What Human Behavior Teaches Us About Trust: A Social Psychologist’s Perspective
Foundations
Contributor / J. Keith Murnighan
Professor of Management and Organizations, Harold H. Hines Jr. Distinguished Professor of Risk Management
Kellogg School of Management / Management
Social psychologists are interested in trust because it allows people to interact more deeply and effectively. In particular, they seek to understand the process of developing trust and what happens when it is broken. A tool developed by economists, called the Trust Game, sheds light on these questions. One key finding is that displaying trust creates a sense of obligation to reciprocate. The greater the risk, the greater the sense of obligation created. But the bond of trust breaks down quickly when players shows little trust.
Transcript
I was trained as a social psychologist. Social psychologists pay attention to normal behavior by normal people in everyday situations and try and figure out why we do funny things the way we do.
My whole approach is about interpersonal interactions. So, I’m looking at small groups, large groups, but people interacting with each other.
Game theory is all about how rational people interact with that. My take — I love game theory because it’s just beautiful, formal, clear theories, with clear assumptions and clear outcomes and clear predictions — it wasn’t designed to be behavioral, and it didn’t include emotional.
So, my research tends to look at, do these beautiful models actually play out behaviorally? Do emotions add to our understanding of what happens and why some of those predictions don’t work?
BUMPER: Fundamental Trust Questions in the Discipline
Our research on trust, folks, is on how trust develops, what happens when there’s a breach. And those, for us, are the two biggest questions because what we’d like to see is people benefit and have efficient, effective interactions.
Game theory is like that as well; you want to maximize your outcomes. And we’d like people to actually maximize what they get in their interactions, and trust has a lot to do with that.
If you can build trust, you can interact more deeply, more effectively, and people get more out of the situation.
Anyway, recently — not that recently, 1995 — a group of three wonderful economists created what we now call the Trust Game. And the Trust Game is a simple situation where two people will interact — they can interact face to face or anonymously.
One person, often referred to as player one — we don’t use the word “trust” in our experiments because we don’t want to cue that — player one gets an endowment, say, of 10 dollars.
They have a choice of how much to send to player two. They can send anything from 0 to all the 10 dollars.
They know that however much they send is going to be tripled on its way to player two. So, if they send the whole 10, player two is going to have 30 dollars, so they’re creating more.
Player two then has a choice to return as much or as little as they want to player one.
So, the question is (this is a great game) — player one trusts; player two reciprocates — what are the factors that lead to player one trusting? What are the factors that lead to player two reciprocating?
And one of the major findings that we have found and others have found as well is that the more player one trusts, the more risks they take, the more player two reciprocates.
So, player twos have this feeling that if player ones have trusted them so much, they’re obligated.
If player one doesn’t trust them much, their obligation goes way down fast. But if they’ve been trusted a lot, obligation feelings come up, and they’re much more likely to send back a high amount.
BUMPER: Looking Forward in the Discipline
Studies of trust have proliferated. And researchers are now trying to slice and dice different kinds of trust: Is distrust the opposite of trust? Or is it something altogether separate on its own?
There’s affective trust, more of a feeling; there’s cognitive trust, where you think it out and calculate.
So, there’s debate about concepts that — for instance, one definition of trust talks about integrity, benevolence and competence.
But when we ask people who do they trust, all three come out pretty high and they overlap in ways. So, are they really separable concepts? I don’t know.
You can have trust in your head; you can have cognitive trust, affective trust, feelings that somebody has integrity.
I am impressed mostly when people behave and act as if they trust someone because, for me, everything’s about behavior. The most important things are about behavior.
Your feelings and your thoughts, yes, take up a lot of people’s time. But I want to see what’s happening in action. And certainly, from a business standpoint, that’s more important.
There are debates in the field, and there will continue to be debates. And people will niggle and nitpick. And let’s just see what people do in important situations, and I think that will be the telling factor and build more understanding.
My whole approach is about interpersonal interactions. So, I’m looking at small groups, large groups, but people interacting with each other.
Game theory is all about how rational people interact with that. My take — I love game theory because it’s just beautiful, formal, clear theories, with clear assumptions and clear outcomes and clear predictions — it wasn’t designed to be behavioral, and it didn’t include emotional.
So, my research tends to look at, do these beautiful models actually play out behaviorally? Do emotions add to our understanding of what happens and why some of those predictions don’t work?
BUMPER: Fundamental Trust Questions in the Discipline
Our research on trust, folks, is on how trust develops, what happens when there’s a breach. And those, for us, are the two biggest questions because what we’d like to see is people benefit and have efficient, effective interactions.
Game theory is like that as well; you want to maximize your outcomes. And we’d like people to actually maximize what they get in their interactions, and trust has a lot to do with that.
If you can build trust, you can interact more deeply, more effectively, and people get more out of the situation.
Anyway, recently — not that recently, 1995 — a group of three wonderful economists created what we now call the Trust Game. And the Trust Game is a simple situation where two people will interact — they can interact face to face or anonymously.
One person, often referred to as player one — we don’t use the word “trust” in our experiments because we don’t want to cue that — player one gets an endowment, say, of 10 dollars.
They have a choice of how much to send to player two. They can send anything from 0 to all the 10 dollars.
They know that however much they send is going to be tripled on its way to player two. So, if they send the whole 10, player two is going to have 30 dollars, so they’re creating more.
Player two then has a choice to return as much or as little as they want to player one.
So, the question is (this is a great game) — player one trusts; player two reciprocates — what are the factors that lead to player one trusting? What are the factors that lead to player two reciprocating?
And one of the major findings that we have found and others have found as well is that the more player one trusts, the more risks they take, the more player two reciprocates.
So, player twos have this feeling that if player ones have trusted them so much, they’re obligated.
If player one doesn’t trust them much, their obligation goes way down fast. But if they’ve been trusted a lot, obligation feelings come up, and they’re much more likely to send back a high amount.
BUMPER: Looking Forward in the Discipline
Studies of trust have proliferated. And researchers are now trying to slice and dice different kinds of trust: Is distrust the opposite of trust? Or is it something altogether separate on its own?
There’s affective trust, more of a feeling; there’s cognitive trust, where you think it out and calculate.
So, there’s debate about concepts that — for instance, one definition of trust talks about integrity, benevolence and competence.
But when we ask people who do they trust, all three come out pretty high and they overlap in ways. So, are they really separable concepts? I don’t know.
You can have trust in your head; you can have cognitive trust, affective trust, feelings that somebody has integrity.
I am impressed mostly when people behave and act as if they trust someone because, for me, everything’s about behavior. The most important things are about behavior.
Your feelings and your thoughts, yes, take up a lot of people’s time. But I want to see what’s happening in action. And certainly, from a business standpoint, that’s more important.
There are debates in the field, and there will continue to be debates. And people will niggle and nitpick. And let’s just see what people do in important situations, and I think that will be the telling factor and build more understanding.
Related Videos
Trust Breaches: Timing and Recovery
Applications
Contributor / J. Keith Murnighan
J. Keith Murnighan
Management
Breaches,Reputation Management,Social Psychology,Vulnerability,Reciprocity,Broken Trust
We were interested in the long-term development of a relationship. And I’m an old fan of Hollywood movies, in particular, romantic comedies. So, Rosalind Russell and Cary Grant and His Girl Friday and a bunch of these movies.
When you watch them, it’s very easy to see that when there’s discord between the two stars early on, romance later is going to be supreme and fantastic.
One of the theories that we had was that if things are bad early, they may have a higher potential later on. Now, there’s an alternative way to think about it, of course — if things are bad early, they just don’t have a chance and they don’t go anywhere.
So, we actually tried to test the two, and we had interactions between people where there was a trust breach early or a trust breach after many cooperative choices.
And it turned out that the early trust breaches were devastating and led to less future cooperation, less trust.
When we saw breaches late, we thought that this might really, really damage things. It does immediately; it’s a shock when somebody breaches trust.
But our guess — and all we have is a guess at this point because we haven’t studied it enough — is that when someone breaches trust, they exhibit strength. And what happens with strength is we do respect it.
And when someone is trustworthy and strong — i.e., someone who’s trustworthy who doesn’t have to be — we tend to trust them more.
BUMPER: Overcoming a Breach in Trust
One of the things that should happen any time trust is broken is you should pursue and try and find out what happened, simply asking questions, because it can be the result of miscommunication, misunderstandings, a variety of things that are simple explanations that have nothing to do with a person’s trustworthiness.
However, if it was intentional, it tells you a lot about the person. And in business situations, what we find when there’s a serious trust breach, trust is gone, when it really comes to the crunch.
You might work with the other person; you might cooperate with them; you might have a beneficial relationship. But if you have to really trust them, it’s unlikely to happen.
BUMPER: Using the Rational Model to Decide Whom to Trust
The rational model suggests what rational people will do to benefit the most themselves. However, we’re not all completely rational, and we actually use very subtle cues and a lot of information to determine whether someone is trustworthy.
So, if you hire a lawyer, hopefully you’ve done some homework and you found out about that lawyer’s reputation, and that allows the trust process to develop more quickly.
Any time you trust someone, you’re vulnerable. You take the first step in whatever it is — sharing a secret, providing some information about a work project, or loaning money or a car. Any time you trust someone, you’re vulnerable.
The question is, how vulnerable do you want to make yourself?
The model tells you to be very cautious. And if you continuously get positive information, trust can continue to grow — again, to a limit.
At some point, things will level off because you don’t want to trust one person too much. It’s the old saying, “Too many eggs in one basket,” for instance.
In business situations, it truly pays to have many people you can trust a moderate to high amount and nobody you have to depend upon too much.
BUMPER: How Much Vulnerability? It Depends
How much do I actually want to loan my best friend? I’d be happy to loan him 5,000 dollars, 10,000 dollars. But when it starts to hurt me, can hurt me badly, do I really want to loan them 100,000 dollars? That would be tough.
In my classes, I often ask people, “Would you loan me 10 dollars?” And if they say yes, I say, “Would you loan me 20? Would you loan me 100?” There’s always a limit.
One time, I did ask someone if they’d loan me 10, and they said, “No.” I said, “How about 5?” They said, “No.” I said, “How about 2?” They said, “No.” I said, “How about 1?” They said, “Maybe.” That was a very cautious person.
Most of us are willing, with people we know, to take some risks. But there’s always a limit.
When you watch them, it’s very easy to see that when there’s discord between the two stars early on, romance later is going to be supreme and fantastic.
One of the theories that we had was that if things are bad early, they may have a higher potential later on. Now, there’s an alternative way to think about it, of course — if things are bad early, they just don’t have a chance and they don’t go anywhere.
So, we actually tried to test the two, and we had interactions between people where there was a trust breach early or a trust breach after many cooperative choices.
And it turned out that the early trust breaches were devastating and led to less future cooperation, less trust.
When we saw breaches late, we thought that this might really, really damage things. It does immediately; it’s a shock when somebody breaches trust.
But our guess — and all we have is a guess at this point because we haven’t studied it enough — is that when someone breaches trust, they exhibit strength. And what happens with strength is we do respect it.
And when someone is trustworthy and strong — i.e., someone who’s trustworthy who doesn’t have to be — we tend to trust them more.
BUMPER: Overcoming a Breach in Trust
One of the things that should happen any time trust is broken is you should pursue and try and find out what happened, simply asking questions, because it can be the result of miscommunication, misunderstandings, a variety of things that are simple explanations that have nothing to do with a person’s trustworthiness.
However, if it was intentional, it tells you a lot about the person. And in business situations, what we find when there’s a serious trust breach, trust is gone, when it really comes to the crunch.
You might work with the other person; you might cooperate with them; you might have a beneficial relationship. But if you have to really trust them, it’s unlikely to happen.
BUMPER: Using the Rational Model to Decide Whom to Trust
The rational model suggests what rational people will do to benefit the most themselves. However, we’re not all completely rational, and we actually use very subtle cues and a lot of information to determine whether someone is trustworthy.
So, if you hire a lawyer, hopefully you’ve done some homework and you found out about that lawyer’s reputation, and that allows the trust process to develop more quickly.
Any time you trust someone, you’re vulnerable. You take the first step in whatever it is — sharing a secret, providing some information about a work project, or loaning money or a car. Any time you trust someone, you’re vulnerable.
The question is, how vulnerable do you want to make yourself?
The model tells you to be very cautious. And if you continuously get positive information, trust can continue to grow — again, to a limit.
At some point, things will level off because you don’t want to trust one person too much. It’s the old saying, “Too many eggs in one basket,” for instance.
In business situations, it truly pays to have many people you can trust a moderate to high amount and nobody you have to depend upon too much.
BUMPER: How Much Vulnerability? It Depends
How much do I actually want to loan my best friend? I’d be happy to loan him 5,000 dollars, 10,000 dollars. But when it starts to hurt me, can hurt me badly, do I really want to loan them 100,000 dollars? That would be tough.
In my classes, I often ask people, “Would you loan me 10 dollars?” And if they say yes, I say, “Would you loan me 20? Would you loan me 100?” There’s always a limit.
One time, I did ask someone if they’d loan me 10, and they said, “No.” I said, “How about 5?” They said, “No.” I said, “How about 2?” They said, “No.” I said, “How about 1?” They said, “Maybe.” That was a very cautious person.
Most of us are willing, with people we know, to take some risks. But there’s always a limit.
Trust and Leadership
Applications
Contributor / J. Keith Murnighan
J. Keith Murnighan
Management
Leadership,Reputation Management,Social Psychology,Swift Trust,Broken Trust, Vulnerability
Every leader wants to come across as competent. And what a lot of research suggests is every leader should also want to come across as warm — as interpersonally warm and caring.
So, this nice combination of competence and warmth is dynamite for leaders. It gets really positive responses from people who not only think you care about them but also understand that you know what’s going on. And that’s a powerful combination.
And if you’re a leader and have been appointed a leadership position for good reasons, people are going to think you’re smart and they’re going to give you some credit.
And think about this — think about leaders who will tell you, “You must earn my trust” versus leaders who say, “We’re starting out with 100 percent trust. You have a good track record. I’ve seen your reports. I have high expectations from you. You’ve been fantastic. Let’s take it from there” — altogether different from starting at ground zero.
BUMPER: Building Trust within Teams
When I work with companies and with leaders, I’m always focusing on the teams that they lead — the immediate groups of people that they lead. And what I want to have happen is for those teams to coordinate themselves well and trust each other.
And if they can coordinate themselves well and trust each other, they’re going to be able to take advantage of whatever abilities they have.
Add on some training programs where their abilities increase — absolutely tremendous bottom-line impact over the long term.
You can’t predict short-term — internal dynamics and their effect on the bottom line, in the short term, is not always obvious.
But in the long term, a really smoothly functioning team that’s well coordinated and trusts each other and has ability — whew! — fantastic.
BUMPER: Exploring Automatic Trust
Automatic trust is where you encounter someone, and for some reason, things click. And you find and believe that they’re trustworthy right away.
Our brains have all of these connections — interwoven connections in our brain — that activate when you say something like a person’s name or when they’re wearing a pair of glasses that you recognize that, for some reason, you have a positive association about.
So, automatic trust is a situation where you get a cue that all of sudden leads you to be more trusting than you would be otherwise. And there are lots of those cues that are possible.
BUMPER: Automatic Trust and Likeability: Creating Schemas
Liking is one thing, but we also have categories where a very likeable person is not particularly trustworthy. And we all know of them. They’re great to have a party with, but you would never loan them your car.
So, we’re pretty good at creating what we call “schemas” for different kinds of people. And we have schemas for professors; we have schemas for nerds; we have schemas for CEOs.
And we have schemas for people who are likeable but not trustworthy and schemas for people who are trustworthy but not so likeable.
Automatic trust can facilitate a better result, when you encounter one of the many, many people who truly are trustworthy, because if I come on as more trusting of you, you’re more likely to reciprocate and be more trusting of me. And we accelerate a trust-development process to both of our mutual benefit.
So, this nice combination of competence and warmth is dynamite for leaders. It gets really positive responses from people who not only think you care about them but also understand that you know what’s going on. And that’s a powerful combination.
And if you’re a leader and have been appointed a leadership position for good reasons, people are going to think you’re smart and they’re going to give you some credit.
And think about this — think about leaders who will tell you, “You must earn my trust” versus leaders who say, “We’re starting out with 100 percent trust. You have a good track record. I’ve seen your reports. I have high expectations from you. You’ve been fantastic. Let’s take it from there” — altogether different from starting at ground zero.
BUMPER: Building Trust within Teams
When I work with companies and with leaders, I’m always focusing on the teams that they lead — the immediate groups of people that they lead. And what I want to have happen is for those teams to coordinate themselves well and trust each other.
And if they can coordinate themselves well and trust each other, they’re going to be able to take advantage of whatever abilities they have.
Add on some training programs where their abilities increase — absolutely tremendous bottom-line impact over the long term.
You can’t predict short-term — internal dynamics and their effect on the bottom line, in the short term, is not always obvious.
But in the long term, a really smoothly functioning team that’s well coordinated and trusts each other and has ability — whew! — fantastic.
BUMPER: Exploring Automatic Trust
Automatic trust is where you encounter someone, and for some reason, things click. And you find and believe that they’re trustworthy right away.
Our brains have all of these connections — interwoven connections in our brain — that activate when you say something like a person’s name or when they’re wearing a pair of glasses that you recognize that, for some reason, you have a positive association about.
So, automatic trust is a situation where you get a cue that all of sudden leads you to be more trusting than you would be otherwise. And there are lots of those cues that are possible.
BUMPER: Automatic Trust and Likeability: Creating Schemas
Liking is one thing, but we also have categories where a very likeable person is not particularly trustworthy. And we all know of them. They’re great to have a party with, but you would never loan them your car.
So, we’re pretty good at creating what we call “schemas” for different kinds of people. And we have schemas for professors; we have schemas for nerds; we have schemas for CEOs.
And we have schemas for people who are likeable but not trustworthy and schemas for people who are trustworthy but not so likeable.
Automatic trust can facilitate a better result, when you encounter one of the many, many people who truly are trustworthy, because if I come on as more trusting of you, you’re more likely to reciprocate and be more trusting of me. And we accelerate a trust-development process to both of our mutual benefit.
Rationality and Reciprocity in Trust: Key Findings
Research
Contributor / J. Keith Murnighan
J. Keith Murnighan
Management
Reciprocity,Breaches,Social Psychology,Reputation Management
BUMPER: Key Research in the Discipline
There have been a bunch of articles on trust that are really important. In 1995, Mayer, Davis, and Schoorman wrote a theory paper on trust that introduced integrity, competence and benevolence as the three critical elements of trust.
Prior to 1995, there had been very little research on trust, for no good reason. Early ’60s and ’70s, there was a lot of work on personality and trust, and trust propensities, and how likely it is someone will trust.
And people would answer questions like, “I believe most people are honest” and other questions like that, and they would try and relate people’s responses to those to their behavior.
And there were moderate predictability but not very impressive. I think that’s kind of why the research on trust dissipated after the early ’70s.
Mayer, Davis and Schoorman were great at coming up with a very noteworthy article that said, “Let’s start studying trust again.”
And so, since the ’90s, there’s been an enormous amount of work on trust.
BUMPER: Trust and Rationality
Gary Alan Fine in our sociology department has written a book on the Minnesota Mycological Society.
And one of the things that he focuses on in this book that relates to trust is that new members of the society — after they’ve gone out mushroom hunting, they come back, they cook their mushrooms and eat them — new members will eat other people’s mushrooms when they haven’t even met the other person. Old members will not.
So, new members want to join, want to become part of the group, want to become members, want to be well thought of, so they say, “yes,” when they’re offered mushrooms that they don’t recognize by someone they’ve never met.
It’s pretty darn dangerous! People don’t choose theirs and are easier to pass it up when they’re already established members — pretty cool.
BUMPER: Trust and Reciprocity
“Attributions of Trust and the Calculus of Reciprocity” — one of my favorite article titles — Madan Pillutla and Deepak Malhotra and I varied how much trustors trusted and then observed how much trustees responded and found very clear indications that the more you trust, the more people respond and reciprocate, especially when they think you’re doing it in a smart way.
That was one of our first studies, and I just thought, very neat and clean and tells us a lot about the trust-reciprocation process.
There have been a bunch of articles on trust that are really important. In 1995, Mayer, Davis, and Schoorman wrote a theory paper on trust that introduced integrity, competence and benevolence as the three critical elements of trust.
Prior to 1995, there had been very little research on trust, for no good reason. Early ’60s and ’70s, there was a lot of work on personality and trust, and trust propensities, and how likely it is someone will trust.
And people would answer questions like, “I believe most people are honest” and other questions like that, and they would try and relate people’s responses to those to their behavior.
And there were moderate predictability but not very impressive. I think that’s kind of why the research on trust dissipated after the early ’70s.
Mayer, Davis and Schoorman were great at coming up with a very noteworthy article that said, “Let’s start studying trust again.”
And so, since the ’90s, there’s been an enormous amount of work on trust.
BUMPER: Trust and Rationality
Gary Alan Fine in our sociology department has written a book on the Minnesota Mycological Society.
And one of the things that he focuses on in this book that relates to trust is that new members of the society — after they’ve gone out mushroom hunting, they come back, they cook their mushrooms and eat them — new members will eat other people’s mushrooms when they haven’t even met the other person. Old members will not.
So, new members want to join, want to become part of the group, want to become members, want to be well thought of, so they say, “yes,” when they’re offered mushrooms that they don’t recognize by someone they’ve never met.
It’s pretty darn dangerous! People don’t choose theirs and are easier to pass it up when they’re already established members — pretty cool.
BUMPER: Trust and Reciprocity
“Attributions of Trust and the Calculus of Reciprocity” — one of my favorite article titles — Madan Pillutla and Deepak Malhotra and I varied how much trustors trusted and then observed how much trustees responded and found very clear indications that the more you trust, the more people respond and reciprocate, especially when they think you’re doing it in a smart way.
That was one of our first studies, and I just thought, very neat and clean and tells us a lot about the trust-reciprocation process.