Why We Trust Strangers Even Though We Think We Shouldn’t

The psychology of trusting and cooperation.

Posted Jul 11, 2019

Viacheslav Nikolaenko/Shutterstock
Source: Viacheslav Nikolaenko/Shutterstock

Trusting strangers is an integral part of modern society. I order food in a restaurant, and I trust that the server will bring me something safe to eat. Meanwhile, the server trusts that I’ll pay her when I’m done. Likewise, the restaurant owner trusts that the server will show up for work on time, and the server trusts that her boss will pay her at the end of the month. Without a general willingness to extend our trust to people we don’t know well, our modern world would collapse.

According to the rational actor model developed by economists, you should assume that other people will take advantage of you if they can possibly get away with it. Further, the model states that it’s in your best interest to take advantage of another’s trust anytime you can get away with it. That means you should only trust someone to the extent that you can compel them to follow through with their promise.

This model forms the motivation for contract law and the role of government in enforcing agreements. You and your roofer sign a contract before he replaces the shingles on your roof. If you refuse to pay when the work is done, he can take you to court. And if the roof leaks, you can sue for damages.

While we may sign contracts for major business transactions, in most everyday situations we rely instead on unsecured trust. We trust barbers, plumbers, and dry cleaners to render the services we request, and they trust that we’ll pay them when the job is done. Only in certain cases do we act according to the rational actor model—for example, in fast-food restaurants we’re expected to pay first before we get our meal.

Although people rarely follow the rational actor model in real life, they generally do so in hypothetical situations. For instance, research participants are often presented with imaginary scenarios in which they need to decide whether to trust a stranger or not. Most of the time, they choose not to trust the other person.

Further, these participants explain their decisions in terms of the rational actor model—that is, they say you can’t trust strangers because it’s in their best interest to take advantage of you. However, when faced with real situations in which they can choose to trust another person or not, they disregard their knowledge that strangers can’t be trusted.

Trust is studied in the laboratory using what’s appropriately called the trust game. The game involves two players. First, player A is given $5. She’s then told that if she gives the $5 to Player B, the amount will be quadrupled to $20. Player B can then keep all $20 or give back $10 to Player A.

In the ideal situation, A gives the $5 to B, and B gives $10 back to A, so that each comes out $10 ahead. This, of course, is the basis of friendship as a mutually beneficial relation. But friends know each other and have an ongoing relationship; moochers are quickly “unfriended.”

However, the trust game involves two strangers who’ve never met before and will likely never meet again. Moreover, depending on how the game is set up, they may never even meet at all, as is the case when the experimenter shuttles the money back and forth between two rooms.

According to the rational actor model, Player A should keep the $5, in which case Player B gets nothing. And if Player A does hand over the money, Player B should keep all $20, in which case Player A gets nothing. But this isn’t what happens. In the large majority of cases, Player A trusts Player B, who reciprocates.

If we ask Player A how likely it is that Player B will reciprocate, estimates average around 50-50. In other words, people greatly underestimate the likelihood that the other person can be trusted. And yet they choose to trust anyway. Ironically, if the trust game is restructured as a lottery in which players can bet $5 on a 50 percent chance of winning $10 and a 50 percent chance of gaining nothing, almost everyone refuses to play. Yet, notice that the payouts in the lottery are exactly the same as they are in the trust game. Apparently, there’s something about interacting with another person that makes us willing to take the risk.

The mystery of why people trust strangers even though they think they shouldn’t has been studied by University of Michigan psychologist David Dunning and his associates, and they explore it in a recent issue of Current Directions in Psychological Science. Dunning and colleagues rule out several common explanations for why people trust strangers.

The first explanation has to do with altruism. This involves doing something that benefits someone else at an expense to yourself. Certainly people are capable of engaging in altruistic acts. We volunteer our time, donate to charities, and help old ladies cross the street. We may even have an evolved propensity toward helping others in need, because doing so makes us feel good. (Behaviors that feel good typically have an evolutionary basis.)

Initially, the trust game appears to be driven by altruism: We incur the risk of losing all we have in the hope that both of us will benefit. But recall that when the trust game is recast as a lottery, people refuse to play. And this remains the case even when Player A is told that another person will receive some benefit regardless of the outcome of the lottery.

The second explanation has to do with public reputation. People often make public displays of generosity to enhance their social prestige. Benefactors donate large sums to hospitals and universities, and in return, get their name on a building. And in ongoing relationships with other people, our reputation is all we’ve got to demonstrate our trustworthiness. However, participants in the trust game behave the same way whether they make their decision in private or in front of others. We take the risk to cooperate in the trust game even when that risk entails no apparent social benefits.

In contrast to the standard explanations, Dunning and colleagues argue that people cooperate in the trust game because they believe that’s what they’re supposed to do. In other words, they’re following the social norm that you should trust another person unless you have a clear reason not to. This proposal is supported by self-reports of the participants themselves. Those who choose to trust Player B often say they did so because they felt it was the right thing to do, even though they thought the chance of betrayal was fairly high. And those who choose to keep the $5 for themselves often report feeling guilty about being selfish.

Further corroboration for the social norm explanation involves behavior under stress. Under normal conditions, cooperation runs at around 80 percent. However, when the players are subjected to high levels of stress, cooperation drops significantly. As we all know from personal experience, success in the struggle between doing what’s right and what feels good depends on how much stress we’re experiencing. For example, we’re faithful to our new diet for several days, but when problems at work or home build up, we’re back to eating junk food.

In the concrete jungles we live in, most people we interact with are strangers. Yet for society to work, we have to trust these strangers not to take advantage of us. As Dunning and colleagues point out, it’s not so much about giving others the benefit of the doubt. Rather, it’s about giving in to the social norm that you should trust strangers unless you have a good reason not to. This simple rule makes the modern world go round.


Dunning, D., Fetchenhauer, D., & Schlösser, T. (2019). Why people trust: Solved puzzles and open mysteries. Current Directions in Psychological Science. Advance online publication. DOI: 10.1177/0963721419838255