Babies as Young as Four Months Judge Closeness by Watching Saliva Sharing

Long before a baby can walk, talk, or even reliably grab a toy, an astonishingly sophisticated social calculation is already underway. According to new research from the University of California, Santa Barbara, conducted in collaboration with Harvard University and MIT, infants as young as four months old appear to use one of the most intimate human behaviors — the sharing of saliva — as a cue for judging how close two people are to one another. The findings, published in the journal Evolution and Human Behavior, suggest that the foundations of our ability to map social networks may be in place far earlier than many scientists had assumed, and that even the youngest members of our species come equipped with abstract expectations about which relationships matter most.

The study, titled Four- to 6-month-old infants use saliva-sharing interactions to infer social intimacy, was led by Brandon Woo, a researcher in UC Santa Barbara’s Department of Psychological & Brain Sciences. Woo, a developmental psychologist by training, has built his career around a deceptively simple question: what does the world look like from a baby’s point of view? How do infants, who cannot yet ask questions or explain themselves, make sense of other people’s actions, minds, and relationships? The new work extends a line of research suggesting that babies are not passive observers of the social scene around them but active interpreters of it, drawing inferences about who belongs to whom from surprisingly subtle behavioral evidence.

The inspiration for the research came from an observation drawn from anthropology and psychology across cultures: people in close relationships tend to be more comfortable sharing saliva with one another than strangers are. Kissing, sharing a fork, eating from the same bowl, licking the same ice cream cone — these behaviors are largely reserved for family members, romantic partners, and intimate friends. Woo frames the intuition with a vivid example: if you saw two people licking the same ice cream cone, you would probably guess that they know each other very well. The question he and his collaborators pursued was whether babies, who have had almost no time to learn such cultural conventions, make the same kind of guess.

The team had already established that older infants possess this ability. In previous work, Woo and his collaborators found that by around eight months of age, babies who watched two people engage in saliva-sharing behaviors — such as kissing, or sharing foods and utensils — inferred that those individuals would respond to each other’s distress. In other words, the infants seemed to understand that saliva sharing marks a thick relationship, one in which the participants are bound to help and comfort one another. What remained unknown was how early in development this capacity emerges, and that question mattered for reasons that go to the heart of a long-running debate about what babies know.

Four- to six-month-old infants present a demanding test case. Compared with the older babies studied previously, they have had far fewer opportunities to observe saliva-sharing interactions in their daily lives. They are also less likely to look at faces during social scenes, which limits the visual information they can take in. And they are typically fed foods — milk, purees — that are not shared with other people in the way a toddler might share a snack. If infants this young still draw the right conclusions from saliva-sharing scenes, the underlying expectation cannot plausibly be the product of extensive personal experience. It would instead point to an early-emerging, and possibly abstract, framework for thinking about human relationships.

To probe this, the researchers designed two experiments. In the first, a human protagonist placed an orange slice in their own mouth, took it out, and then inserted the fruit into the mouth of an actor puppet. This sequence represented a scenario of shared saliva, since the food had contacted both mouths. The contrasting condition involved the same kind of protagonist taking a ball and passing it to the puppet — an act of sharing, but one involving no saliva. After the interaction, the puppet expressed distress, and the researchers measured which person the infant looked at longer: the individual who had shared the orange slice or the individual who had shared the ball. Longer looking in infant studies is commonly interpreted as reflecting expectation, violation of expectation, or heightened interest, making it a standard window into the infant mind.

The second experiment was designed to separate saliva itself from food, a crucial methodological control. Here, a human protagonist put a finger into their own mouth, then into a puppet’s mouth, and then back into their own — a direct exchange of saliva with no food involved. In the contrast condition, the protagonist performed similar actions with a different puppet but touched foreheads instead, a form of intimate contact that does not involve saliva. The protagonist then began to cry, and the researchers observed which puppet the infants expected to come to the rescue. Woo notes that the finger-in-mouth scenario is not a common feature of children’s everyday lives, which made it a strong test of whether babies were truly tracking saliva exchange rather than simply responding to familiar mealtime routines.

The results were consistent across both experiments. When the actor expressed distress, the infants looked longer at the thick relationship partner — the individual who had shared saliva — rather than at the partner who had shared the ball or merely touched foreheads. In the food-sharing experiment, that meant babies expected the orange-slice sharer to respond to the puppet’s cries. In the finger experiment, it meant they expected the puppet whose mouth had received the protagonist’s finger to intervene. Woo describes the pattern as striking: even with less familiar actions, the babies were making inferences about closeness, and the knowledge they applied seemed to generalize across very different kinds of interactions.

That generalization is perhaps the most theoretically significant aspect of the findings. Woo emphasizes that it is fascinating that infants seem to possess abstract knowledge about relationships that applies across different actions. The babies were not simply learning that a specific behavior, such as food sharing, predicts a specific outcome. Instead, they appeared to extract a relational principle — that saliva sharing signals intimacy and mutual obligation — and to deploy that principle flexibly across scenarios ranging from the utterly familiar, like sharing food, to the unusual, like a finger placed in a puppet’s mouth. Abstraction of this kind, at an age when infants are still mastering basic motor control, poses a serious challenge to accounts of social development that place heavy weight on accumulated experience.

The work also speaks directly to a longstanding disagreement among scholars of human development about how much babies understand of the world. As Woo puts it, some scholars argue that babies are blank slates that know nothing, and this research is part of an effort to determine just how babies begin to make sense of the social world, of social relationships, and what role experience plays in that process. The evidence that four-month-olds already treat saliva sharing as a marker of a thick relationship suggests that at least some of the architecture of human social knowledge is present from very early in life. Because human survival and cooperation depend on understanding who will help whom, an early sensitivity to the signals of intimacy would make evolutionary sense, and the study’s publication in Evolution and Human Behavior reflects that framing. For parents, the image is a striking one: the infant in the stroller, seemingly absorbing nothing more than sights and sounds, may already be quietly charting the social network around them, deciding who is close to whom from a licked spoon, a shared bite, or a kiss.

Subject of Research: Infant social cognition and the use of saliva-sharing cues to infer relationship closeness

Article Title: Young infants infer the closeness of social relationships via saliva sharing interactions

Article References: Young infants infer the closeness of social relationships via saliva sharing interactions. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: infant cognition, social relationships, saliva sharing, developmental psychology, Evolution and Human Behavior, UC Santa Barbara, social networks, relationship inference, early development, distress response, human sociality, intimacy cues