like anxiety and guilt for not fulfilling a ‘social duty’ or responsibility – may account for much of the ‘excessive trust’ observed between strangers every day.”
Again and again in laboratory variations of the so-called trust game (see sidebar) hundreds of students at Cornell in Ithaca and in Cologne disregarded risks, perceptions of others’ trustworthiness and even the 50/50 odds of a coin flip – and trusted anonymous strangers to do the right thing with their $5 or €5 bills.
Reporting online in the May 2014 Journal of Personality and Social Psychology, Dunning and his research collaborators wrote: “Trusting another person proved to be the most popular option ... trust was seen as the act that people should do, and this perception predicted who would trust versus not ... the action of trust was seen as more respectful and less insulting than the two alternatives.”
The researchers define trust as “a psychological state comprising the intention to accept vulnerability based on the chance of reward from positive intentions or behavior of another.”
Excessive trust, the researchers said, is borderline irrational according to a pure economic analysis – but happens millions of times a day around the world – in eBay transactions, farm stand sales and a thousand other human interactions between people who most likely will never meet.
“Despite the protests of philosophers and economists that it is irrational to do so,” the psychologists wrote, “people trust strangers and those strangers reward that trust.”
“If the farmer trusts you to pay for the corn,” Dunning adds, “it’s probably safe to assume it’s not wormy.”
The article, “Trust at Zero Acquaintance: More a Matter of Respect Than Expectation of Reward,” is by Cornell postdoctoral fellow Joanna E. Anderson, Dunning and Cologne’s Thomas Schlösser, Daniel Ehlebracht and Detlef Fetchenhauer. The study was funded by the National Science Foundation and by the Social Sciences Research Council of Canada.
Get Cornell news delivered right to your inbox.
Subscribe