Conceived and designed the experiments: MFHS JAS. Performed the experiments: MFHS. Analyzed the data: MFHS JAS. Contributed reagents/materials/analysis tools: MFHS JAS. Wrote the paper: MFHS JAS.
The authors have declared that no competing interests exist.
Human cooperation is a key driving force behind the evolutionary success of our hominin lineage. At the proximate level, biologists and social scientists have identified other-regarding preferences – such as fairness based on egalitarian motives, and altruism – as likely candidates for fostering large-scale cooperation. A critical question concerns the ontogenetic origins of these constituents of cooperative behavior, as well as whether they emerge independently or in an interrelated fashion. The answer to this question will shed light on the interdisciplinary debate regarding the significance of such preferences for explaining how humans become such cooperative beings. We investigated 15-month-old infants' sensitivity to fairness, and their altruistic behavior, assessed via infants' reactions to a third-party resource distribution task, and via a sharing task. Our results challenge current models of the development of fairness and altruism in two ways. First, in contrast to past work suggesting that fairness and altruism may not emerge until early to mid-childhood, 15-month-old infants are sensitive to fairness and can engage in altruistic sharing. Second, infants' degree of sensitivity to fairness as a third-party observer was related to whether they shared toys altruistically or selfishly, indicating that moral evaluations and prosocial behavior are heavily interconnected from early in development. Our results present the first evidence that the roots of a basic sense of fairness and altruism can be found in infancy, and that these other-regarding preferences develop in a parallel and interwoven fashion. These findings support arguments for an evolutionary basis – most likely in dialectical manner including both biological and cultural mechanisms – of human egalitarianism given the rapidly developing nature of other-regarding preferences and their role in the evolution of human-specific forms of cooperation. Future work of this kind will help determine to what extent uniquely human sociality and morality depend on other-regarding preferences emerging early in life.
Since Darwin, the evolutionary emergence and stability of human cooperation – which presents an outlier in the animal kingdom in terms of its scale – has puzzled biologists and social scientists
In addition to recognizing ultimate mechanisms that explain why and under which conditions cooperative behaviors are adaptive, a critical charge in building a scientific understanding of human cooperative tendencies is identifying psychological dispositions and traits that enable the operation of such mechanisms in the first place. As such, empirical research using psychological methods is very important for understanding how humans become such cooperative beings over the course of ontogeny. Recently, a range of prosocial dispositional attitudes or “other-regarding preferences” have been identified and promoted as likely candidates to explain why human cooperation has been maintained and developed to a large scale
With respect to fairness, several studies in experimental economics using bargaining games suggest that adults consider fairness issues in their decision making
Past work indicates that other-regarding preferences may emerge fairly late in ontogeny, suggesting the need for a protracted period of socialization. For instance, when required to distribute goods between themselves and a recipient, children do not distribute goods equally until roughly middle childhood
Similarly, experimental evidence suggests that sharing tendencies also develop later in childhood. For example, a recent study demonstrated that it is not until 25 months of age that toddlers voluntarily share resources with an adult who makes her desire explicit
The current experiment investigated the emergence of sensitivity to fairness, and the willingness to share goods altruistically, in 15-month-old infants. Despite the work discussed above, there are several reasons to believe that such other-regarding preferences may emerge early in the course of development. At an evolutionary level such preferences may have been crucial for our hominin ancestors to enable and maintain cooperation in small groups, and later, in larger groups of genetically unrelated individuals, to introduce norms (e.g., how to share spoils after a group hunt) that fostered group cohesion, and to motivate group members to enforce those norms. At a developmental level, infants often evaluate events on the basis of underlying social and physical principles, before they can produce behavior consistent with these principles
We investigated 15-month-old infants' sensitivity to third-party fairness using a resource distribution task in a violation-of-expectation (
The study followed a within-subjects design with each infant tested first in the VOE paradigm and subsequently in the sharing task.
In the VOE paradigm, infants watched two movies in which an actor allocated continuous (milk) or discrete (crackers) resources to two recipients in a 23-s distribution phase; the outcome of the resource distribution was occluded by a black screen (see
In the introductory phase of the crackers movie (milk movie), the distributor greeted the recipients, lifted the bowl of 4 crackers (the pitcher with 10 ounces of milk) while saying “Yummy!” (A). Then, the recipients moved their plates (glasses) toward the distributor asking “Please?”. During the distribution phase (B), the distributor then allocated crackers (milk; exact amount occluded by a black screen) to each recipient via a single movement to each side. The distributor then held up the empty bowl (pitcher) up saying “All gone!” (C). During the test phase, a still frame depicted a fair (D; crackers: 2 crackers each; milk: 5 oz each) and an unfair (E; crackers: 3 crackers vs. 1 cracker; milk: 8 vs. 2 oz) outcome in succession (order counterbalanced), with the actors displaying neutral facial expressions, whereas the post-test phase showed the same displays devoid of a social context, hence symmetrical (F) and asymmetrical (G) outcomes in succession (order counterbalanced).
Since preliminary analyses yielded no effect of movie type (crackers, milk) on infants' looking to test and post-test outcomes, the data were collapsed across movie type. All statistical tests were performed two-tailed. Analyses focused on looking times collapsed across both movies (
Infants' mean looking times to the test and post-test outcomes in the VOE paradigm are depicted in
Infants' attention to the 23-s distribution phase during the first (
Two experimenters (one familiar and one unfamiliar) conducted the sharing task. Two toys were placed on the wooden table 54 cm apart (position counterbalanced). In the preference phase (
In the preference phase (A), the infant chose one of the two toys (only one reachable at a time) - her preferred toy. After the infant had taken one toy, the familiar experimenter gave the infant the other (non-preferred) toy (not depicted here). In the request phase (B), an unfamiliar experimenter asked for a toy while looking directly at the infant. Here, the infant shares her preferred toy (“altruistic sharing”).
Regarding a potential relation between infants VOE performance and their sharing status, it is important to note that if one assumes that infants merely used a formula in the VOE paradigm (e.g., a 15% time difference leads to a 15% difference ratio in outcomes), we would expect no interrelation between infants' VOE preference and their morally relevant sharing behavior, since moral issues of fairness in the VOE paradigm would be irrelevant. To examine the relation between infants' sharing behavior and their VOE performance, we performed two sets of analyses. In the first analysis, we contrasted altruistic sharers and selfish sharers/non-responders. This analysis assumes that infants who shared the preferred toy were motivated by altruistic concerns, whereas those who shared the non-preferred toy, or did not respond at all, were motivated by selfish concerns. Ninety-two percent of altruistic sharers looked longer to the unfair outcome (paired sign test,
VOE preference | ||||
Unfair | Fair | Total | ||
|
Altruistic | 11 | 1 | 12 |
Selfish/no response | 10 | 16 | 26 | |
|
21 | 17 | 38 |
The above analysis assumes that both selfish sharers and non-responders were motivated by the same factor: a reluctance to share the preferred toy. However, it is alternately possible that non-responders were comprised of a heterogeneous group whose performance on the sharing task was governed by factors ancillary to selfish or altruistic concerns. Indeed, there are multiple reasons that infants may fail to respond in our task: because they do not understand the experimenter's request, because they are distracted or inattentive, because they are struggling to decide which toy to select under the allotted time constraints, and/or because they suffer from stranger anxiety.
To investigate whether non-responders differed from responders (i.e., altruistic and selfish sharers) on at least one of the dimensions listed above, we coded all infants for behaviors indicative of stranger anxiety
Thus, in our second analysis we directly contrasted altruistic sharers' with selfish sharers' VOE performance. We found a significant association between sharing status and VOE performance (
VOE preference | ||||
Unfair | Fair | Total | ||
|
Altruistic | 11 | 1 | 12 |
Selfish | 2 | 12 | 14 | |
|
13 | 13 | 26 |
The current study provides the first evidence that by at least 15 months of age, human infants possess the rudiments of a sense of fairness in that they expect resources to be allocated equally when observing others (third-party fairness). Furthermore, our findings suggest that sharing non-essential resources (at high or low personal costs) with an unfamiliar adult is also prevalent at this age, which dovetails with natural observations of sharing behavior with familiar adults in young infants
Were infants merely responding to the test events as violations of non-moral conventions (e.g., that goods are usually divided into equal amounts), there would be no reason to expect a tight interconnection between infants' evaluations of the test events and their prosocial behavior. Thus, we suggest that infants evaluate events along morally relevant dimensions, and not just according to whether such events are consistent or inconsistent with conventional norms. Moreover, the fact that infants' sensitivity to fairness and altruism were interrelated not only lends support to theoretical claims of a close alliance in other-regarding preferences
Taken together, the present findings strike a new path in social-moral development, because they suggest that in addition to instrumental helping
With respect to the evolution of cooperation, one mechanism, indirect reciprocity
The treatment of participants in this paper was in accordance with the ethical standards of the American Psychological Association. The subjects' parents provided written informed consent, and the study was approved by the Institutional Review Board at the University of Washington, Seattle, WA (Application #24231).
Forty-seven healthy full-term 15-month-old infants (
Movies were recorded with three female actors (a distributor and two recipients), and presented on a 21-inch television monitor. The props used in the movies were four Graham crackers, white plates, a transparent bowl (crackers movie), and milk, transparent glasses (volume of 10 oz.), and a transparent glass pitcher (milk movie), respectively.
A green Lego brick (4 cm width, 7 cm length) and a female doctor toy (4 cm width, 8 cm height) were used as resources, and a wooden table (38×92 cm) served as the location from which infants could choose one toy.
Infants sat on their parent's lap (80 cm from the display). Parents were instructed to remain silent, and to close their eyes during the experiment. The movies consisted of an initial greeting by a (fourth) female actor (to attract infants' attention), a familiarization phase (including an introductory phase and a 23-s distribution phase), two test trials (a fair and unfair outcome), and two post-test trials (a symmetrical and asymmetrical outcome). Movie order, test outcome shown first, first distribution side, and location (right, left) of the unfair/asymmetrical outcome were counterbalanced across infants.
The procedure of the sharing task is outlined in the
All sessions were coded on-line, recorded and additionally coded from videotape by a second independent observer.
The minimum-look criterion (accumulated) to the distribution phase was 18.4 s (80% of 23 s). Infants' looking to the test and post-test trials was timed on-line until they looked away for 1 consecutive second (maximum trial length: 30 s; minimum-look criterion: 2 s). The second independent observer coded all subjects for reliability (interobserver agreement: 95%).
The secondary observer coded which toy infants chose, whether infants shared a toy or not, which toy (preferred vs. non-preferred) infants handed the second unfamiliar experimenter (interobserver agreement: 100%). The secondary observer additionally coded subjects for behaviors indicative of stranger anxiety in the request phase (concerned/fearful facial expressions, avoiding looking at the requestor, crying, looks to the parent; interobserver agreement: 94%).
Portions of these data were presented at the 2011 meeting of the Society for Research in Child Development, Montreal, Quebec. We would like to thank E. Blumenthal, J. Yun, N. North, and B. Blackman for their support in stimulus creation, coding, and data collection. We also thank E. Blumenthal for helpful comments on an earlier draft of this manuscript, and M. Schreiber for drawing