Group Membership Modulates Early Expectations about Retaliatory Harm

• In Event: 1-035 - Friends and enemies: Children’s developing understanding of how relationships inform behavior, inferences and expectations.

Integrative Statement

Social interaction among humans is rife with instances of retaliation. But how do humans judge how much retaliation is appropriate for a given misdeed, and what factors affect such judgments? In adults and older children, the relationship between the wrongdoer and the victim, such as belonging to the same social group, seems to modulate the magnitude of retaliation that is considered appropriate. However, while some experiments suggested that belonging to the same group reduces this magnitude, others suggested the opposite (for review, see McAuliffe & Dunham, 2016). The present experiments contributed to this issue by examining, very early in development, how much retaliation was deemed appropriate in response to the same misdeed when perpetrated by an outgroup or an ingroup wrongdoer.
In Experiment 1, 18-month-old infants watched live events involving two experimenters, E1 and E2, who wore either different outfits (outgroup condition) or similar outfits (ingroup condition). In both conditions, E1 built three towers with four blocks each, while E2 watched and ate crackers. After E1 finished her third tower, she stole one of E2’s many crackers and left. In retaliation, E2 knocked down 1 block, 1 tower, 2 towers, or 3 towers. Infants in the outgroup condition detected a violation in all but the 2-tower event, whereas infants in the ingroup condition detected a violation in all but the 1-block event. Infants thus expected E2 to knock down fewer blocks when retaliating against ingroup E1 (1 block) as opposed to outgroup E1 (2 towers) for the theft of a cracker.
Experiment 2 sought to rule out the possibility that these expectations were independent of the initial theft. Here, E1 no longer stole E2’s cracker, but E2 still knocked down 1 block, 1 tower, or 2 towers, in what was now unprovoked as opposed to retaliatory harm. Infants in the outgroup condition detected a violation in all but the 1-block event, whereas infants in the ingroup condition detected a violation in all three events. These results differed significantly from those of Experiment 1 and indicated that (a) infants expected E2 to direct more harm toward E1 when provocation had occurred (Experiment 1) than when no provocation had occurred (Experiment 2), but (b) group membership modulated these expectations, with less retaliatory and less unprovoked harm being deemed acceptable when directed at ingroup E1 as opposed to outgroup E1.
Experiment 3 sought to replicate the results of Experiment 1 and tested 2.5-year-old toddlers using a slightly different procedure: E1 and E2 were now puppets (e.g., a monkey and a duck in the outgroup condition, two monkeys in the ingroup condition), and E2 had 3 apples to match E1’s 3 towers. As before, toddlers expected E2 to knock down 2 towers when retaliating against outgroup E1, but to knock down only 1 block when retaliating against ingroup E1.
Together, these results indicate that group membership modulates both infants’ and toddlers’ expectations about retaliatory harm: Less retaliation is expected for the same misdeed when perpetrated by an ingroup as opposed to an outgroup wrongdoer.

Authors

• Fransisca Ting, UIUC
  Presenting Author

• Zijing He, Sun Yat-sen University
  Non-Presenting Author

• Renee Baillargeon, University of Illinois at Urbana-Champaign
  Non-Presenting Author