Poster #99 - Children’s Generalizations of Negative Information About a Group to Its Novel Members

• In Event: Poster Session 17

Abstract

Past research shows that children tend to make positive inferences about a new person in the absence of any other information (e.g., Aboud, 1988), however intergroup attitudes and testimony may change this tendency (e.g., Dunham et al., 2011).

In this preregistered study, we ask 1) whether children expect a member of a social group to behave in a similarly negative way to the behaviors of the other members of that group, 2) if children’s initial attribution about the novel member of that social group (i.e., the perpetrator group) can be modified by an informant’s testimony in the opposite direction, and 3) whether the identity of the informant (i.e., being a member of the perpetrator, victim or a neutral group) would change the effectiveness of the testimony.

Children (N = 111, 56 female, Mage = 6.91, Range = 5.8-8.1) were introduced to two novel groups and were presented with two scenarios in which members of one of the groups exhibited a negative behavior toward the other group (e.g., making fun of some members). Children were asked if another member from the perpetrator group (i.e., target) would also engage in one of these negative behaviors, and whether they were a little or very sure of their answer. Then, an informant, who was either from the victim, the perpetrator, or a new neutral group, presented a contrasting opinion about the target. Children were asked again about their opinion about the target and how certain they are about it. Children’s answers were coded as -1 or +1 (for negative and positive attributions, respectively) and these were multiplied by 1 (a little sure) or 2 (very sure). We calculated a difference score by subtracting the second response from the first. We also coded response changes in terms of the valence of the attribution (e.g., -2 to 1) and certainty (e.g., -2 to -1).

Data collection is ongoing (target N = 120), but our preliminary results suggest that children’s positivity bias toward novel targets after having witnessed negative behaviors of other group members was only marginally significant (X² (1, N = 111) = 3.252, p = .071). Overall, children tended to change their certainty after the conflicting testimony of the informant (X² (1, N = 111) = 29.27, p < .0001), but only children whose first attribution was negative changed the valence of their attributions after conflicting testimony (p = .003). Finally, informants’ group membership had a significant effect only for those children, who initially made a negative attribution about the target (F(2, 43) = 3.037, R² = 0.083, p = 0.05). Specifically, the informant from the perpetrator group had a greater influence on children’s attributions from negative to positive compared to the informant from the neutral group (p = 0.04). Differences between the perpetrator and the victim informants or the victim and the neutral informants were non-significant (ps > .2).

These findings might help us better understand the mechanisms that play a role in the formation and the malleability of children's social biases.

Authors

• Halide Sena Koçyiğit, Boğaziçi Üniversitesi (BOUN)
  Presenting Author

• Pinar Aldan, Yale University

• Gaye Soley, Boğaziçi Üniversitesi (BOUN)