Current Research

The Emory Infant and Child Lab is currently focusing primarily on cross-cultural research concerning social and moral development. We are interested in what we view as a crucial step taken by children in relation to others during the pre-school years. This step is what we call the "ethical stance".

For example, we established that by 5 years of age children develop a sense of equity and fairness. The question is whether this development is universal or, on the contrary, variable across cultural contexts. Is the emerging sense of fairness comparable across highly contrasted cultural circumstances in which children grow? Are all children taking an ethical stance toward others at approximately the same age?

Our goal is to address such questions and contribute to a better understanding of what might be the roots of moral development and moral values, the roots of the "ethical stance" taken by children in the pre-school years.

-A full list of Dr. Rochat's published works can be viewed here

More information on our current research:

Early Moral Development


The study on reputation is an extension of our previous studies on sharing, where we hope to learn more about what children understand about sharing. One of the things we are most interested in is how children think about reputation when they share do they act differently when they know other people are watching? To explore this idea, we ask children to distribute a pile of coins into two banks one for them, and one for the experimenter. We vary the conditions to include a clear bank, an opaque bank, and a noisy bank. We then study how the child acts in each of the different conditions.


Where does property begin and where does it end? This is a perennial conundrum in the history of philosophy. This question, however, continues to elude the scrutiny of cognitive and developmental scientists, particularly cultural psychologists. Aside from philosophical speculations and psychoanalytical interpretations based on case studies, there is no good account on the developmental origins of entitlement, the claim of ownership, how and why children eventually learn to relinquish property via sharing and negotiation. Aside from armchair speculative views, little empirical work exists on the origins of property, nor on the origins of sharing.

More to the point, no systematic research exists on the influence of the great variety of cultural contexts in which children grow. We are interested in documenting the development of the sense of property in preschool age children (between 3 and 5 years of age) growing up in highly contrasted cultural and economic environments. These environments are richer or poorer with markedly different ways of promoting collective values, different ways of relating to others that depend on a more or less individualistic sense of self of the individuals growing up in these cultures and variety of socio-economic backgrounds.

When young children are reported to begin to say mine! typically at around 20 months class children, what they say is de facto that it is not yours. Property is inseparable from the idea of potentially relinquishing it via sharing and negotiation. By definition, property is the necessary prerequisite and constitutive element of what characterizes social exchanges (by analogy, responsibility and autonomy are the necessary prerequisites of punishment).

Property and sharing are thus two inseparable, mutually defining terms. Pioneer field works of early anthropologists all converge in pointing to the importance of object and people transactions, in particular the importance of sharing and gifts in small, primitive societies all over the world (see the seminal writings of M. Mauss, B. Malinowsky, F. Boas, M. Mead, or C. Levi-Strauss).

Gifts, sacrifices, and benevolence are the epitome of active social cohesion and stability maintenance by individuals within a group. However, their forms vary across cultures and it is not clear how children learn the particular forms of sharing within the culture of their parents.

Here we consider that when children refer to when they begin to say mine!is fundamentally dynamic, not normative or static. The argument is that early on, children discover the social power of possession and property, in particular the potential for exchange it represents. They discover the power of social control through the claim of ownership. Accordingly, to talk about possession and ownership it is to talk about the self in relation to others. The idea is that without self-consciousness, nor any consciousness of others, there would be no property question, nor any ownership issue. When young children are reported to begin to say mine!, what they say is de facto that it is not yours. But is it the case in all cultures and for all children growing up in various socio-economic circumstances? The question remains wide open.

Strong Reciprocity

For several years, children have participated in our strong reciprocity study in which children shared coins with a stingy and a generous puppet. We have analyzed our American sample, as well as a small sample of children from the South Pacific. Our results demonstrate that 3-year-olds prefer to keep more coins for themselves and do not distinguish between the stingy and generous characters. However, by 5-years, children prefer to share fairly and will take advantage of the opportunity to punish the selfish puppet. Furthermore, American children are more assertive about punishing the stingy puppet than are children in small-scale, collectivistic cultures. We think these results are exciting and shed some light on the origins of fairness in childhood.

Economic Games

The lab is starting a new series of experiments investigating children's attitudes toward competition, risk, and fairness. We have already collected data from two sites in the South Pacific and are in the process of testing children in the Emory area. We have already noticed some interesting differences between our non-Western samples, and we can't wait to learn more. We have a lot more to do, but so far it appears that children in collectivistic, small-scale traditional societies are fairly egalitarian (i.e., fair-minded) and unlikely to be competitive or prone to risk-taking behavior. We expect to find that American children are more competitive and risky, perhaps as a result of growing up in a more individualistic culture.

Impact of Race and Gender on Social Behavior

The current study is comparing African American children, between ages 3-5 in predominantly African American preschools to those in predominantly Caucasian preschools. Children are asked to report the difference and identify their preference between varying pairs of dolls. They are also asked to distribute eight goldfish between the two dolls that have not eaten all day long. We are using a forced choice game to examine the impact of racial and gender identification/preferences on pro-social behavior. All children participate in 6 conditions:

    (1) black girl doll and white girl doll
    (2) black boy doll and white boy doll
    (3) black girl doll and black boy doll
    (4) black girl doll and white boy doll
    (5) black girl doll and white boy doll
    (6) white girl doll and black boy doll

Preliminary results indicate no difference in preference/identification between African American children in predominantly Caucasian schools and their counterparts in predominantly African American schools. Gender preference for in-group seems to be more constant than preference for racial in-group. Thus far, is seems that African American girls have a stronger affinity towards their racial outgroup doll (Caucasian) than African American boys.

Uncanny Valley

Uncanny Valley Hypothesis As robots increasingly resemble humans, they may fall into an "uncanny valley," in which their realistic appearance and/or behavior elicit uncanny feelings, which researchers describe as eerie, repulsive, threatening, and disgusting. Is uncanny valley a real phenomenon? Although some studies demonstrate a U-shape relation between robots' human likeness and their likability, others fail to detect this relation. In our research, we tested the existence of the uncanny valley by plotting participants' ratings of emotional responses to 89 human and artificial faces against ratings of human likeness of the same faces. Our findings indicate a nonlinear relation between faces' human likeness and the emotional responses they elicit that resembles the uncanny valley, corroborating the uncanny valley hypothesis. In addition, a visual looming task was used to implicitly measure participants' emotional responses to the faces. Results further corroborated the uncanny valley hypothesis. Finally, we conducted a reaction time-based sorting task, in which participants were asked to sort as quickly as possible the same faces used in the previous two tasks as either real or unreal while their reaction times were recorded. The results show that the faces that elicited the longest reaction times were also those that elicited the strongest negative emotions, which suggests that participants' uncertainty about a face's (real or unreal) category may contribute to their emotional responses to this face.

Audience Perception

Past research shows that the perceived presence of an audience (others watching) influences or modifies one's behavior. In other words, because we assume that others will evaluate our behavior, we modify our behavior when others are watching in order to maximize self-presentation. While there is ample evidence that we come to perceive others as evaluators, little is known as to when this emerges in development. Specifically, when do children begin to systematically change their behavior when another is watching as an index of an emerging sensitivity to others' evaluation, and what contributes to its trajectory? My research focuses primarily on exploring these questions in late infancy as well as finding precursors to the developmental trajectory. I plan to continue investigating this phenomenon as well as inter-individual difference in childhood.

Ethnic Familiarity of Novel Faces

Background and Aims: Previous studies have documented a shift in the ability to discriminate faces as a function of early exposure, indicating that there may be emerging face categories in the first months (de Haan, Belsky, Reid, Volein, & Johnson, 2004; Slater et al., 2000). In particular, Pascalis, de Haan, & Nelson (2002) report a change between 6- and 10-months, in infants ability to perceptually discriminate among monkey faces. We further investigated the mechanism underlying this development by recording 6- and 10-month-olds visual exploration of novel adult human faces that were either ethnically familiar or unfamiliar (black or white individuals) using a non-invasive (Tobii) eye tracking device.

Methods: Twenty-seven infants were tested using a Visual Paired Comparison (VPC) procedure modeled after Pascalis and co-authors (2002). We used a within subjects design and presented infants with two blocked conditions of faces from either same- or other- ethnic faces. Infants were familiarized to a face, followed by preferential looking test trials in which the Familiar was paired with a Novel face. In order to determine visual discrimination, we recorded and analyzed infants visual exploration of Novel and Familiar faces either from same- and other- ethnic faces.

Results: We analyzed total gaze duration to Novel and Familiar faces during test trials, as a function of whether it matched or did not match the infants ethnicity. Confirming past studies, 6-month-olds displayed visual discrimination with preferential looking for Familiar faces regardless of matching or non matching ethnicity (black or white). In contrast, 10-month-olds did not show any signs of discrimination in any of the conditions. Consistent with previous developmental studies, 10-month-olds gaze distribution becomes more evenly distributed across regions of the face, compared to gaze patterns of 6 month-olds.

Conclusions: These findings confirm that there is a general developmental shift in face processing between 6- and 10-months that is independent of ethnic familiarity. More studies are needed to understand the potential effect of exposure and familiarity on face processing and social perception in the first year of life. We interpret these findings as indicating a potential shift in what constitutes a familiar face, with familiarity becoming restricted to primary caretakers in the life of the infant by 10-months of age.

Social Dominance

Both humans and non-human animals are able to determine social hierarchies and social dominance between individuals fairly automatically. For example, we are able to know that a big guy is more dominant physically than a little guy. At the same time, we also know that there is strength in numbers, where a little guy may beat the big guy if he has a lot of friends by its side. A remaining question is, when in development do we begin to understand social hierarchies based on size and number? The Social Dominance study explores this question using eye-tracking technology. By testing 6-11 month olds and measuring looking duration between computer-animated social scenarios, we are able to see which social scenario an infant is more "surprised" by as indexed by longer looking times. For example, we would expect infants to look longer at scenarios where a little guy "bows" for a big guy as opposed to the big guy bowing down to the little guy, because this scenario would be more unexpected. The study is currently ongoing.