People Learn to Punish Moral Violations

Outrage at a moral violation may be common, but we do learn how to respond.

Posted Sep 12, 2018

mohamed_hassan CC0 via Pixabay
Source: mohamed_hassan CC0 via Pixabay

When you see someone do something that you consider to be morally wrong, chances are that you will experience some sense of outrage or frustration at their action. It is common to have a strong emotional response to violations of moral rules.

What you will do about it, though is another matter. You might choose to let it go, particularly if you feel that you have little power over the other person. You might choose to help anyone who was harmed by the moral transgression. You might also choose to punish the transgressor.

A paper in the August, 2018 issue of the Journal of Experimental Psychology: General by Oriel FeldmanHall, my former Ph.D. student Ross Otto, and Elizabeth Phelps explored whether people learn to punish others for a (simple) moral violation.

They had people play a variation on a simple economic game they called the Justice Game. In this game, there are two players. I’ll call them Player A and Player B. Player A is given $10 and can split it with Player B in any way he or she chooses. As an example, suppose Player A elects to keep $6 and to give $4 to the other player.

Player B then has three options. If they are happy with the split, they can Accept the division. If they are unhappy with the split, they have two options. One is to Compensate, in which case the money they get is raised to same level that Player A took. So, in this example, both players would get $6. The third option is to Reverse, so that the payoffs given to each player swap. In this example, Player A would now get $4, and Player B would get $6. Notice that this last option is the only one that explicitly punishes Player A for an uneven split of the money.

In the first experiment, participants played this game on-line. The experiment took place in three phrases. In the first phase, the participant was assigned the role of Player B. They played several rounds of the game in which Player A tried several uneven splits of the money in amounts that were unfavorable to player A. Player B then had the option to Accept, Reverse, or Compensate. Consistent with previous research on this game, participants chose to Compensate about 70% of the time. That Is, participants did not punish Player A, they just raised their own compensation to the same level as that of Player A.

So, in this first phase of the game, the participant was the victim of a minor moral transgression. Player A had the chance to treat Player B equally and chose not to do so. In general, people’s response to this uneven split was to even the score, but not to punish the other player.

Now, things get interesting.

In the next phase of the study, participants were assigned to one of three conditions. Now, they play the role of a third individual (Player C). In this round, Player A gets $10 and makes a split with Player B. Player C (the participant) now gets to decide on the response (Accept, Compensate, or Reverse) for Player B. After they make their choice, Player B lets player C know what they would have done. So, Player C gets feedback on how their own decision about whether to punish player A relates to the someone else’s decision. As in the first round of the task, the players tend to elect to Compensate Player B.

The participants are assigned to one of three conditions. In a control condition, Player B randomly selects from among the responses, so this provides little information about what they would do. In another condition, Player B decides they would prefer Compensate as a decision on 90% of the trials. In a third condition, Player B decides they would prefer Reverse as a decision on 90% of the trials. In this last condition, participants are getting feedback that the other player would punish Player A often.

Finally, all participants do another round of the game playing as Player B. This final round allows the experimenters to see any changes in performance after witnessing the responses of another player. Participants in the Control condition and the condition in which the other player generally chooses Compensate do not change their behavior. They still choose Compensate on 70% of the trials. But, when participants witness another player choosing to punish, their tendency to Reverse (and punish Player A) goes up to 36%, which is substantially higher than it was in the first round of the game.

The paper replicates this finding several times and also extends it by demonstrating that you can get the same increase in the tendency to punish just by observing the actions of another player without having to actually choose for them.

This set of findings suggests that people learn how to respond to the transgressions of other people. In these studies, the natural tendency of the participants was to deal with unfairness by making the outcome even. But, when participants witnessed someone else punishing the unfairness, it increased their own tendency to punish unfairness as well.

It will be interesting to see this research extended to more serious transgressions. In law enforcement, for example, there has been a pendulum between using the justice system for punishment and using it for rehabilitation. At any given moment, each of us reaches our own belief about how strongly a wrong must be punished. That belief is shaped in part by what we see other people doing.


FeldmanHall, O., Otto, A.R., & Phelps, E.A. (2018). Learning moral values: Another's desire to punish enhances one's own punitive behavior. Journal of Experimental Psychology: General, 147(8), 1211-1224.