Moderators of the disapproval of peer ‘punishment’ Kimmo Eriksson, Per Andersson, Pontus Strimling To appear in Group Processes & Intergroup Relations.

Moderators of the disapproval of peer

‘punishment’

Kimmo Eriksson, Per Andersson, Pontus Strimling

To appear in Group Processes & Intergroup Relations

Personality and morality

Social dilemmas

Who makes an effort for the greater good?

Are such efforts generally approved of?

The intuitive answers have at least some truth to them.

Personality and morality

Social dilemmas

Who makes an effort for the greater good?

Are such efforts generally approved of?

Peer ‘punishment’

Who makes an effort to harm a peer who neglects the greater good?

Are such efforts generally approved of?

The intuitive answers have at least some truth to them.

No clear intuitive answers!

Answers suggested by prior research

• Who uses peer ’punishment’ is – not related to prosociality– correlated across strategic situations and

targets

• Peer ’punishment’ tends to be disapproved of– as measured by experiments on second-order

punishment in economic games– as measured by judgments of vignettes on everyday social dilemmas

• but higher approval if punishment is a collective concern

• Prior methods are not wholly satisfactory!• Unclear what problem people have with peer ’punishers’, given that the

direct effect of their actions for the group tends to be positive.

Eriksson et al. (2014) Review of Behavioral Economics

Strimling & Eriksson (2014) In van Lange et al. (eds.), Reward and Punishment in Social Dilemmas.

HypothesisA problem people have with peer ’punishers’ is that they come across as aggressive (bullies); such people are on the whole viewed as a problem rather than an asset to have around because they will also ’punish unfairly’.

Hypothesis & PredictionsA problem people have with peer ’punishers’ is that they come across as aggressive (bullies); such people are on the whole viewed as a problem rather than an asset to have around because they will also ’punish unfairly’.

P1: Collectively supported ’punishment’ – indicating that it is less driven by the punisher’s personality – should be disapproved of less.

P2: More severe ’punishment’ - although potentially more effective - should be disapproved of more.

P3: More aggressive people should identify more with the ’punisher’ and therefore disapprove less of peer punishment.

Our method: abstract animations (inspired by Heider & Simmel, 1944)

Weak – Individual – Physical

Example of how one participant described his/her judgment of the situation

“blue triangle was quite aggressive, 'pushing' and 'shoving' the purple triangle... the behavior of the purple triangle was less than exemplary but there were other ways to deal with this situation”

Study 1: Variations of ”physical punishment”

Severity: Strong / WeakResponsibility: Individual / Collectiveas well as no punishment at all

2 × 2 + 1 = 5 variations

Strong instead of Weak

Example quote

”Blue had some serious anger issues!!”

Collective instead of Individual

Approval scale1. I think the BLUE triangle’s behavior was appropriate 2. I would like to spend time with a person who

behaves like the BLUE triangle 3. If a person who behaves like the BLUE triangle

belonged to my group I would consider that person to be a problem (rather than an asset) for the group

4. I think BLUE is someone who is generally trustworthy

5. I think BLUE is someone who is generally angry6. I think BLUE is someone who takes others’ interests

into account7. I think BLUE is someone who would punish others

unfairly8. I think BLUE is someone who generally follows

standard norms of behavior

Cronbach’s alpha = .92

N = 500 respondents (100 per condition)

G: All versions of punishment were less approved of than no punishment at all.

N = 500 respondents (100 per condition)

G: All versions of punishment were less approved of than no punishment at all.

P1: Collective → less disapproval

N = 500 respondents (100 per condition)

G: All versions of punishment were less approved of than no punishment at all.

P1: Collective → less disapproval

P2: Severe → much greater disapproval

G: All versions of punishment were less approved of than no punishment at all.

P1: Collective → less disapproval

P2: Severe → much greater disapproval

P3: High aggression respondent → less disapproval

Economic instead of Physical

N = 500 respondents (100 per condition)

Similar results!

G: All versions of punishment were less approved of than no punishment at all.

P1: Collective → less disapproval

P2: Severe → much greater disapproval

N = 500 respondents (100 per condition)

Similar results!

G: All versions of punishment were less approved of than no punishment at all.

P1: Collective → less disapproval

P2: Severe → much greater disapproval

P3: High aggression respondent → less disapproval

Summary of findings

Main effect of peer punishment compared to no punishment:•Across physical and economic kinds of ’punishment’ we found general moral disapproval of animated peer punishers

Three moderators were identified:•More disapproval of individually deployed punishment•More disapproval of more severe punishment•More disapproval expressed by low aggression individuals

Quick preview of new data

I decide you get 25 %

Costly 2nd party ’punishment’ after receiving an unfair split in the dictator game

Spend my share on reducing his Spend my share on reducing his payoff to zero?payoff to zero?

After choice: Why did you make this choice? to be fair it’s the morally right

thing to earn as much as

possible to defy the other to punish the other to protect myself

Conclusions

Consistent with the previously found tendency to morally disapprove of peer ’punishment’, there is a common moral motive to refrain from using it.

Consistent with the previously found moderating role of aggression, •punitively motivated punishment was associated with high trait aggression•and morally motivated restraint was associated with low trait aggression.