Babies Spotted Selfish Sharing Long Before They Can Talk, Study Finds

Long before children learn to say the word “mine,” they appear to have already worked out how the world’s resources ought to be divided. A new study of infants in their second year of life suggests that babies as young as 14 months not only notice when a sharing arrangement is unequal, but also seem to distinguish between a person who keeps the lion’s share for themselves and one who generously hands most of it away. The findings, published in Frontiers in Developmental Psychology, offer some of the clearest evidence yet that the seeds of human fairness may be planted far earlier than researchers once assumed.

The research, conducted by Ha Na Yoo of the University of Utah Asia Campus and Hyun-joo Song of Yonsei University in South Korea, tackled a puzzle that has lingered in developmental psychology for years. Most previous studies of infant fairness relied on so-called third-party scenarios, in which a neutral distributor hands out resources to two other people. In those experiments, infants reliably stared longer at unequal splits than at equal ones, a classic sign of surprise that researchers read as a violated expectation. But real life is rarely so tidy. When a toddler watches a sibling divide a pile of crackers, the person doing the dividing is usually one of the people eating them. The question was whether infants apply the same equality standard when the divider’s own interests are on the line.

That distinction matters more than it might first appear. In a third-party allocation, fairness is essentially a bookkeeping problem: does each recipient end up with the same amount? In a first-party sharing situation, the distributor pays a personal cost with every item given away. Sharing therefore involves ownership, self-interest, and the temptation to keep more than one gives. If infants merely applied a generic equality rule to any distribution they witnessed, they would treat all unequal outcomes as violations. But if their expectations are more socially sophisticated, they might react differently to a selfish split than to a selfless one, even when the arithmetic is identical.

To find out, the team recruited 46 infants aged 14 to 19 months from the metropolitan area of South Korea and seated them roughly 60 centimeters from a small live-action stage. The setup followed the violation-of-expectation paradigm, a workhorse method in developmental science. Infants tend to look longer at events that contradict what they expect, so longer gazes toward one outcome are taken as evidence that the outcome surprised them. Caregivers held their babies on their laps with eyes closed, ensuring that no parental cues could steer the infants’ attention, while two trained observers who were blind to the experimental conditions recorded looking times through hidden peepholes with inter-observer agreement around 90 percent.

In the first experiment, infants watched a familiarization scene in which two actors sat side by side, each with a tray. The actor on the right, wearing a red shirt, held six red cubes on a plate; the actor in blue had none. The cubes were deliberately presented on plates to emphasize that they were possessable goods belonging to one person. Then came the test events. In the equal event, the sharer moved three of her six cubes to her partner’s tray, one by one, each transfer taking two seconds. In the unequal event, she handed over just one cube out of six, keeping five for herself. Both events lasted 12 seconds, and the order of presentation was counterbalanced across infants.

The results were striking. Infants in the experimental condition looked reliably longer at the unequal event, averaging about 33 seconds of attention compared with roughly 24 seconds for the equal event, a statistically significant difference. Crucially, the researchers included a control condition designed to rule out a mundane explanation: perhaps babies simply prefer visual symmetry, with equal numbers of objects on two trays, regardless of any social meaning. In the control condition, a single actor performed the exact same movements, shifting three cubes or one cube from one tray to another, but no second person was present to receive them. Under those asocial circumstances, infants’ looking times for the symmetric and asymmetric outcomes did not differ at all. The asymmetry itself held no special interest; only the social act of unfair sharing did.

That left the researchers with a deeper question. Was the infants’ surprise in the sharing condition driven purely by a violation of the equality norm, or did it reflect an expectation of prosocial behavior, a sense that a person ought to give generously to another? If infants expect altruism, then an unequal split that favors the recipient, with the sharer keeping only one cube and giving away five, should be just as unsurprising as an equal split, or perhaps even more expected. If infants care only about numerical equality, however, that generous-but-unequal outcome should register as a violation just as strongly as the selfish one.

Experiment 2 put this to the test with 27 additional infants in the same age range. The procedure mirrored the first experiment, but the unequal event was reversed: the sharer now gave five of her six cubes to the recipient, keeping only one for herself. The researchers then compared these infants’ responses with those from the experimental condition of the first experiment using a repeated-measures analysis of variance. The interaction between event type and experiment was significant, and follow-up comparisons showed that infants in the second experiment looked equally long at the equal and the altruistic-unequal events, with no reliable difference between them. Unlike their peers in the first experiment, these babies did not treat the unequal outcome as a violation, even though the numerical ratio of the split, one versus five, was the mirror image of the selfish case.

Taken together, the two experiments sketch a picture of early moral cognition that is more nuanced than a simple equality detector. Infants in their second year appear to flag selfish-unequal sharing as norm-violating while tolerating, or at least not objecting to, unequal sharing that benefits the other person. The authors suggest that infants may process not just the numerical division of resources but also the intentions behind the allocation, distinguishing selfish offers from selfless ones. This interpretation aligns with recent theoretical work arguing that distribution and sharing are psychologically distinct: third-party distribution primarily invokes outcome equality, whereas first-party sharing inherently involves ownership, personal cost, and the regulation of self-interest, opening the door to richer evaluations of generosity and selfishness.

The findings also plug into a broader developmental framework. Researchers such as Kathleen McAuliffe and colleagues have proposed that human fairness develops from an early sensitivity to unequal outcomes toward increasingly explicit, norm-based concerns about cooperation across childhood. Studies have shown that children as young as three protest unequal allocations and that older children increasingly engage in costly fair behavior even when it conflicts with self-interest. The new results suggest that some of the cognitive foundations for that trajectory, including a sensitivity to the social meaning of unequal splits, may already be in place during infancy. At the same time, the authors urge caution: looking-time measures reveal expectations, not explicit moral judgments, and later fairness behavior draws on executive function, strategic reasoning, and culturally shaped norms. The team also notes limitations, including the use of neutral red cubes whose desirability was not independently verified, a somewhat constrained sharing scenario, and a cross-sectional design that cannot trace developmental change over time. Future work with desirable resources, richer sharing cues, and longitudinal designs will be needed to map how these early expectations about selfish and selfless sharing mature into the egalitarian instincts that shape human cooperation later in life.

Subject of Research: Infants' expectations about fair and unfair resource distribution in first-party sharing situations

Article Title: Infants' expectations about resource distributions in a sharing situation

Article References: Yoo, H. N., & Song, H.-J. (2026). Infants' expectations about resource distributions in a sharing situation. Frontiers in Developmental Psychology, 4, Article 1800452. https://doi.org/10.3389/fdpys.2026.1800452

Image Credits: AI Generated

DOI: 10.3389/fdpys.2026.1800452

Keywords: infancy, distributive fairness, sharing, violation-of-expectation, prosocial behavior, resource allocation, moral development, developmental psychology, equality norm, altruism, social cognition, looking time