Poster group
Details of individual items:
poster
Do toddlers take their social partners' line of sight into account whenengaging in joint attention? If pointing involves an understanding of therelationship between seeing and knowing, specific modifications incommunication should appear when joint attention is obstructed. In thisstudy, obstacles were introduced in order to manipulate the visibility oftoddlers' targets to their social partner.Thirty one toddlers aged 18-36 months were tested with an adult socialpartner and two animated clown mannequins 250 cm. away from participants.The clowns were activated from the control boot according to apre-established sequence of movement / pause. Whereas both clowns werevisible to toddlers throughout the experiment, there were three conditionsvarying the number of targets visible to the adult social partner: bothclowns (condition 1), one clown only (condition 2), none of them(condition 3). These three conditions were created by orienting indifferent ways a large screen between target/s and social partner, so thatthe screen would/would not be an obstacle along the adult's line of sighttowards the target/s. Eye contact between toddler and adult was alwayspossible, although toddlers had to turn 90 degrees approximately to lookat the adult. Subsequently, gestures, visual checking (before, during andafter pointing) and vocalisations or words were analysed.The main results show that pointing frequency significantly increases incondition 3, when none of the targets is visible to the social partner.This increase is significant in the older group (31-36 months) also incondition 2, when only one of the two targets is visible to the adult. Atall ages, in condition 2 the majority of pointing is directed to thetarget not visible to the adult. The proportion of pointing associatedwith visual checking with the adult is higher when joint attention isobstructed (conditions 2 and 3). Specifically, in condition 3 thefavourite timing for checking with the adult is just before pointinginitiation, thus showing a deliberate effort to enhance the likelihood toengage the social partner in joint attention. Finally, vocalisations(both associated with pointing and isolated) increase in condition 2 andagain in condition 3. The frequency of isolated words is higher than thatof words associated with pointing in condition 3 only for 31-36 montholds. The frequency of isolated words increases from condition 1 tocondition 2 and then again to condition 3. Although this tendency ispresent in the younger participants, it becomes significant only in theolder group.In conclusion, specific alterations appear in communication when one orboth targets are not visible to the social partner. While showing thetoddler's heightened effort to engage the 'blind' adult in jointattention, such alterations suggest an understanding of the link betweenseeing and knowing. Results are discussed in relation to both languagedevelopment and developments in joint attention highlighting toddlers'understanding of attention and knowledge of others.
poster
Infants use eye contact (or gaze direction) for various communicativefunctions. The manner in which infants use eye contact may indicatedifferent communicative intents depending on the situation. For instance,an infant may alternate their gaze between an active object and an adult inorder to initiate Joint Attention, or to share an experience with another.In contrast, the infant may alternate their gaze between an inactive objectand an adult in order to Request the object, which might be considered amore instrumental function. The present study examined the use of gazedirection in a group of infants at 6 and 15 months of age to determine ifdifferences in function exist in the first two years of life. Participants included 66 typically developing infants who visited the labat 6 and 15 months of age and engaged in various interactions with boththeir caregivers and experimenters. At 6 months, the infants werepresented with a toy that was activated by their caregiver three times,with brief pauses between activations. Infants' gaze direction was codedas to whether the infant gazed at the toy, their caregiver or other duringeach toy presentation. At 15 months, the infants participated in the EarlySocial Communication Scales (Mundy & Hogan, 1996), which allows foropportunities for both Joint Attention and Requesting behaviors.Requesting behaviors are typically coded when objects during the ESCS areinactive, whereas Joint Attention behaviors are typically coded whenobjects are active.At 6 months, the infants spend a larger proportion of time gazing at thetoy (M 81%) than at their caregivers (M 12%) overall [F (1, 65) 679.78, p < .0001]. However, the infants tended to gaze at their caregivermore when the toy was inactive than when it was active [F (1, 65) 22.57,p < .0001]. This was also the case when the frequency of instances ofalternating gaze between the toy and caregiver was examined. The infantstended to have more instances of alternating gaze between the toy and thecaregiver when the toy was inactive (M 1.84) than when the toy was active(M .80) [F (1, 66) 41.56, p < .0001]. These results indicate that theinfants might be more inclined to use eye contact for instrumental opposedto more social purposes at 6 months of age when an object is involved inthe interaction.At 15 months, the infants produced more Joint Attention behaviors involvingeye contact (M 15.74) than Requesting behaviors involving eye contact (M 10.66) during the ESCS [F (1, 35) 17.83, p < .0001]. Gaze direction at6 months did not predict the number of Joint Attention or Requestingbehaviors involving eye contact at 15 months. These results indicate thatthe function of eye contact or gaze becomes more socially oriented in thesecond year of life.
poster
Infant ability to attend to objects is a necessary prerequisite to coordinating actions with others. In particular, infant coordinated attention, or gaze alternation between an adult and toys, is one indication that infants are capable of understanding another's intentions. Both mutually focused visual attention and cooperative interchanges with toys are beneficial for language learning. During mutually focused activities, both mothers and infants are more verbal, mothers label objects to which infants are attending, and mothers offer more support (Hustedt & Raver, 1998). This study examined the specific relations among joint attention, toy interactions in free play, and emerging language. Infants were videotaped in a laboratory at 11 and 14 months (n 103). During the first 10-minute session, infants were seated in a high chair next to mother and given a set of nesting cups, while mothers filled out questionnaires. Joint visual attention was coded from this session in which eye direction could be detected. Infants then engaged in free play on the floor for 10 minutes, while mother sat nearby. From this session, toy interactions were coded. Infant language was assessed using language items from the Bayley Scales of Infant Development (Bayley, 1969). Measures of joint attention included: duration of focused mother and infant attention, mutual or overlapping attention towards toys or the other's face, and coordinated attention (the total duration of mutual looks to the toys, interspersed with infant looks to mother's face). Measures of toy interactions included: initiations with toys (offering or showing toys to the other), responses (initiations which were accepted or acknowledged), and coordinations (initiations which were accepted and returned to the play partner). At 11 months, duration of joint attention was associated with maternal initiations and coordinations and with infant responses and coordinations during toy interactions. Infant attention to mother's face was related to infant initiations and responses during toy interactions. Unfocused infant attention was related to fewer infant initiations with toys. At 14 months, focused maternal attention was related to maternal initiations during toy interactions, but focused infant attention was not related to toy interactions. Coordinated attention was related to maternal initiations and infant responses. Joint attention was highly stable from 11 to 14 months. The strongest predictor of joint attention at 14 months was infant-initiated joint attention at 11 months ( r .62**), a measure reflecting both infant ability to direct and focus attention and mothers' responsiveness to infant visual focus. At 11 months, maternal and infant attention to toys and infant-initiated joint attention were associated with higher infant language scores. Maternal attention to the infant's face and unfocused infant attention were associated with lower language scores. Attention measures were not related concurrently to language at 14 months, but aspects of toy interactions at 11 months were related to language at 14 months. These results indicate that language development is supported by infant attention towards objects and mother's tendency to follow infant attentional focus. Both mother and infant attention skills were related to maternal responses and coordinations during toy interactions, which supported language development over time.
poster
The development of pointing with the index finger (or canonical pointing) by human infants is a well-studied process, with remarkable congruence across studies. In addition to canonical pointing, however, human children often extend their arms and multiple fingers toward distal objects, in a variety of natural and laboratory contexts. These behaviors are variously coded as 'reaches,' 'whole-hand points,' or 'indicative gestures' by developmental researchers. Recent research into the gestural communication of nonhuman primates has revealed that representatives of all great ape species (bonobos, chimpanzees, gorillas, and orangutans), as well as representatives of several monkey species, extend arms and fingers toward distal objects of interest (usually, but not always, food items). Although several individual nonhuman primates exhibit high incidences of index-finger extensions in laboratory contexts, the majority of captive nonhuman primates extend multiple fingers. These behaviors are commonly termed 'reaches.' That these 'reaches' are communicative in function for nonhuman primates has been well-demonstrated: Several laboratories report that these gestures are not emitted in the absence of an observer, hence they are unambiguously communicative in function. Evidence from the human developmental literature, reviewed here, strongly suggests a communicative function for many of the so-called 'reaches' exhibited by human infants, as well. The term 'reach' has a dual meaning: on the one hand, it has a structural meaning that describes an outstretched arm and hand; on the other hand, it has a functional meaning that implies an attempt at prehension or grasping. In neither sense of the term 'reach' is a communicative function necessarily implied. Therefore we recommend distinguishing communicative gestures (whole-hand points) from attempts to grasp objects (reaches). This poster comprises a meta-analysis of pointing behavior by apes, monkeys, and human infants. Reports of pointing by nearly 100 captive apes and monkeys will be tabulated. Because chimpanzees constitute the best-studied species with respect to pointing behavior, most results involve chimpanzee-human comparisons. Among the findings: (a) human infants exhibit high frequencies of pointing with the whole hand, (b) chimpanzees exhibit higher levels of gaze alternation between an observer and an indicated distal object than do human infants from 8 to 18 months of age, and (c) chimpanzees are markedly less vocal than are human infants while pointing. These results will appear in the December, 1999 issue of the Journal of Comparative Psychology. The finding that chimpanzees are less vocal than human infants while pointing to distal objects was previously mentioned during a paper presented at the International Conference on Infant Studies, in Atlanta, GA.
poster
The interactions between infants and caregivers in joint attentionepisodes have been observed to relate to language development (Tomasello &Farrar, 1986). Individual differences in caregiver behavior within jointattention episodes, especially in the tendency to follow and elaborate onthe attentional focus of the infant may contribute to this linkage(Tomasello & Farrar, 1986). Research has also indicated that individualdifferences in early developing infant responding to joint attention andother communication skills may influence the tendency of caregiver andinfants to engage in joint attention interactions (Markus, Mundy, &Morales, in press). It has also been well established that the capacity ofthe child to respond to the pointing and gaze direction of an adult earlyin infancy significantly predicts language acquisition (Mundy & Gomes, inpress; Mundy, Kasari, Sigman, & Ruskin, 1995; Morales, Mundy, & Rojas, inpress). Child ability to determine adult attentional focus may influencethe extent to which child-caregiver dyads engage in episodes of jointattentional focus, thereby optimally benefiting from caregiverresponsiveness, and facilitating child language acquisition (Markus, Mundy,& Morales, in press). While the extent to which caregiver-child dyadsengage in episodes of joint or coordinated attention has been linked tosubsequent child vocabulary size, perhaps infant tendency to initiate jointattention relates to caregiver-child ability to establish and maintainepisodes of joint attentional focus. This study was designed to furtheraddress the possibility that child capacity to direct adult attentionalfocus may also facilitate adult responsiveness to the attention of the child. >Participants were 21 infant-parent dyads. Infant tendency to initiatejoint attention was assessed at 12 months of age during an abbreviatedversion of the Early Social-Communication Scales (ESCS; Mundy & Hogan,1996). The critical measure here, to be referred to as initiating jointattention, was the tendency of the child to establish eye contact,alternate looks between active object spectacles and the tester's eyes,point to objects or pictures, and show objects as bids for joint attention. Child vocabulary development was assessed using the MacArthurCommunicative Development Inventory (MCDI; Fenson et al., 1994) at 12, 18,21, and 24 months of age. Infant-parent free-play interactions were codedfor episodes of joint attention and attentional factors associated withmaternal references to objects at 18 months (see Tomasello & Farrar, 1986). Free play sessions were coded for joint attention episodes, defined asepisodes of 3 seconds or more, initiated by either member of the dyad, inwhich both members simultaneously focused on the same object and the childindicated awareness of their joint focus. In addition, cognitivedevelopment was assessed using the Bayley Scales of Infant Development-IIat 12 and 24 months of age.>We expect to find that dyadic joint attention episodes and the influenceof caregiver behavior on language may relate to child tendency to initiatejoint attention. Data analysis will further examine the impact of childnonverbal communication skills on the regulation of caregiver.
poster
Thatcher (1994) has suggested that measures of EEG coherence reflect the dynamic interrelations among cortical sites that may be associated with cyclical aspects of brain and behavior maturation.Coherence is a measure of phase synchrony, or shared activity between spatially separated EEG generators. Maturation cycles of the left hemisphere may be characterized by a pattern of'integration of differentiation'. In this a reduction in coherence among proximal cortical sites, indicative of differentiation of processes, is followed by an increase in more distant corticalcoherence, indicative of the integration of cortical processes. Alternatively, maturation cycles of right hemisphere functions may be characterized by 'differentiation of integration' or a reduction ofdistal coherence followed by and increase in proximal coherence among cortical areas. Furthermore, different types of left and right hemisphere coherence measures may be markers of brainorganizational processes that support the development of different skills and provide important information about individual differences in behavioral development (Bell & Fox, 1996; Fischer &Rose, 1994; Mundy, Card & Fox, 1999; Thatcher, 1994). Aspects of this model have recently been supported in a study of early communication development (Mundy et al. 1999). In that study a coherence measure, indicative of the differentiationof left frontal and central activity during a baseline attention task at 14 months of age, was predictive of infants' tendency to initiate bids for joint attention at 18 months of age. The present studyextended this line of investigation to examine the relations between 14 month EEG coherence and language development. EEG data were acquired from 29 full term healthy infants at 14 monthsduring a baseline attention task. Data on infant joint attention skill development was gathered at 18 months and language development was appraised at 24 months via parent report on the MacArthurCommunication Development Inventory (MCDI). The results indicated that a 14 month measure of differentiation, or decreased coherence between left frontal and central activity, was not only associated with initiating joint attention (IJA)at 18 months ( r -.48, p < .005), but also correlated with a dichotomous measure of whether children were reported to display higher or lower expressive vocabularies based on a median split of24 month MCDI data ( r -.48. , p <.01), as well as with reports of their sentence complexity ( r -.44, p < .02). The 18 month IJA measure was also related to this dichotomous 24 vocabularymeasure ( r .48, p < .02) and the sentence complexity measure ( r .40, p < .03). Significant effects for the 14 month coherence measure, but not for the 18 month IJA measure, were observedwhen both of these variables were entered into in multiple regression analyses of the median split and sentence complexity variables. These results support the utility of coherence measures in the study ofearly brain behavior relations. They suggested that processes associated with the differentiation of left frontal and central brain maturation appeared to mediate the relation between infant jointattention skill and language development in this study. The implication of these results for understanding early social-communication development will be discussed.
poster
The ability of infants to follow the visual regard of others may reflectthe development of important social, cognitive, and self-regulatory skillsassociated with the ability to acquire language. Individual differences inthe ability of infants to follow the visual regard of others have beenshown to be predictive of language ability in the second year of life(Morales, Mundy, & Rojas, 1998; Mundy & Gomes, 1998; Mundy, Kasari, Sigman,& Ruskin, 1995). While gaze following prior to 12 months of age istypically limited to targets within the visual field of the infant, betweenthe ages of 12 and 18 months of age infants develop the capacity to locatetargets which are outside their visual field, or behind them (Butterworth &Cochran, 1980; Butterworth & Jarrett, 1991).Prediction of later language development from the ability of infants torespond to joint attention (RJA) has typically been computed using infantresponses to targets both within (to the left and right of the infant) andoutside (behind the infant) their visual field. For infants older than 12months of age, however, the developmentally more advanced task of locatingtargets outside the visual field may be a stronger predictor of languagedevelopment than the task of locating a target within the visual fieldwhich emerges at an earlier age. To test this hypothesis, fifty-one, healthy, full-term infants wereassessed. RJA was evaluated at 15 months of age using the Early SocialCommunication Scales (Mundy & Hogan, 1996). During the ESCS, RJA wasassessed using eight pointing trials in which the experimenter attempted todirect the infant's attention to four cartoon characters placed on thewalls to the left, right, and behind the infant. The percent of trials inwhich the infant looked toward the target was calculated for the combinedleft and right trials and for the combined behind trials. Language abilitywas assessed at 24 months of age using the Reynell Developmental LanguageScales (RDLS).Infants responded correctly to the left/right trials (M81.4%) to asignificantly greater extent than they did to the behind trials (M36.1%)[t(38)8.394, p<.001]. Multiple regression analysis indicated that thecombination of left/right and behind trials significantly predictedexpressive language score on the RDLS [R2.303, p.001] but not verbalcomprehension score [R2.042, p.461]. In addition, behind trials were astronger predictor of expressive language score than were left/right trials[B.510, p.002; B.080, p.611, respectively]. Percent correct on behindtrials remained significantly correlated with expressive language evenafter controlling for accuracy on left/right trials [partial-r .479,p.001]. These results suggest that 15 month old infants are quite capable oflocating targets within their visual field but are only just beginning tobe able to locate targets outside their visual field. This variability inthe ability of infants to locate targets outside their visual field mayprovide unique information about their developing expressive languageskills and deserves further study.
poster
Controversy exists over the meaning of the gaze alternation and visualchecking between adult and toy that infants exhibit during jointattention episodes. When an infant points at an object and then looksback at the adult, it would indicate an awareness of the adult as anintentional being or it could be a conditioned response acquired becausethe adult generally provides some reinforcement when the infant lookstoward the object. Visual checking and pointing behaviour were examinedin a procedure designed to elicit joint attention between mothers andinfants. Thirty 16-month-olds and their mothers participated in aface-to-face interaction. One of two remotely controlled toys,positioned equidistant from the infants' midline was activated randomlythrough the interaction. When the toy was activated, mothers wereinstructed to either look at the activated toy, look at the inactivetoy, or look at the child and say 'Yes, look at that'. By looking atthe active toy, their attentional focus matched their infant's and bylooking at either the inactive toy or their child, their attentionalfocus did not match their infant's. When the toy became inactive,mothers resumed interaction with their infants. Infants pointedsignificantly more often when their mothers were looking at them thanwhen they were looking at either the active or the inactive toy, though,pointing across the entire interaction did not differ across the threegroups. Checking behaviour did not correspond with pointing behaviour;infants were more likely to check with their mothers when their motherswere looking at a toy (either active or inactive) than when theirmothers were looking at them. These results do not indicate thatinfants were pointing to direct their mother's gaze when their mother'sattentional focus did not match theirs since there was no increase inpointing when their mothers looked at the static toy as opposed to theactive toy and infants did not show increased checking in conditionswhere they showed increased pointing. However, infants did not seem tobe aware when their mother's attentional focus was directed away fromthem as evidenced by more checking behaviours when their mothers lookedat the toys. It was concluded that 16-month-olds have an appreciationof another's attentional focus but do not understand it well enough todirect it themselves when their mother is not immediately attending tothem.
poster
Mothers' use of affect as a mediator of their infant's attention towardsobjects during mother-infant object play was examined. It was proposed that object play whichis couched withinfamiliar affective interactions provides infants with a familiar routewith which to perceive andinterpret the object. Twenty-three mother-infant dyads were videotapedduring a five minute freeplay session when the infants were either 3 or 6 months old. Each motherwas provided with thesame toy and instructed to interact with her infant as she wouldnormally. Infant attentional stateswere coded for the amount of time the infant spent unengaged, onlooking,engaged with theirmothers or engaged in passive or active joint attention (Bakeman &Adamson, 1984) whilemother and infant affect and gaze direction were coded independently. The behaviour of the 6-month-olds and their mothers was also coded for mother and infantactions on the toy. In bothsets of dyads, correlations between joint attention and individualmeasures of engagement wereexamined. Correlations confirm that affective sharing was established:there were significantpositive correlations between the percentage of time the mother a)lookedat her infant and herinfant looked at her (r .63); smiled at her infant and her infantsmiled at her (r .71); andc)looked at the toy and her infant looked at the toy (r .76). In thelarger sample, active jointattention was positively correlated with mother's smiling when lookingat the toy (r .46) andinfant looking at the toy (r .54) but was not related to infantsmiling when looking at the toy (r .04) and was negatively related to the infant smiling when looking atthe mother (r -.46). Inthe smaller sample, mother's animating the toy was positively correlatedwith her infant'slooking at them (r .74) and negatively correlated with infant'slooking at the toy (r -.76). Mother's animating the toy was also marginally negatively correlatedwith her infant's smilingwhen looking at the toy (r -.61) as well as marginally negativelycorrelated with joint activeattention (r -.55). It appears that the affective sharing duringmother-infant object play wasmore common during social games which used the toy as a prop than duringactive jointattention. These results confirm the use of a familiar interpersonalrelationship as a mediator inthe introduction of objects to infants but place more emphasis on thesocial games betweenmother and infant rather than the object play per se.
poster
The goal of this study was to examine the effect of joint visualattention on infants' behavior during subsequent events. Thirty-four infants (M age: 11.1 months, range: 9 months to 13 months, male: 14, female: 20) and their mothers participated as subjects. These subjects were divided into three groups, which was control group, experimental group 1, and experimental group 2. A personal computer was used to present stimuli on a 17 inch CRT and to record eye movement direction by means of a mouse click by the experimenter. Mother-infant pairs were shown a pair of line drawing stimuli on the CRT two times. In the control group, the mother never paid attention tothe stimuli. In experimental group 1, the mother pointed to one stimulus in the first presentation, but did not point to it in the second presentation. In experimental group 2, one of two stimuli blinked during the first presentation but not the second presentation. The experimenter monitored the infant's eye movements from outside of the room through the one-way mirror and recorded the infant's gaze direction by clicking the computer mouse during the presentation of thestimuli. When the infant looked at the right-side stimulus, the experimenter clicked right mouse. When the infant looked at the left-side stimulus, the experimenter clicked the left mouse. There wasrecorded every 50 m-sec. The total looking time for each stimulus, right and left, was calculated automatically by the computer. The infants gazed longer at the stimulus pointed to by their mothers in the first presentation. In the second presentation, during which mothers never paid attention to the stimuli, the infants gazed longer at the stimulus which had been pointed to by the mothers in the first presentation. Infants gazed at the blinking stimulus longer in the first presentation. However, there was no difference in looking timetoward the two stimuli in the second presentation. The present study explored whether joint visual attention between mother and infant affects the infant's responses when the same stimuli were presented for a second time. More specifically, the focus of the study was on the infant's looking behavior after the occurrence of joint visual attention with the mother. The major findings of the current study are that (1) the infant gazed at the stimulus that was pointed to by his/her mother longer than the stimulus that was not pointed to, and (2) such joint visual attention seems to affect the infant's responses when the same stimulus pair is presented again later. (3) The blink of the stimulus on the computer attracted the subjects' attention, i.e., the babies looked at the stimulus which was blinking longer than the stimulus which was not blinking, however, (4) the blinking stimulus did not affect the infant's looking behavior when the same stimulus pairs were shown subsequently.
poster
Early environments that provide settings for children and adults to participate in joint interactions acknowledge the importance of interactive learning and the opportunities that joint shared participation provides to scaffold the child's learning. Scaffolding or guided support, describes the Vygotskian assistance given by the more competent to the less competent partner.The engagement of young children in the context of everyday activities with significant adults emphasises the importance of both personal and interpersonal processes in early learning environments. Many researchers have analysed the scaffolding that occurs in book reading interactions but not the early learning that occurs within the routinised activities such as the eating of a snack and changing clothing.The study reports on patterns of joint attentional sequences of mother - child dyads at 12 months of age in the context of a number of activities and specifically in the context of book reading, toy play, eating a snack and changing clothing. Video taped sequences of mother - child interactions in these everyday activities were analysed and coded for the dimensions of initiation, termination, focus and length of sequence. Episodes of non joint participation by mother and child were also coded.Research findings demonstrated that at 12 months of age time spent by mother and child in shared interactions far exceeded the time spent in non joint episodes. These findings very clearly demonstrated the responsive role of the mother in maintaining joint attentional sequences. Significant differences in mean length of joint attentional sequences were related to the ways in which the adult and child engaged in the interaction. Patterns of interaction also differed in relation to the focus activity. At 12 months of age the child terminated interactions more frequently demonstrating a balance of power in terms of the child.Analysis of non joint interactions where mother and child attended to a different focus identified some interesting findings and particularly the pattern of the mother initiation following a non joint episode. Gender differences in male and female mother- child dyads were also evident as early as 12 months of age. The three patterns of initiation , initiated by the child, initiated by the adult following into the child's focus and the adult with the adult's own focus have all been shown by the study to be equally productive of rich joint attentional sequences.Everyday activities often just taken for granted are important contexts for early learning and as demonstrated in this study children and mothers do choose to spend substantial periods of time in shared interactions. Brief interactions along with more sustained exchanges are critical to early learning and later understandings by the competent individual.
poster
The ability to follow a change in another personD5s looking behavior in order to look where they are looking, is by definition, the establishment of coordinated looking or joint visual attention. While several developmental accounts of jointvisual attention exist emphasizing infantsD5 representational abilities (e.g., Butterworth & Cochran, 1980; Butterworth & Jarrett, 1991) or infantsD5 ability tolearn the significance of the attention-directing gesture (Corkum & Moore, 1995;1998), the purpose of this experiment is to examine the perceptual basis of joint visual attention. Research by Deak, Flom, Pick, (submitted), investigatingthe perceptual basis of joint visual attention demonstrates that 18- and 12- month-olds establish more instances of joint visual attention with a more expansive or more elaborate gesture compared to a less expansive gesture. Specifically infants in all three age categories looked more frequently toward the correct target when the adult looked and pointed toward the target object than when the adult only looked toward the object. Therefore one possibility regarding the perceptual basis of joint visual attention concerns infantsD5 ability to perceive or pick-up that information specifying where the adult is looking.The current experiment varied the affective expression of the adult attempting to redirect the infantsD5 focus of visual attention. The significance of varying the affective expression is that it requires the infant to attend to different 'types' of information, i.e., information specifying the affective expression ofthe adult, and information specifying where the adult is attending. Therefore ifinfantsD5 participation in joint visual attention is affected by their ability todetect the attention-directing gesture, it is predicted when infantsD5 attention is captured by other information, e.g., affective, they will establish fewer instances of joint visual attention compared to a circumstance where their attention is not focused on other information. Three affective expressions were varied between subjects, positive or happy, negative or sad, and neutral or no affective expression. Forty-eight 7-month-olds served as participants. Each infant was seated facing an experimenter who aligned their head and eyes toward a visual target while posing one of the three affective expressions. A significant effect was found regarding infantsD5 frequency of joint visual attention and affective expression, p <. 05. Neutral affective expressions elicited more 'correct looks', i.e. infant looking to the same target as the experimenter, than positive affective expressions, p 3D .009 or negative affective expressions, p 3D .067. The interpretation is that infants attended to different information in each of three affective conditions. In other words, infants attended to information perceived as the most 'interesting' or the 'easiest to detect'. For instance, when the experimenter displayed affective information (i.e., sad or happy) in parallel with the attention-directing information, infants focused their visual attention on the affective information. Similarly, in the absence of affective information (i.e., neutral) infants focused their visual attention on the information specifying where the experimenter was attending. InfantsD5 perceptual abilities, in particular the ability to pick-up another personD5s attention-directing gesture, is one important factor in the emergence of joint visual attention.
poster
Infants start to follow the gaze of another person to targets outside ofthe visual field at about 9 months of age. This behavior has beeninterpreted as evidence for the earliest understanding in infants thatpeople can look at things. An alternative view is that when they firststart to follow gaze, infants treat adult head turns as predictors of whereinteresting sights will occur and that it is only at the end of infancythat the understanding of gaze as a visual perspective develops. Thelatter interpretation was tested in an experiment in which the gazefollowing of infants of 12 months and toddlers of 24 months was compared intwo conditions. For both conditions, participants were initially shown anattractive moving toy on the appropriate side after an adult head turn toleft or right. In this way, participants observed that the adult head turnpredicted the appearance of an interesting sight to the side. After afirst phase of four such trials, a second test phase of trials waspresented. In one condition (toy removal), the toys were removed and therewas nothing to see to the side. The adult continued to turn to the sidefor 20 more trials (ten to each side). In the other condition (head turnremoval), the adult stopped turning to the side but the toys were activatedfor 20 more trials with ten activations for each toy. Participant headturns during the test phase were recorded. Results showed that infantsproduced significantly more head turns towards the side in the head turnremoval condition than in the toy removal condition. The reverse patternwas seen in the toddler group. These results are interpreted to mean thatfor infants, gaze following results from an expectancy that somethinginteresting will occur to the side after an adult head turn. In contrast,by two years, children interpret adult head turns to mean that the adult islooking at something.
poster
Social referencing is the use of another's perception of a particularsituation when developing one's own understanding of that situation(Feinman, 1982). Although social referencing is known to develop during theend of the human infant's first year, the mechanisms that underlie itsemergence are not yet fully specified. This study takes a dynamic systemsview of social referencing in an attempt to closely examine the behaviorsthat are occurring and changing at the time referencing emerges. Beginningwhen they are seven-months-old, five infants are assessed on a weekly basisfor a period of four months. Referencing behavior is observed every week,while language development and joint attention capabilities are measuredevery two weeks. Other factors, including temperament, maternalsensitivity, and aspects of the home environment are measured at thebeginning, middle, and end of the study. The goal is to determine how theserelated behaviors are changing over the course of change in referencingbehavior in order to pinpoint individual differences in the emergence ofsocial referencing. Preliminary analyses of this ongoing study reveal arelation between the emergence of referencing behavior and joint attention.A significant increase in the number of infant looks between the stimulusand the referee in referencing paradigms is observed at the same time thatthe infant becomes more capable of following the mother's gaze. There alsoappears to be a relation between this increase in looks and the infant'sengagement in various games and routines, as measured by the MacArthurCommunicative Development Inventory. Data has also indicated that infantswill reference in the laboratory setting, which is a rather unfamiliarcontext, before they will reference in their own home, a finding that hasbeen hypothesized in other studies, but has not yet been supported.