top of page

Trust is rewarding. 
But if our trust is betrayed, revenge is also rewarding

How should we think about trust? Are we innately trusting or suspicious? Are we able to trust strangers? What happens when trust is broken? Do we seek revenge? If we do seek revenge, why?


We start with this experiment conducted by Ernst Fehr and a team of Swiss researchers. Participants are observed playing a simple "trust game", a simple test to see if strangers trusted and reciprocated trust. If and when trust is broken, the response of participants was recorded. Participants wore brain scanners as they made decisions, which offers an unfiltered look into participants' true answers.

1) The Trust Game has 2 participants who do not know nor ever meet each other.

  • Participant A is given $10

  • She has 2 options:

    • Keep the $10​. If she does so, the game ends

    • Give the $10 to Participant B

55 (6).png
55 (5).png

2) If Participant A picks the second option - passing the $10 to Participant B - this sum will be quadrupled by the experimenter. In other words, B will receive $40.

55 (4).png

3) Now, it is B's turn to be given an option of what to do with the money:

  • He can keep the whole $40 sum to himself (which means A gets nothing)

  • Or he can split the $40 sum equally, so both participants get $20

55 (3).png

4) Ok now let's pause this for a moment. If you are participant A, what would you do? Keep the $10 for yourself? Or give the $10 to Participant B, with the hope that he reciprocates and shares the $40 bounty?

Traditional economic theory suggests that if Participant A is rational, she will choose to keep the $10 for herself. This is because the 2 participants will never meet. If B is given the money, a rational Participant B will choose not to share it. He can walk away with the money without A ever finding out who he is. A knows that B will likely the option of just leaving with the full $40. Therefore, a rational A will choose to just pocket the initial $10, rather than risking B not sharing the money with her.

Unsurprisingly, rational economic theory is incorrect. 

Almost all Participant As choose to give the $10 to Participant B. 

download (2).png

5) Now imagine if you were a Participant B. Would you simply keep the $40 to yourself and leave? Or would you reciprocate and share the money equally between you and Participant A.

About half of the participant Bs choose to share the $40 bounty, leaving both participants with more than they otherwise would if they didn't trust the other.

download (3).png

6) So what happens if the Participant B is that terrible bastard that walks away with the $40, leaving Participant A with $0? This is a 

Participant A will never know who B is. But in one version of this experiment, the experiments provide Participant As with a choice to punish errant Bs:

  • For every $1 that A contributes, B will have $2 taken away from him

  • So instead of receiving money, A has to now take out her own money just to punish B.

  • Will you do it if you were participant A?

The answer is unanimous. Every participant A in this scenario chooses to punish B. They were willing to spend their own money to pursue revenge.

Some important takeaways

There are 4 important takeaways:

1. We are generally trusting, even of people we do not know or have never met.


Almost all A participants decided to pass the initial $10 to B Participants, trusting them to reciprocate. But beyond the experiment, we can observe this in our daily lives. When we travel, we trust that pilots or bus/uber/cab/train drivers are well that day, and will not get into an accident. We trust that the people who prepare our food and drinks have washed their hands and the ingredients. We trust that the doctor we see is assessing our health capably. We trust that the financial institutions and our bankers and brokers are not siphoning our money (like Bernie Madoff)


2. Trust is generally beneficial for us

In fact, trust is what enables our societies to run. Imagine how life would be like if this fabric of trust does not exist, and we had to
individually make sure of everyone we ever deal with. 


And as the experiment shows, trust is generally beneficial for us. If A participants never trust B participants, the most they would ever walk away is $10, less than the $20 bounty they could have gotten if they trusted B, and if B reciprocated.

This is demonstrated in the classic Prisoners' Dilemma model (below) - a high amount of mutual trust creates the most mutual benefit - the combined outcome for both parties is the best. But if there is a loss of trust where one party tries to maximise his/her own benefit, and if the other party as the ability to retailiate, both parties are likely to end up with a far worse outcome.

But trust brings not just tangible benefits. It brings intangible ones as well. B participants who decide to share the bounty (hence, walking away with only $20) recorded higher levels of pleasure neurochemicals (dopamine) than those who walked away with $40. Even though they tangibly received less, they gained more pleasure through the additional avenue of gaining A's trust. 



3. If trust is broken, we tend to seek revenge. And we will seek revenge even when there is a cost to ourselves

There seems to be a strong evolutionary root to this. Since human trust and cooperation has been beneficial for us, we have evolved to punish those who violate trust and cooperation. And we are motivated to do so ourselves. Remember, for most of human history, there were no lawyers and judges, no police, no agreed-upon way of punishment. 

The pursuit of some form of punishment as revenge also has an altruistic element. Punishment is a signal to others to be wary of the offender, that he/she is not trustworthy. And we can still see examples of this today. Think about online reviews - for hotels, restaurants, shops, uber, Airbnb, attractions, etc. Or every time told someone else when we feel like someone had betrayed our trust.  

Finally, we are inclined to punish offenders even at a cost to ourselves. A participants were willing to spend additional money to ensure that offending B participants received less money.  

4. Not only do we seek revenge to punish the offender, this act actually brings us pleasure

Here's where the brain scans come in useful. In participants that expressed a strong desire to punish offending Bs, the brain region (the caudate nucleus) that plays a prominent role in the activation of the sensation of rewards. Moreover, it was observed that when this brain region activates strongly, participants would be willing to punish offending Bs even when it costs them. In other words, the anticipated pleasure of punishing an offending B is higher than the certain and tangible cost of punishment. 

Again, we can observe this playing out in our own lives. When someone betrays your trust, you might hold a grudge against him, looking for ways to exact revenge, even at the cost of your own well-being. Another example, for those who have lost trust in the government, they are willing to go against the government even when policies are for their own benefit. 

People punish not just for justice, but because we feel good when a punishment is meted out to someone whom we think have broken our trust. And it feels good enough that we tend to seek revenge even if it costs us. 

Read more at our section on Trust and Revenge.

bottom of page