Edited by: Susanne Leiberg, University of Zurich, Switzerland
Reviewed by: Jessica Sommerville, University of Washington, USA; Nikolaus Steinbeis, Max-Planck Society, Germany
*Correspondence: Alex Shaw, Social Cognitive Development Lab, Department of Psychology, Yale University, 527 Chapel St. Apt A2, New Haven, CT 06511, USA e-mail:
This article was submitted to Frontiers in Emotion Science, a specialty of Frontiers in Psychology.
This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in other forums, provided the original authors and source are credited and subject to any copyright notices concerning any third-party graphics etc.
Fairness concerns guide children's judgments about how to share resources with others. However, it is unclear from past research if children take extant inequalities or the value of resources involved in an inequality into account when sharing with others; these questions are the focus of the current studies. In all experiments, children saw an inequality between two recipients—one had two more resources than another. What varied between conditions was the value of the resources that the child could subsequently distribute. When the resources were equal in value to those involved in the original inequality, children corrected the previous inequality by giving two resources to the child with fewer resources (Experiment 1). However, as the value of the resources increased relative to those initially shared by the experimenter, children were more likely to distribute the two high value resources equally between the two recipients, presumably to minimize the overall inequality in value (Experiments 1 and 2). We found that children specifically use value, not just size, when trying to equalize outcomes (Experiment 3) and further found that children focus on the relative rather than absolute value of the resources they share—when the experimenter had unequally distributed the same high value resource that the child would later share, children corrected the previous inequality by giving two high value resources to the person who had received fewer high value resources. These results illustrate that children attempt to correct past inequalities and try to maintain equality not just in the count of resources but also by using the value of resources.
Fairness is certainly important to human society (Boyd and Richerson,
One goal of the current research is to examine whether children focus on the norm of sharing equally themselves, or on trying to make the overall distribution of resources equal by correcting existent inequalities. If they do correct previous inequalities in order to equate outcomes, a second goal of the current research is to investigate whether children take resource value into account when trying to minimize inequality between others. Do they correct inequalities by trying to make the count of resources equal—giving cars to make up for unfair ticket-giving—or do they attempt to make the overall value of the resources as equal as possible?
We know that children are biased toward equal distribution of resources, but there has been very little research on how children respond to existent inequalities. Research with infants using looking time measures suggests that by the second year of life infants expect resources to be distributed equally between two agents, as long as both agents are highlighted as possible recipients (Geraci and Surian,
We also know very little about how children divide resources that differ in value, despite the fact that many forms of exchange involve resources that are not equal in value. Most research on children's and even adults' equality concerns has focused on decisions that involve distributing a single type of resource, for example, distributing a sum of money or a sum of cookies rather than having a person divide some money and some cookies (for reviews, see Damon,
What little work that has been done on children's understanding of value has examined how children's preference for a resource influences their willingness to give resources to another person. We know that children, like adults, demonstrate preferences for some goods over others (Harbaugh et al.,
In Experiment 1, we first investigated whether children correct existent inequalities in order to minimize inequality in outcomes. Children were asked to share two resources with two non-present recipients who had already received resources from the experimenter. The experimenter gave three resources to one of the recipients and one to the other recipient. If children try to make outcomes equal, then they should give both erasers to the recipient with fewer resources (giving unequally but correcting the inequality) rather than giving one to each recipient (maintaining the inequality by giving equally themselves). We investigated this question in 6- to 8-year-old children because past research has demonstrated that it is at this age that children become comfortable sharing unequally with third parties, at least based on merit (e.g., Hook and Cook,
If children do give more resources to those who currently have fewer resources, it would be unclear if they do so in order to keep the count of the resources equal, or in order to keep the value of the resources equal. To investigate if children use value to determine how to equalize outcomes, we included two conditions in which children were sharing resources that were slightly more valuable (jar of Play-Doh, Medium Value Condition) or much more valuable ($20 bill, High Value Condition) than the resources that were initially shared unequally by the experimenter. If children want to keep the count of resources equal, then children should respond similarly in all conditions, by giving two resources to the recipient with fewer resources and thus equalizing the count of resources. If instead children care about keeping the value of resources as equal as possible, then, as the value of the resources to be shared increases, children should become increasingly likely to share equally themselves by giving one resource to each recipient. We investigated these questions in this study.
Participants included 84 children aged 6 to 8 years old. Of these participants, 28 were in the Equal Value Condition (
Two buckets were placed in front of the participant and the experimenter said (modeled on Shaw and Olson's ( Thanks for playing this game with me. Earlier today, two kids named Mark and Dan did a great job cleaning up their room and we want to give them erasers as a prize. The problem is I don't know how much to give them; can you help me with that? We are going to decide how many erasers Mark and Dan will get. Mark's erasers go in this bucket and Dan's erasers go in this bucket. We have six erasers. I am going to give these four erasers and you are going to give these two erasers. I'll go first. We have one for Mark, one for Dan. One for Mark, and one more for Mark. Now it's your turn; here are two erasers. Give them however you want.
Each time Mark or Dan's name was used, the experimenter pointed to the corresponding bucket. During the allocation phase of the task, the experimenter placed an eraser into the corresponding bucket when noting who was receiving the eraser (Mark or Dan). The erasers were colorful and shaped like fun things children like, such as turtles, sports balls, and ice cream cones, and have been used in previous research on decision-making in children (Shaw and Olson,
In order to investigate the influence of value on children's decisions, we had two additional conditions in which children distributed resources that were slightly more valuable (Medium Value Condition) or much more valuable (High Value Condition) than the resources (erasers) that were shared unequally by the experimenter; all other aspects of the design of these conditions was identical to the condition described above. The slightly more valuable object in the Medium Value Condition was a 3 oz jar of Play-Doh, and the much more valuable resource in the High Value Condition was a $20 bill. Children could not make the value of resources equal in these conditions, since both Play-Doh and a $20 bill are presumably worth much more than two erasers (for empirical verification that children see these items as more valuable than erasers, see Experiment 4), but they could ensure that the inequality did not increase. Specifically, they could give one high value resource to each recipient rather than giving two to the person with fewer low value resources, if they were interested in maintaining the smallest overall inequality.
Because no children chose the option of giving two erasers to the person with more erasers, we conducted analyses with just the two strategies that children used—sharing equally by giving one to each recipient, or giving two to the person with fewer resources. We first conducted a 3 × 2 Yates-corrected χ2 test on children's responses in the Equal, Medium, and High Value Conditions, which revealed a main effect of condition, χ2 (2,
We then examined whether children's responses differed between pairs of conditions by conducting Yates-corrected χ2 tests. A 2 × 2 Yates-corrected χ2 test revealed that children in the Equal Value Condition were more likely to give two erasers to the disadvantaged recipient than children in the High Value Condition, χ2 (1,
We next conducted binomial tests to compare children's responses to chance responding. The binomial test on the Equal Value Condition revealed that children gave two to the recipient with fewer erasers (24 out of 28) more often than giving one to each recipient (4 out of 28),
Children corrected inequalities created by an experimenter, and did so by attempting to equate the value, not just the count, of the resources distributed. When children distributed resources that were equal in value to those shared unequally by the experimenter, children gave more resources to a recipient who had received fewer resources in order to correct the existent inequality. Children could have ignored the inequality that was created by the experimenter and simply focused on the norm of giving equally themselves, since we know from past research that children have a tendency to share equally with others (Damon,
We next asked whether, when children try to make outcomes equal, they try to simply make the count of resources equal, or whether they consider the value of the resources. Our results suggest that children do use value when deciding how to share. Children behaved differently when sharing resources that were much more valuable (e.g., $20 bills) than the unequally shared erasers, giving the resources equally themselves rather than attempting to correct the past inequality. Perhaps children distributed the more valuable resources differently because they realized that giving the disadvantaged child two $20 bills would actually make the outcome even more unequal, though now in the other recipient's favor.
Although we interpret these results as indicating that children minimize inequality between others by using value, this is not the only possible interpretation. One alternative possibility is that children were confused in the Medium and High Value Conditions because they were required to match distributions involving multiple resources. However, the fact that children differentiated between the Medium and High Value Conditions speaks against this alternative—the resources used in both the Medium and High Value Conditions were different from the resources that were distributed by the experimenter, yet children treated these two conditions differently, suggesting they used some sense of value to guide their decisions. However, this by itself does not provide enough evidence to rule out the possibility that children were confused in these conditions. Perhaps children were indeed confused in the Medium Value Condition, and only behaved differently in the High Value Condition because there is something special about money that makes children more likely to share equally or pay attention to the value of resources. Previous research with adults indicates that when people distribute money, as compared to other resources, they are likely to think in terms of market exchanges (DeVoe and Iyengar,
Participants included 56 children aged 6 to 8 years old. Of these participants, 28 were in the Lower Value Condition (
The procedure for Experiment 2 was very similar to that used in Experiment 1. Again the experimenter gave out four resources unequally, giving three to one recipient and one to the other, using the script described in Experiment 1. Then, the participant was told to share two resources with the two recipients. We did, however, make two changes. First, we used different resources in Experiment 2. In the Higher Value Condition, the experimenter gave out four lower value resources (four small fruit-flavored candies) and the participant gave out two higher value rewards (two full-sized chocolate candy bars). This method was similar to the Medium and High Value Conditions from Experiment 1, so we predicted a similar pattern of results—that participants would be less willing to give more resources to the person who had fewer resources. In the Lower Value Condition, the experimenter gave out high value resources (four full-sized chocolate candy bars), and the participant gave out two lower value rewards (two small fruit-flavored candies). If children's responses in the previous experiment were merely being driven by confusion about how to distribute a resource different than the one involved in the original inequality, or by money priming them to think about value, then they should respond at chance or give one lower value resource to each recipient as they did in the Medium and High Value Conditions from Experiment 1. However, if children try to equate value to minimize inequality of outcomes between others, then they should instead give two low value resources to the recipient who received fewer higher value resources. Giving more low value resources to the recipient with fewer resources would not make the distribution equal, but is the most equal option available to children. We deliberately used different types of candy because we wanted to ensure that children were responding to value, not merely thinking about the resources in terms of large and small quantities of the same resource (for empirical verification that children see the chocolate bars as more valuable than the small fruit candies, see Experiment 4).
A second change from Experiment 1 to Experiment 2 was that we now presented the resources on pieces of paper (5 × 8″) rather than placing them in buckets. We modified this aspect of the design to reduce the memory load required to complete this task.
Again, because very few children chose the option of giving two resources to the person with more resources (only one child who was in the Higher Value Condition and no children in the Lower Value Condition), we again conducted our analyses focusing on the strategies that children used—sharing equally themselves by giving one to each recipient, or giving two to the person with fewer resources
We next conducted binomial tests to compare children's responses to chance responding. A binomial test on children's choices in the Higher Value Condition revealed that children did not show a preference, with about half the children giving two to the person with fewer resources (15 out of 27) and about half of the children giving one to each recipient (12 out of 27),
We again found that children correct previous inequalities in order to minimize inequalities in outcomes between recipients, and do so by using the value of the resources at their disposal. When children were presented with an inequality involving high value resources, but only had a few low value resources with which to address it, children gave two to the person with fewer resources since this was the best way to minimize inequality. However, when children were presented with an inequality involving low value resources, but only had a few high value resources with which to address it, children were much less likely to give two to the person with fewer resources. Importantly, in both conditions children were dividing resources that were different from the resources that were distributed by the experimenter, so the results cannot be explained by confusion involving the distribution of different resources (which was common to both conditions). In fact, the same resources were used in both conditions; what differed between conditions was which resource was distributed by the experimenter and which was distributed by the participant. These results suggest that children focus on trying to equalize outcomes, and that they do so by using the value of resources.
Although the results thus far are consistent with children using value to determine how to minimize inequality in outcomes, children could be using an even simpler heuristic—size of resource. In Experiments 1 and 2, the high value resource was physically larger than the low value resource, and so children may have been using resource size, not value, to guide their decisions. In Experiment 3, we dealt with this confound by matching the volume and surface area of the high and low value resources.
Participants included 56 children aged 6 to 8 years old. Of these participants, 28 were in the Lower Value Condition (
The procedure for Experiment 3 was the same as Experiment 2, except that we used different resources: chocolate bars, and pieces of cardboard cut to the same size as the chocolate bars. In the Lower Value Condition, the experimenter gave out high value resources (four chocolate bars; three to one recipient and one to the other) and the participant gave out two lower value rewards (two pieces of cardboard). In the Higher Value Condition, the experimenter gave out lower value resources (four pieces of cardboard; three to one recipient and one to the other) and the participant gave out two higher value rewards (two chocolate bars). If children in the previous experiments were trying to equate the volume or surface area of the resources, then children should behave similarly in the Higher and Lower Value Conditions here. However, if children in the previous experiments were trying to minimize inequality in outcomes by using value, then they should be more likely to share equally themselves by giving one resource to each recipient when distributing the higher value reward as opposed to the lower value reward (for empirical verification that children see the chocolate bars as more valuable than cardboard, see Experiment 4).
Again, because no children chose the option of giving more resources to the recipient with more resources, we conducted our analyses on children's two strategies of giving more to the person with fewer resources and giving one to each recipient. A Yates-corrected χ2 test revealed that children in the Higher Value Condition were more likely to give one resource to each recipient than children in the Lower Value Condition, χ2 (1,
We next conducted binomial tests to compare children's responses to chance responding. A binomial test on children's choices in the Higher Value Condition revealed that children did not show a preference, with about half the children giving two to the recipient with fewer resources (13 out of 28) and about half of the children giving one to each recipient (15 out of 27),
These results again indicate that children are motivated to create equal outcomes, not just to give equally themselves, and that they use value, not just the volume or surface area of resources, to decide how to create equal outcomes for others. When a resource was of lower value than the resources involved in the original inequality, children gave more to the recipient who received fewer resources originally in order to correct the previous inequality. However, children were much less likely to give more to the recipient with fewer resources if the resources they were distributing were more valuable than those involved in original inequality, presumably because they understand that this would make things more unequal.
However, one limitation of Experiments 1 through 3 is that we did not have an empirical measurement of value. We deliberately chose resources that seemed more valuable to adults; however, we do not know if children actually think these resources are more valuable. In Experiment 4 we ask children explicitly about which items they think that another child would prefer and how many of the less preferred items they think one would need to trade in order to get the more preferred item.
One other open question from the previous experiments is whether children distributed high value resources differently then low value resources because they treat high value resources differently in general or because they noticed that the high value resources were more valuable than the originally distributed resources. Perhaps children just maintain the status quo by sharing equally when they are given certain resources to share, regardless of the value of resources shared by the experimenter. To examine this possibility in Experiment 4 we had children distribute the high value resource from Experiment 3 (a chocolate bar) in a situation in which equal sharing was not the option that minimized inequality—where an experimenter shared three chocolate bars with one recipient and one chocolate bar with the other. If children treat certain resources differently regardless of context, then they should give one chocolate bar to each recipient as they did in Experiment 2 and 3 when sharing chocolate bars. However, if what matters is the relationship between the value of resources already distributed and the resource children are sharing, then they should now give two chocolate bars to the person with fewer resources because this would minimize inequality between the two recipients.
Participants included 28 children aged 6 to 8 years old (
The procedure for Experiment 4 was the same as Experiment 1 Equal Value Condition, except that the equal value resource was now the chocolate bars from Experiments 2 to 3 rather than erasers. That is, the experimenter gave out four chocolate bars, three to one recipient and one to the other, and the participant gave out two of the same kind of chocolate bar. In the previous experiments chocolate bars were treated as a high value resource in comparison to small fruit candies and cardboard. Therefore, if children are simply more inclined to maintain the status quo when distributing objectively valuable resources like chocolate bars, then we should see children giving one chocolate bar to each recipient as they had in Experiments 2 and 3. However, if what children are attempting to do is to equate value, then we should see them giving two candy bars to the recipient with fewer candy bars.
After completing the Equal Value Condition, children completed an explicit measure of value. We asked children to decide which resource they thought Mark would prefer, resource
Again, because no children chose the option of giving more resources to the recipient with more resources, we conducted our analyses on children's two strategies of giving more to the recipient with fewer resources and giving one to each recipient. We conducted binomial tests to compare children's responses to chance responding. A binomial test on children's choices in the equal value condition revealed that that children chose to give two to the recipient with fewer resources (24 out of 28) more often than giving one to each recipient (4 out of 28),
We next conducted binomial tests on children's responses to which resource Mark would prefer. Children thought that Mark would prefer: a jar of Play-Doh to an eraser (26 out of 28),
These experiments demonstrated that 6- to 8-year-olds are more concerned with making the outcome of a resource distribution equal than with giving equally themselves. They also demonstrated that children consider value when responding to inequalities. Experiment 1 showed that children will give unequally themselves in order to minimize inequality of outcome. Children gave two resources to the recipient with fewer resources so that both recipients would have three resources, rather than giving equally themselves and maintaining the inequality. This result is consistent with past research demonstrating that children will give unequally in some circumstances, such as when others have done more work (Damon,
Experiments 1 and 2 also investigated what measure children use to determine how best to minimize inequality. These experiments illustrated that children use the
It is worth noting that while children became less likely to give both resources to the recipient with fewer resources as the value of the new resources increased, in Experiments 2 and 3 about half the participants still attempted to equate resource count rather than resource value when sharing the high value resource (large chocolate bar). It is unclear why children gave mixed responses in this case, though there are several possibilities. One possibility is that some children placed different value on the items they were asked to share. If children thought the chocolate bars were about as valuable as the fruit candies or cardboard, then it would be unsurprising that they attempted to equate count rather than value. However, a more likely possibility is that children did not know which norm to apply to this situation and so were forced to choose between two conflicting norms: should I equalize the count or value of the resources? This conflict in norms may have made children confused about what to do and led to their chance responding when distributing the higher value rewards. However, what is important about these results is that children did differentiate between distributing resources that had higher and lower value than the original inequality, suggesting that at least some children take resource value into account when deciding how to minimize inequality in outcomes between others.
The current findings are interesting to consider in light of recent work demonstrating that children are fair partly in order to signal to others that they are fair. Shaw et al. (
The results of our experiments demonstrate that the value of resources influences children's sharing behavior, but they do not address how children determine the value of resources in the first place. The first strategy that children likely use to determine value is to simply use their own preferences as a guide for how to share with others. That is, they know what they like, and think that the things they like are valuable and that the things they dislike are not valuable. This strategy is likely a large part of children's early understanding of value, but as they get older they may use more sophisticated variables to determine how resources are valued. One possibility is that children use some aggregate sense of others' preference for a resource, analogous to the adult concept of demand—recognizing that the more others want a resource, the more valuable it is (Baumol,
Now that we know that children can use value to guide their equality judgments, we can investigate whether or not children use value in other domains such as trade. Trade is ubiquitous in modern society and simpler forms of bartering were also very prevalent before the advent of currency (Fagan,
Despite remaining questions, the current research demonstrates that children do not treat all inequalities equally—they use resource value, rather than just resource count, when deciding how to share with others.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
We would like to thank Nina Slywotzky, Anna Merrill, Melanie Fox, Danielle DeLee, Matt Choy, Zoe Liberman, Suzanne Horowitz, Alia Martin, Kelcey Wilson, and Alex Chituc for assistance running the participants in these studies. This research was supported by a grant from the University of Chicago's ARETE Initiative/A New Science of Virtue Program.
1We analyze the results here without including the one child who gave the non-standard response of giving more to the person with more resources, but the pattern of results remain the same if we conservatively run the analyses counting this child as having given more to the person with fewer resources.