Precision in Procedure: The Influence of Exactness on Fairness Judgements

• In Event: How Much is Fair: Integrating the Development of Numerical Cognition with Children's Understanding of Fairness

Abstract

To distribute resources in a fair way, children not only have to decide what distribution to produce, but also how to produce it. For young children, generating different types of distributions may be challenging, particularly before they understand how to count. Consistent with this, recent research shows that children’s ability to count influences how they act in fairness tasks (Jara-Ettinger, et al., 2016). Yet, it is unknown if children only see counting as a way to solve problems in first-person situations, or if they also consider counting when evaluating the fairness of other agents distributing resources.

We explored this question in a paradigm where four- to six-year-olds learned about two students who cleaned up a classroom; one student worked really hard, while the other student did not work very hard. Participants then watched two different teachers give seven cookies to the hard worker and three cookies to the not-hard worker. While both teachers always produced identical outcomes, the way they produced this outcome varied.

In Experiment 1 we first tested the role of attention. Both teachers produced identical distributions by splitting the cookies into two piles in an approximate way. However, only one of the teachers looked at the cookies when distributing them, whereas the other teacher was distracted (Fig.1a-b). Participants (n=48) judged the attentive teacher as more fair than the inattentive one (91.7%; 95% CI:85.4-100%; Fig.2) revealing a sensitivity to attention.

Experiment 2 contrasted the attentive teacher from Experiment 1 with a new teacher that distributed cookies by counting (Fig.1b-c). Participants (n=48) now judged that the counting teacher was more fair than the attentive one (72.92%; 95% CI:60.42-85.42%; Fig.2), revealing a sensitivity to counting beyond attention.

Participants in Experiment 2 may have favored the counting teacher simply because she put more effort into the task, instead of considering that she counted to produce a precise distribution. We tested this possibility in Experiment 3. Similar to Experiment 2, participants (n=40) watched an attentive and a counting teacher produce identical distributions. The attentive teacher, however, first had to move ten objects resting on top of the cookie box (thus matching the number of actions and effort across teachers; Fig.1c-d). Participants continued to identify the counting teacher as more fair (70%; 95% CI:57.5-85%; Fig.2), showing a particular sensitivity to counting beyond effort alone.

Finally, Experiment 4 tested if children see counting as intrinsically fair, or as revealing an intent to be precise. The experiment was identical to Experiment 2 with the difference that both teachers now gave three cookies to the hard-working student and seven cookies the not hard-working student. Participants (n=48) now judged that the counting teacher was less fair than the attentive one (68.75%; 95% CI:56.25-83.28%; Fig.2).

Collectively, our results show that children make nuanced fairness judgements that go beyond outcomes alone, and that counting distinctly reveals an agent’s intent to produce a precise distribution. Logistic regressions revealed no significant age effects. These findings show that the connection between number and fairness is rich and emerges early in development.

Authors

• Colin Jacobs, Yale University
  Presenting Author

• Rosie Aboody, Harvard University
  Non-Presenting Author

• Julian Jara-Ettinger, Yale University
  Non-Presenting Author