Symposium
Chair: Tricia Striano
Discussant: Jacqueline Nadel
A key transition in social cognitive development occurs at around 9months of age when infants start to tune into the attention of the peoplearound them (e.g., Carpenter, Nagell, & Tomasello, 1999; Striano & Rochat,1999). Joint attention is indexed by social behaviors such as imitativelearning, social referencing, gaze following, and gestural communication,all of which involve the coordination of attention between people andobjects. Joint attention, or triadic social behavior, is thought to indexfirst signs of the intentional stance, infants' consideration of people asintentional and goal directed agents (e.g., Tomasello, 1995). The intentional stance is considered as an important milestone indevelopment as it announces unique forms of human social cognition such aslanguage, symbolic play, and theory of mind. However, while recentprogress has been made in documenting dramatic changes in infants' socialcognitive repertoire by the end of the first year, little is knownregarding what might prepare infants for such developments. Much researchsuggests that social understanding begins long before the emergence ofjoint attention behaviors at the end of the first year. For instance,research indicates that between 2 and 6 months of age infants developabilities to monitor and form expectations about people in the context ofdyadic, face-to-face, exchanges (e.g.,Muir & Hains, 1999; Nadel &Tremblay-Leveau, 1999; Rochat & Striano, 1999; Tronick etal., 1978). Onepossibility is that information infants gain about people through earlydyadic relations (i.e., emotional resonance, contingency detection,recognition of emotional expressions) eventually leads to an understandingof the intentions, motives, and goals that drive people's behavior.In the symposium, four different research teams will present their recentfindings on the early development of social expectations and socialcognition. The goal of the symposium is to present an overview ofdeveloping social expectations and cognition, with a focus on determiningwhat information and experiences might prepare infants to recognize andconsider the intentions and motives that guide people's behavior. Twopresentations focus on the early development of social expectations andintentions prior to 9 months of age. One of these presentations will drawfrom evidence suggesting that as early as 3 months of age, young infantspick up and recognize dynamic facial and vocal expressions from an overallflow of dynamic information that is used to infer people's intentions.The other will focus on the role of mothers affective style as adeterminant of infants developing social expectations and discriminationof facial expressions. Two other presentations will focus on infantsunderstanding of intentionality toward the end of the first year. Oneof these presentations will draw from evidence showing that by the end ofthe first year, infants can use various cues to respond to an adultsdirection of regard and to infer intentions. The other presentation willtake an evolutionary and clinical perspective, considering recent researchfindings on understanding of others intentions by typically developinginfants, autistic children, and chimpanzees. The presentations will bediscussed in terms of the relation between developing social expectationsand understanding intentions in others.
Details of individual items:
paper
Young infants come to discriminate and recognize the emotional expressionsof others perhaps as early as 3 months of age. Although most researchershave asserted that infants begin to recognize facial expressions asaffective sometime in the second half of the first year, we find thatpresenting infants with dynamic, multimodal presentations of familiarpersons permits them to demonstrate recognition in the first few months.We report on data from two recent studies. In the first, we asked whether3.5-month-old infants would show preferential looking for a facialexpression accompanied by an affectively matching vocal expression whenthe expressions were provided by the infants' own mothers. Infants werepresented with paired happy and sad facial expressions of their own motheraccompanied by her characteristic happy or sad vocal expression, or withthe happy and sad facial and vocal expressions of an unfamiliar female.Infants looked preferentially to the appropriate facial expression (theone accompanied by an affectively matching vocal expression) only whenmaternal expressions were used. In addition, there were differences ininfants' responsiveness to the filmed facial/vocal expressions. Globalaffective coding yielded data that showed that when happy was thesound-specified emotion, infants were more expressive, they showed greatervariability of affective expression, and increased the number ofalternating expressions. They were also rated as experiencing morepositive affect and as more interested and engaged, particularly when theexpressions were portrayed by their own mothers. Furthermore, morespecific measures of infants' smiles and distress bouts showed that theinfants who viewed their own mothers spent more time smiling at the films.Infants produced more full and bright smiles when happy was thesound-specified emotion and particularly when they viewed the happyexpressions of their own mothers. The groups did not differ in duration ofsmiling when sad was the sound-specified emotion. The average duration ofdistress was significantly longer for infants who observed the unfamiliarwoman.More recently, we have been examining infants' recognition of facial andvocal expressions when these are presented by their mothers, fathers, andunfamiliar males and females. Using the same procedure, we present infants(3 to 4 months of age) with happy, sad, angry facial and vocalexpressions. We also videotape the infants to determine whether theyrespond differentially to these persons and to the expressions that areportrayed. Finally, we obtain measures of parental involvement (diaryprotocols and interviews) to determine whether there is a relationshipbetween the amount and quality of time each parent spends with the infantand later performance on the preferential looking task. These data areimportant for theories of infants' ability to recognize expressions,especially with respect contextual effects on such perception.Intentionality is likely to be detected first in intense and frequentinteractions such as those provided by parents of infants.
paper
By 2-months of age, infants start to monitor and form expectations aboutpeople in dyadic, face-to-face exchanges. Research also indicates that ataround this time, infants are also able to discriminate static displays ofadult smiles from neutral expressions, and that such ability hasexperiential correlates. Specifically, discrimination ability is greateramong infants whose mothers smile less frequently, but who encourageattention toward themselves when smiling in the contexts of normalface-to-face interaction. We report on data from a study examining whether infants abilityto discriminate facial expressions was related to their mothers depressivesymptoms since giving birth. Using a preferential looking paradigm, 3-to 6-month-old infants were presented with varying levels of smilingfacial expressions paired with a neutral expression. Infants heart ratewas recorded as they viewed the facial expressions displayed in 10-secondtrials. Looking time toward the smiling and neutral stimuli, as well asheart rate deceleration while looking at the stimuli, were coded. Inaddition, infants and mothers affective expression (e.g., gazing andsmiling) was coded during a face-to-face interaction. We found that more depressed mothers smiled less to their infantsduring the face-to-face interaction. Results of the visual preferencetask indicate that infants were able to discriminate different levels ofsmiling expressions from a neutral facial expression. In addition,infants whose mothers reported symptoms of depression had a greaterability to discriminate smiling from neutral facial expressions. Comparedto infants whose mothers reported no symptoms of depression, these infantsalso exhibited a greater heart rate deceleration to smiling compared toneutral facial expressions. These findings suggest that maternalaffective style is an important determinant of infants emerging socialexpectations and their ability to discriminate and recognize the meaningof facial expressions in others.
paper
Infants rely on their caretaker's direction of regard to initiateand maintain face-to-face interactions during dyadic interactions. Intriadic engagements (joint attention activities) the adult's direction ofregard targets the appropriate object for shared attention. In bothcases, infants use adult direction of regard to infer adultintentionality. We have already established that by 3 months of age,infants use eye contact to regulate their responses during face-to-faceinteractions. In contrast, their use of off-center, directionalinformation develops later. In joint attention studies, infants followthe adult's shift in eye + head orientation reliably by 12 months of age,and eye-direction alone beginning around 18 months of age in a visualreinforcement paradigm. Limitations of previous joint attention studiesinclude poor dyadic scaffolding and placing targets outside the infant'sfield of view. We used a procedure designed to optimize gaze following byyounger infants--12-month-olds--by: (1) placing our targets (TV monitors)within the infant's visual field (about 30 degrees from midline); (2)having the adult interact with the infant, then repeatedly directed theireyes back and forth towards a TV monitor (head stationary) while saying'Look at that' or 'Wow, what's that' for 10 sec or until the infant lookedat the same monitor; and (3) reinforcing a correct look with an 8 sec'Barney' video excerpt. In Experiment 1, infants received two demonstration trialsfollowed by 6 test trials: 2 left, 2 right and 2 look-down control trials,in randomized order. On trial 1, 86% of infants looked in the correctdirection, p < .05; however, over trials performance declined to justabove chance (56% correct). Because the control trials appeared todisrupt performance, infants in Experiment 2 were given 12 left/righttrials, and an improved reinforcer ('Teletubby' excerpts). On trial 1,80% of the infants gaze-followed, and averaged 69% correct across trials.Finally, we asked if infants could follow the adult's direction of regardwhile viewing the back of her head, a manipulation that requires infantsto infer the adult's direction of regard without any facial cues, usingthe same procedure as above. In Experiment 3, infants received 4demonstration trials where the adult faced the infants, engaged them,turned her head to look at the monitor, and reinforced visual following.Next, she turned around shook a rattle over her head to center theinfant's attention, and proceeded to turn to face the target saying 'Wow,what's that.' Infants were reinforced if they followed her direction ofregard within 10 seconds. They performed correctly on 91% of the firsttwo trials and 81% correct across 6 trials. Clearly, when an effectiveprocedure is used, by 12 months of age infants are able to respond toadult direction of regard using a variety of directional cues includingeye + head orientation, eye direction alone, and even projected gazedirection inferred from viewing the back of the adult's head.
paper
One of the earliest mental states to be understood by typically-developinginfants is intention. Tomasello, Kruger, and Ratner (1993) and othershave proposed that a newfound understanding of other people as intentionalagents is what underlies the blossoming of infants social-cognitive skillsin the few months before infants first birthdays. However, it is unlikelythat infants wake up one morning at around age 9 months and suddenly havea new, complete understanding of other people and their intentions.Instead, children must pass through different levels of understanding ontheir way to an adult understanding. These levels are useful not only inexplaining typical development but also in the study of other populations.We present a series of studies that investigated understanding of othersintentions in three populations: typically-developing infants, youngchildren with autism, and chimpanzees. We discuss different levels ofunderstanding of intentions, including understanding of goals versusintentions and understanding of intentions-in-action versus priorintentions. We present a developmental and evolutionary perspective thatincorporates the different levels of understanding of intentions and wediscuss the implications of this perspective for developmental, clinical,and comparative approaches.