Poster group
Details of individual items:
poster
There is a vast field of research regarding the social regulatory capacities of young infants, as assessed in normal mother-infant face-to-face interactions and interactions employing the Still Face (SF) paradigm, with a growing literature regarding the impact of maternal psychopathology on mother-infant interactions and subsequent infant development. We selected women with Borderline Personality Disorder (BPD), which is a severe and chronic psychiatric condition that is marked by affective instability and disturbed interpersonal relationships, to see if we could highlight specific disturbances in mother-infant interactions that would have implications for the infants' emotional development and emerging capacity for relatedness. We were also interested in whether patterns of interaction embedded in the mother-infant relationship may start to generalize to a pattern of relating with a different interactive partner. We recruited eight pregnant women with BPD and 12 pregnant women without any psychiatric diagnosis from the greater London area. Mother-infant (MI) and stranger-infant (SI) face-to-face interactions were videotaped in the research unit when the infants were 2 months old. MI interactions entailed 2 minutes of normal face-to-face play, the SF for 90 seconds, followed by a further 2 minutes of face-to-face play. SI interactions were 2 minutes of face-to-face play. The pre and post SF episodes were rated with the ÔGlobal Ratings for Mother-Infant Interactions at 2 and 4 Months' (Fiori-Cowley & Murray, 1998). All scales are rated from 1 to 5, with 5 indicating a positive quality (i.e., less intrusive, more sensitive). A second approach was to assess infant behaviour across all three episodes of the MI interactions by rating infant affect and gaze. Infant affect was divided into positive, neutral and negative, with negative being subdivided into fussing, distress, and disorganized. Infant gaze was divided into Looks to Mother, Looks Away, Looks to Self, Eyes Closed and Looks Dazed (defined as a Ôfreezing' of facial movement and/or eye contact similar to the type of Ôfreezing' that is found in the Disorganized attachment pattern in the Strange Situation Test). Results from analyses of variance indicated that the BPD mothers were less sensitive to their infants during the pre SF (mean3.30, s.d..44 vs mean4.20, s.d..74; F9.41, p<.01) and post SF episodes (mean3.43, s.d..80 vs mean4.38, s.d..59; F9.55, p<.01) and they were more intrusive in the post SF episode (mean4.00, s.d..60 vs mean4.67, s.d..39; F9.22, p<.01) . Also, the quality of the interactions between the BPD infants and their mothers during the post SF play were of a poorer quality compared to the control group (mean3.00, s.d.1.25 vs mean4.18, s.d.1.20; F4.54, p<.05) . With respect to infant affect and gaze, there were no significant differences between the groups in the pre SF episode. However, during the SF episode, the infants of BPD mothers spent only 5% of the time in a state of positive affect compared to 24% for the infants of control mothers (F7.17, p<.05). They had significantly more looks to mother (mean10.88, s.d.6.85 vs mean3.25, s.d.2.14; F13.25, p<.01) and more looks away than the control infants (mean7.63, s.d.5.95 vs mean1.83, s.d.1.47; F10.67, p<.01) and they had more dazed looks (mean4.25, s.d.5.70, vs mean.42, s.d.1.00; F5.33, p<.05) . The data from the post SF episode suggest that the infants of the BPD mothers were not able to recover as well as the infants of the control mothers. They were in a state of positive affect only 26% of the time compared to 62% for the control infants (F 7.55, p<.01). There was also a near-significant trend for them to spend more time fussing (30%) compared to only 9% for the control infants (F 3.46, p.08). Similar to the SF episode, they had significantly more looks to mother than the control infants (mean10.13, s.d.4.45 vs mean4.33, s.d.2.50; F13.96, p<.01). With regard to the infant-stranger play, the infants of the BPD mothers did exhibit less positive affect than the control infants (mean3.92, s.d.1.24 vs mean4.77, s.d..34; F4.80, p<.05), suggesting some carry-over effect from mother-infant interactions into subsequent interactions, (although this comparison involved only 11 control infants and 6 infants from the index group). Results are discussed in relation to the implications for infant social development and emerging capacity for relatedness.
poster
Parents' sensitive responsiveness, especially under stress, has beenheld crucial to infants' early socio-emotional development. Previousresearch produced contradictory evidence with regard to theeffectiveness of maternal responsiveness in reducing infant distress.Recently, Lewis and Ramsay (1999) investigated the effect of maternalsoothing on infants' reaction to inoculation. Despite stable individualvariations in the amount of soothing, they found no evidence that it waseffective in reducing infants' cortisol or behavioural responses tostress. The present study investigated maternal soothing and its effects oninfant crying in fine detail.First-born, 4-5-month-old infants (12 girls, 18 boys) and their motherswere videotaped at a routine DPT inoculation. Maternal soothing wasanalysed by time sampling using a checklist of 11 behaviours from themoment of inoculation until the infant stopped crying for at least 15seconds. The occurrence of each behaviour was noted for each successive5-second interval with great reliability (kappa ranging 0.61-0.98,M3D0.80). Relative amounts (sum of occurrences/number of 5-secondintervals) were calculated in order to compare dyads. The same infant-mother dyads were observed at home at age 6 months andin the Strange Situation at 12-13 months. Sensitivity and the smoothnessof interactions at home were rated on 5-point scales and averaged acrosscare-taking and play sessions. Quality of attachment to the mother wasassessed independently: the number of dyads across A/B/C/D categorieswere 7/16/4/3, respectively.1. Significant relationships were found between infant quieting, infantsex and the variation of maternal soothing. Face contact andventro-ventral body contact were associated with shorter crying for bothsexes. Soothing speech and stroking were related to quieting for girls,whilst for boys, rocking/swaying was associated with shorter crying.Dressing/diapering was associated with delayed quieting for boys only.With one exception, there were no significant sex differences for therange and mean relative frequencies of various types of soothingbehaviour (mothers smiled more frequently at infant sons).2. Soothing by face contact, ventro-ventral body contact androcking/swaying was associated with the smoothness of interactions athome at 6 months. The relations were stronger for mother-son dyads.3. Maternal soothing at 4 months was weakly related to quality ofmother-infant relationship at 12 months. Mothers who later hadinsecure-resistant relationship with their infant used lessventro-ventral contact after inoculation.In summary, maternal soothing was found influencing infants' behaviouralresponse to the stress caused by inoculation. However, different kindsof soothing behaviour seemed to have different effects on infants ofeither or both sexes. Thus, it was not surprising that the total sum oraverage as measures of maternal soothing were not associated withquieting.Face contact both at the moment of inoculation and afterwards, stroking,rocking the baby, keeping ventro-ventral body contact were significantlyassociated with reduced crying. These types of behaviour are known toreduce crying in non-human primates as well (Hinde, 1974). However,dressing a crying baby, especially a boy, effectively delayed quieting.Indeed, some mothers seemed to be 'tuned out' while getting their infantdressed.
poster
In Lewis and Ramsay's (1999) work, no evidence was found that maternalsoothing is effective in reducing behavioral responses to stress in earlyinfancy. The major question of this study is to verify whether suchintriguing results can be replicated in a context different from theAmerican one. Fifty Italian infants were longitudinally observed duringpediatric vaccinations at 3 and 5 months of age. The coding of bothinfants' and mothers' behavior closely followed the method adopted byM.Lewis and his associates (Lewis & Thomas, 1990; Lewis & Ramsay, 1995ab).The videotapes of the inoculation procedure were coded on a 6 point scalefor infants' peak facial and vocal expressions at successive 5 secondsintervals during the 90 seconds period after perturbation . These data wereused to derive several behavioral response measures: Initial reaction , theaverage score across the first 3 intervals after perturbation; Duration ofmaximum reaction , the longest consecutive number intervals with themaximum affect score.;Quieting rate , the slope of the successive scoresafter the perturbation; Mean Affect Score, in the 90 secs afterperturbation.. Maternal behavior was coded according to a list of 20behaviors which were divided into Proximal (involving physical contact withthe baby) and Distal behaviors. Preliminary statistical analyses showedthat:1. All the indexes for infants are significantly correlated at each agelevel, showing that the coding system is highly reliable also for theItalian context.2. The duration of infants' Maximum reaction is stable between 3 and 5months (r .58, p<.01);3. The intensity of Initial reactions significantly decreases from 3 to 5months (F(1,47) 6.16, p<.001);4. The average scores of both Proximal and Distal maternal behaviors in the90 secs following perturbation do not change according to the infants' age,and they are stable from 3 to 5 months (Proximal, r.40, p<.01; Distal,r.35, p<.05);5. There was a gender effect only for Distal soothing (F(1,47)12.45,p<.001). Italian mothers show a greater amount of distal behaviors (e.g.,words, sounds, smiles, gazes) with boys than with girls;6. The average score of maternal Proximal behavior in the 90 secs followingperturbation is significantly correlated with all the infant measures at 3months (Initial reaction, r.39, p<.01; Duration of maximum reaction,r.41, p<.01; Quieting rate, r.28, p<.05; Mean Affect Score, r.46,p<.01). No correlation between maternal soothing and infant stressreactivity was found at 5 months.These preliminary results seem to indicate that for the Italian contextmaternal Proximal soothing is associated with infant distress only at 3months. Several questions are raised by the above results. Furtheranalyses are currently under way on this corpus of data and the same samplewill be observed at 12 months.
poster
Recently, there has been a great deal of research on the socializationof children's emotions and emotion regulation. An understanding of thesocialization of emotions and its regulation is important because emotionrelated capacities are believed to play a major role in social competence(Cassidy, Parke, et. al., 1992; Eisenberg & Fabes, 1992; 1998; Saarni, 1998;Roberts & Strayer, 1992). Much of the work on the socialization of emotion and its regulationhas focused on the impact of socializers' expression of emotion, parentalreactions to children's emotions, and parents' emotion-related discussionswith their children (Eisenberg, Cumberland & Spinrad, 1998; Eisenberg,Spinrad, & Cumberland, 1998). In terms of emotion-related discussion ofemotion, parents can communicate support, encourage the expression of someemotions over others, and may help children understand the experience andregulation of emotion. Although there is evidence that maternal conversationsabout emotions are linked with children's awareness and understanding ofemotion (Denham, Cook & Zoller, 1992; Denham, Zoller, & Couchoud, 1994; Dunn,Brown, & Beardsall, 1991), there has been little emphasis on the verbalstrategies mothers use to help regulate their children's distress. The purpose of this study was to identify the verbal strategies thatmothers use (i.e., soothing distracting, explaining the situation) duringstressful events to assist in toddlers' emotion regulation. Mothers'responses to their toddlers' emotions were observed longitudinally (whentoddlers were 18 and 30 months of age) and children's emotion regulationabilities were observed at age 4. The first goal of this investigation was todescribe mothers' regulation strategies in early and later toddlerhood and todetermine if the types of strategies mothers use change over this period. Thesecond goal of this work pertained to whether mothers' verbal regulationstrategies predicted children's later emotion regulation. Participants were 42 mother-toddler pairs (23 boys). Children andtheir mothers participated in laboratory assessments when the children were18, 30 and 48 months of age. When the toddlers were 18 and 30 months of age,mothers' verbal attempts to regulate their toddlers' emotions (i.e., labelingor using a regulation strategy, such as distraction or explanation) duringseveral emotion eliciting tasks were transcribed from videotape. Whenchildren were 48 months of age, children's emotion regulation was observed inresponse to a disappointment task. Results indicated that mothers rarely labeled their children'semotions, whereas mothers used a regulation strategy over the time inresponse to children's affect. Moreover, there was a significant decrease inmothers' use of verbal regulation between 18 and 30 months of age (see Table1). In addition, the types of strategies mothers used changed over time. Atthe 18 month visit, mothers were most likely to use soothing in response totheir toddlers' negative emotion. At the 30 month assessment, mothers weremore likely to use an explanation. Preliminary results also demonstratemothers who tended to use regulation strategies at 30 months had children whowere better at masking their negative emotion at age 4. Findings arediscussed in terms of the potentially important role that mothers' verbalregulation plays in the development of emotion regulation skills.
poster
The ability to effectively regulate emotions is a primary goal of socioemotional development (Garber & Dodge, 1991) and is important for many aspects of functioning (Thompson, 1994). Theoretical accounts of the development of emotion regulation strategies suggest that emotion regulation occurs within dyads. For example, it is proposed that infants and young children need external support from caregivers for regulating their emotions (Kopp, 1989). Calkins and Johnson (1998) found that maternal positive guidance was related to the use of distraction and more mother-oriented strategies during frustrating events. In another recent study, Braungart-Reiker, Garwood, Powers & Notaro (1998) found that parents who were more sensitive had 4-month-old infants who used more parent-orientation during the still face paradigm. Other studies examining relationship quality found that mother-infant attachment quality is also related to behavioral regulatory strategies (Braungart & Stifter, 1991; Nachmias et al., 1996). In general, there is far less research on infants' emotion regulation strategies with fathers than with mothers. This is an important gap in the research as fathers may engage their infants in more stimulating, exciting play than mothers (e.g., Lamb, 1977), and thus provide their infants with unique opportunities to practice emotion regulation. This study examined infant emotion regulation strategies with fathers and the relationship between strategy use and infant-father attachment quality.Ninety-one twelve-month-old infants participated in the Strange Situation procedure with their fathers. Immediately following the Strange Situation procedure, fathers and infants completed a competing demands task in which fathers were instructed to complete a questionnaire and infants were left with no toys. Behavioral emotion regulation strategies, infant emotional expressions and father-infant attachment quality were coded from videotapes. Contingency analyses examined the extent to which infant behavioral strategies reduced infants' subsequent distress expressions. Fussing to father, engaging father, social referencing, and self-soothing were all followed by greater decreases in distress than expected by chance, suggesting that those strategies were effective in reducing subsequent distress.A repeated measures MANOVA examining strategy use by attachment quality and child sex revealed a significant interaction between strategy and attachment quality, F(12, 474) 2.97, p. .001. Post hoc tests showed that distraction and leavetaking were used significantly less by resistant infants than by securely or avoidantly attached infants. In order to examine the relations between style of emotion regulation and attachment quality, three clusters of infant emotion regulation styles were created. Individuals in the first cluster, labeled the Self-Distracters, were high on self-distraction, and low on social referencing, engaging father, self-soothing and passive disengagement. The second cluster, termed Father Orienters, was characterized by high social referencing and engaging father, average distraction, and low self-soothing and passive disengagement. Infants in the third cluster, labeled Self-Soothers, were high on self-soothing and passive disengagement, and low on social referencing, engaging father, and self-distraction. A significant chi-square analysis revealed that avoidant infants were most likely to be classified as Distracters, whereas resistant infants were most likely to be classified as Self-Soothers. These data indicate that infants' behavioral emotion regulation strategies with fathers are meaningfully related to father-infant attachment quality. Table 1Observed (and Expected) Frequencies for Changes in Emotional Expression for Each Behavioral Strategy_____________________________________________________________________________Child Strategy Increase Decrease No Change X2 _______________________________________________________________________Fussing to Father 0 (1) 6 (1) 4 (8) 19.45***Engaging Father 25 (25) 45 (30) 139 (154) 10.99**Social Referencing 43 (41) 65 (49) 234 (251) 8.59**Self-Distraction 58 (60) 59 (63) 401 (394) 0.76Leavetaking 6 (3) 1 (3) 18 (19) 4.59Self-Soothing 30 (23) 41 (27) 118 (139) 14.61**Passive Disengagement 8 (8) 3 (8) 56 (51) 4.06_____________________________________________________________________________: ** p. < .01 *** p. < .001 Table 2Chi Square Analysis of Infant Emotion Regulation Clusters by Attachment Quality___________________________________________________________________________Emotion Regulation Clusters Attachment Quality___________________________________________________________________________ Avoidant Secure Resistant _______________________________________________________Self-Distracters 62% 43% 19% (N 8) (N 24) (N 3)Father Orienters 8% 36% 6% (N 1) (N 20) (N 1)Self-Soothers 31% 21% 75% (N 4) (N 12) (N 12)___________________________________________________________________________X2 (4, N 85) 20.30, p < .001
poster
The Bayley Scales of Infant Development (BSID-II) were administered toinfants with prenatal cocaine-exposure and non-exposed controls at 6.5months of age. The examiner's rating of behavior during testing wasassessed with the Behavioral Rating Scale (BRS) of the BSID-II. In thestandardization sample for the BRS, three factors emerged, two of which areconceptually related to behavioral control: Orientation-Engagement andEmotional Regulation. The Orientation-Engagement factor is comprised of 11ratings and the Emotional Regulation factor includes 8 ratings. Bothfactors have items related to behavioral control. TheOrientation-Engagement factor includes ratings of positive affect, interestin test material, initiative with tasks and persistence in attempting tocomplete tasks. The Emotional Regulation factor includes ratings ofnegative affect, adaptation to change in test materials, attention to tasks,frustration with inability to compete tasks, and cooperation. At 4 years of age, a computerized test of vigilance (pictorial CPT,modified to include continuous examiner feedback) was given to some of thechildren. During the test, the examiner recorded the frequency of thefollowing off-task behaviors: head turns away from the computer, non-targetkey presses, getting out of seat, and becoming engaged in non-taskactivities. These observations were compiled into a weighted sum ofoff-task behaviors. For the 64 children receiving both the infant and childhood measures, therewas no difference between cocaine-exposed and control children onOrientation-Engagement, Emotional Regulation or the off-task behaviormeasure. Collapsed across cocaine-status groups, there was a trend forratings indicating poorer Emotional Regulation at 6.5 months to becorrelated (r.20, p<.10) with four-year-old scores indicating more off-taskbehavior. There was no such relation for the Orientation-Engagement factor.Data collection on the four-year old rating is on-going and an additional 40subjects is anticipated.
poster
Normal timing variations in the onset of important developmentaltransitions may be associated with differing patterns of socioemotionalorganization and communication (e.g., Biringen, Emde, Campos, & Appelbaum,1995). Preliminary results from a longitudinal study of toddlers suggest acomplex relation between language and affect expression. Earlier talkerstended to express more positive as well as more anger-related negativeaffect in response to standardized emotion-eliciting events, whereas latertalkers tended to express more fear-related negative affect. This reportbegins to explore the observed difference in fear-related negative affectexpression by examining emotion regulation during one such event, strangerapproach.Emotion regulation develops in the context of early social interactions(Feinman, 1992; Tronick, 1982). During the first three years, childrenincreasingly rely on the emotional expressions of their caregivers toprovide meaning and help them understand the world around them (Barrett &Campos, 1987; Fox, 1994). Children will seek emotional information from acaregiver to help resolve uncertainty and guide subsequent behavior insocial (Boccia & Campos, 1983) as well as nonsocial (Sorce et al., 1985)settings. Individual differences in the child as well as the caregiver arelikely to affect this development (Emde, 1992). Are high fear respondersless likely to rely on mother for guidance in resolving their uncertainty?Are mothers of high fear responders more likely to respond with fear oranger or be emotionally unavailable? Subjects are 80 normally developing, first-born toddlers and theirmothers. Ninety-five percent are Caucasian and from middle classbackgrounds. Toddlers were enrolled at 13 months based on their languagecomprehension and production scores on the Infant Form of the MacArthurCommunicative Development Inventory (CDI) (Fenson et al., 1993). An earliertalker was defined as a toddler whose score fell within the top 25% on theCDI production norms, whereas a later talker was defined as a toddler whosescore fell within the bottom 25% on these norms. Each group has 20 girlsand 20 boys. The MacArthur CDI is a parent-report recognition-based measure oflanguage comprehension and production which parents completed every threemonths when their toddler was between 15 and 30 months of age. At 15, 21,and 30 months, the families were invited to our playroom for the LaboratoryTemperament Assessment Battery (LAB-TAB) (Goldsmith & Rothbart, 1992), astandardized instrument for observational assessment of responsivity to bothpositive and negative emotion-eliciting events. This poster will focus onstranger approach. Coded toddler behaviors include facial expressions,vocalizations and speech, nonverbal bids, and social referencing to mother,as well as coping/distracting behaviors. Maternal responses to eachreference or bid were coded as well and include gaze at/away from child,facial expressions, and speech.Analyses address the relation between the expression of fear-relatednegative affect and the frequency of social referencing and coping behaviorsas well as the nature of maternal responses to toddler references and bids.Final analyses will include all subjects at 15, 21, and 30 months.
poster
Two major theories attempt to explain how children come to regulate theirown emotions. The first emphasizes aspects of the caregiver-childrelationship that predict emotion regulation (Tronick, 1989; Cassidy,1994; Gunnar, 1998). The second emphasizes physiological correlates toemotional patterns and regulation skills (Gunnar et al., 1997; Porges,1997). Little work has investigated the interface betweenbiologically-based and relationship-based processes in emotion regulationdevelopment. In the animal literature, neurophysiological factors havebeen found that influence emotion regulation behaviors, and also developpartially in response to early social relationships (Coplan et al., 1996;Kraemer, 1997). Certain neurophysiological patterns in humans also may beinfluenced by early social relationships (Spangler et al., 1994; Donovan &Leavitt, 1985). One important measure, cardiac vagal tone, particularlysuppression of heart rate variability during challenges, is thought toreflect overall neurophysiological regulation (Porges, 1997). This studyassessed a model with infant cardiac functioning as a mediator in therelation between maternal sensitivity and child emotion regulation. Itwas hypothesized that maternal sensitivity influences child emotionregulation via its influence on infant neurophysiological regulation. A diverse sample of 60 infants and mothers were seen at 3, 6, and 12months of age in the laboratory. Face to face interaction, peek-a-boogames, puppet play, and a routine caregiving task were included to solicitmaternal styles of interaction at 3 and 6 months. Mothers' interactionswith their infants at 3 months were coded using the Emotional AvailabilityScales (Biringen, Robinson & Emde, 1989) as adapted for young infants(Little, 1995). Mothers' behavior with infants at 6 months was codedusing the Parent Behavior Coding Scales for Dyadic Play at 6-months,adapted from Fish, Belsky, and Stifter (1991). Tasks at the 12-monthvisit were designed to elicit children's emotion regulation patterns insituations that should be frustrating, fearful, and positive(respectively) for most infants. Frustrating tasks included prohibitionof a fun toy and a brief arm restraint. For fear, children wereapproached by a research assistant in a clown costume. Positive affectwas solicited in a puppet play. Twelve month child emotion regulation wascoded using measures of child emotionality and regulation. Children'scardiac patterns were collected during all three visits. Data on vagaltone and heart period at baseline and during episodes of two types ofexternal challenge were analyzed. Attentional challenges involvedexposing infants at each visit to only one, age-appropriate, low levelstimulus. Emotional challenge consisted of children's reactions tomaternal Still Face at 3 months, the toy prohibition and arm restraint at12 and 6 months, and clown task described above at 12 months. Analyses will focus on relations among early maternal sensitivity, infantcardiac patterns at baseline and during challenge, and 12 month emotionregulation. Results will be discussed in terms of implications for theoryof emotion regulation processes of development and potential for clinicalintervention.
poster
Recent developmental models of child compliance have highlightedthe role of early emotional regulation on subsequent behavioral compliance(Stifter et. al., 1999). While data has suggested that infants low inregulatory abilities are more likely to become noncompliant, less is knownabout the relationship between emotional regulation and parentalcharacteristics in the development of noncompliance (Stifter et al.,1999). Moreover, the influence of regulatory behaviors such as crying andthe mediating effects of maternal anxiety on early developmental outcomeshas received little, if any, attention (Conway & McDonough, 1999; Weinberg& Tronick, 1998). In this study we examined the role of early emotional regulationand parental characteristics in the prediction of behavioral compliance.One hundred and six infants and mothers participating in the MichiganFamily Study, a longitudinal study examining the relationship betweenregulatory problems and environmental risk factors, were assessed at 7 and15 months (McDonough et al., 1997). Parents completed questionnairesregarding infant crying behavior, maternal anxiety and depression, andbehavioral compliance (Infant-Toddler Social and Emotional Assessment;Briggs-Gowan & Carter, 1998). The results were that the more infants cried at 7 months, thelower the compliance ratings at 15 months (r-.20, p<.04). Moreover, thehigher the level of maternal anxiety at 7 months, the less compliant theirchildren were at 15 months (r-34, p<.001). To examine the degree to which infant crying and maternal anxietyat 7 months predicted noncompliance at 15 months, regression analyses wereperformed. When each of the predictors were tested alone, longest cryduration significantly predicted behavioral noncompliance (F(1, 105)4.26,p<.04; R2.04) and maternal anxiety significantly predicted behavioralcompliance (F(1, 105)13.51, p<.001; R2.11). However, when both of thepredictors were entered into the model, the effect of longest cry durationbecame non-significant while maternal anxiety remained highly significant(F(2, 105)7.86, p<.001); R2.11, b-.31). Even when maternal depressionwas used as a control, the effect of maternal anxiety remained highlysignificant in the full model (F(3, 105)5.51, p<.001); R2.14, b-.26).Moreover, the longest duration of infant crying significantly predictedmaternal anxiety ((F(1, 105)4.51, p<.001; R2.03). These patterns alsoheld when the total amount of 24-hour crying was used instead of longestcry duration. These findings indicate that the relationship between earlyemotional regulation and noncompliance is not direct, but is mediated byfactors such as maternal anxiety, and that maternal anxiety is animportant predictor of early developmental outcomes. Moreover, these datahighlight the importance of assessing transactional and contextualfactors, such as parental characteristics, in studies examining therelationship between early emotional regulation and subsequent behavioraloutcomes. ReferencesConway, A., & McDonough, S. (1999, April). Stress and soothing: Caregiving correlates of infant crying. Poster session presented at the biennial meeting of the Society for Research in Child Development, Albuquerque, NM.McDonough, S. C., Rosenblum, K., Devoe, E., Gahagan, S., & Sameroff, A. (1997, April). Parent concerns about infant regulatory problems: Excessive crying, sleep problems, and feeding difficulties. Paper presented at meetings of the International Conference on Infant Studies, Atlanta, GA. Stifter, C.A., Spinard, T. L., & Braungart-Rieker, J. (1999). Toward a developmental model of child compliance: The role of emotion regulation in infancy. Child Development, 70(1), 21-32. Weinberg, M. K., & Tronick, E. Z. (1998). The impact of maternal psychiatric illness on infant development. Journal of Clinical Psychiatry, 59(2), 53-61. Figure 1-Predictors of Behavioral Compliance Maternal Anxiety .20* -.33***Longest Cry Duration -.13 Compliance (minutes) Maternal Anxiety .20* -.33***Total 24-Hour Crying -.14 Compliance (minutes)
poster
Studies of social referencing have indicated that emotional messages ofothers to situations influence infants' responses to those events. It isstill unclear, however, how various social messages (facial expressions,vocal parameters) influence infants' responses. Especially in vocal tone,results of previous studies have been often inconsistent. Mumme et al.(1996) found that vocal tone was related to infant regulation only infearful conditions. On the other hand, Walden and Knieps (1991) indicatedthat the amount of vocal tone was higher in positive conditions, but itsrelationship with regulation was unclear. Although those few studies available suggested that the quality ofvocal tone in positive vs. fearful conditions might differ, we still do notknow which aspect is related to the difference. This study focused on oneaspect of vocal tone, intensity of hedonic tone, and its relation to infantregulation. In addition, a characteristic of vocal tone revealed by previousstudies was reassessed; whether parents send more unequivocal vocal tone inpositive conditions (Walden & Knieps, 1991). Twenty-six 18-month-old infants (15 boys) and their mothersparticipated in a social referencing experiment. Three novel stimuli (aslinky string, a wooden snake, and a box of sirens) were paired withpositive or fearful messages, which the mother expressed to the infantwhenever the infant looked toward her. Two observers coded both the'negativity' and the 'positivity' of the first three messages, and overallintensity scores were calculated by taking absolute values ofpositivity-negativity scores. High scores in each condition, therefore,indicated highly intense emotional message. Affect regulation (5-pointscale), behavior regulation (7-point scale), and looking pattern data werealso obtained. Data of three parents were excluded due to the failure ofgiving appropriate fearful messages. Results indicated that parents delivered more intense messages in thepositive than in the fearful condition (t1.91, p.069). The correlationbetween intensity in the positive and the fearful conditions was notsignificant (r.34, p>.11). The intensity of vocal tone was significantlyand negatively correlated with both affect and behavior regulation in thepositive condition, but not in the fearful condition (see Table 1). Theresult was congruent with Walden and Knieps (1991)'s study, in which thecharacteristics of positive messages were negatively related to infantregulation. When high and low intensity groups were formed based on medianintensity scores, the high-intensity group looked toward mothers morefrequently, and thereby received emotional message more often (t-2.27,p<.05). Subsequent regression analysis, however, showed that the impact ofthe amount of message received was minimal (see Table 2). Additionalanalyses showed that parents tended to deliver more intense positive messageto girls than boys (t-1.96, p.067), suggesting that the results ofprevious studies showing gender differences in regulation might be partlydue to the different intensity of messages parents deliver (Carpenter,1997). The present study shows that the intensity of vocal tone is relevant ininfant regulation of positive affect. The amount of appropriate vocalmessage, as well as the intensity of vocal tone, were related to infantregulation only in the positive condition.Table 1. Correlations between message intensity and infant regulation andlooking pattern-------------------------------------------------------- Message Condition ----------------------- Positive Negative--------------------------------------------------------Infant Regulation Affect Regulation -.56** -.30 Behavior Regulation -.65*** -.10Infant Looking Pattern Frequency of looking to parents .31 .21 Frequency of looking to strangers -.06 -.20 Frequency of looking to stimuli .25 -.22--------------------------------------------------------** p<.005, *** p< .001Table 2. Intensity and looking patterns as predictors of infant regulationin positive condition--------------------------------------------------------------------------------- Affect Regulation BehaviorRegulation ------------------- --------------------- r beta t r betat---------------------------------------------------------------------------------Intensity of Positivesage -.56** -.56 -3.1** -.65** -.65 -3.9***# of Looking Toward Parents -.18 -.01 - .02 -.30+ -.11 -.63# of Looking Toward Strangers -.24 -.27 -1.5 -.05 -.17 -.10# of Looking Toward Stimuli -.09 -.05 - -.35+ -.20 -1.20---------------------------------------------------------------------------------Note: None of the variables is significant in negative message condition.+p<.10, ** p<.005, ***p<.001
poster
Studies of social referencing show that by 12 months of age infants look to their parentsor experimenters for information when faced with ambiguous or mildly threatening stimuli. Thisresult has been found with diverse types of stimuli, including the visual cliff (Sorce, Emde,Campos, & Klinnert, 1985), mechanical robots (Gunnar & Stone, 1984; Walden & Ogan, 1988),animals (Hornik & Gunnar, 1988), and unfamiliar persons (Feinman & Lewis, 1983). However,considerable variability exists in the frequency of infants' looks to parents, as well as infants'behavioral and emotional reactions in these contexts. Differences in stimulus ambiguity have beenoffered as an explanation, yet ambiguity remains ill-defined in the existing literature. Even lessattention has been directed toward identifying how differences among stimuli influence infants instudies of social referencing. Both predictability and personal control (Gunnar, 1980; Gunnar,Leighton, & Peleaux, 1984) have been suggested as factors that reduce fear, thus, they mayimpact social referencing. Although these studies did not focus on social referencing, they indicatethat stimulus attributes influence perceptions of and reactions to unfamiliar objects. Sixty 24-month-old infants and parents saw two exemplars of three different types ofstimuli (robots, animals, and strange adults) in a counterbalanced order. All stimuli met a minimaldefinition of ambiguous, all were novel and did not elicit extreme reactions in pretesting. Onestimulus of each type was paired with happy parent cues and one stimulus was paired with fearfulparent cues. Infants were exposed to each stimulus for 90 sec. The robots were 12' to 16' tall,made mechanical noises, and were controlled remotely from behind a curtain. For the toy animals,an experimenter placed each object in the room and left for the duration of the trial. Included inthis group was a turtle in a cage, an octopus that shook, and a weasel that rolled around in a box.For the stranger, an unfamiliar adult entered the room with one of three hand puppets. The adultsmanipulated the puppets and periodically invited the infants to play with them using eye contactand verbalizations. Infant looks, affect, and behavior were coded from videotape. Results indicated that infants responded to the different stimuli types in important ways.One, infants' behavioral and emotional responses showed stronger within-group than between-group correlations (see Table 1). That is, infants responded more similarly to stimuli within agroup than across stimulus groups before social cues were given. Two, infants' referencing looksto their parents differed according to stimulus type. Looking at parents took up 42% of the trialtime in the robot condition, 36% and 22% of the trial time for the animal and social conditions,respectively. The difference in referential looking suggests that infants perceived the stimulidifferently. Three, infants regulated their behavior and affect differently according to stimulustype. Infants regulated their behavior and affect significantly in the animal condition, whereasregulation did not reach significance for robots or strangers. Results are discussed in terms of the importance of defining aspects of stimuli that elicitdifferent patterns of infant referential looking and regulation. Because 'uncertainty' and'ambiguity' are multidimensional, a framework is proposed for describing stimuli, which includes personal control, stimulus attributes of predictability and intensity (i.e. sound or speed), and socialcontext (such as mother present vs absent). It is suggested that such a framework will allowdescription and comparison of stimuli in terms of dimensions that underlie ambiguity and threat,which will help us to better understand the conditions under which social referencing occurs. References Feinman, S., & Lewis, M. (1983). Social referencing at 10 months: A second-order effecton infant's responses to strangers, Child Development, 54, 878-887. Gunnar, M. (1980). Control, warning signals, and distress in infancy. DevelopmentalPsychology, 16, 281-289. Gunnar, M., Leighton, K., & Peleaux, R. (1984). Effects of temporal predictability on thereactions of 1-yr-olds to potentially frightening toys. Developmental Psychology, 20(3), 449-458. Gunnar, M. & Stone, (1984). The effect of positive maternal affect on infant responses topleasant, ambiguous, and fear-provoking toys. Child Development, 55, 1231-1236. Hornick, R., & Gunnar, M. (1988). A descriptive analysis of infant social referencing.Child Development, 59, 626-634. Sorce, Emde, R. Campos, J. & Klinnert, M. (1985). Maternal emotional signaling: It'seffect on the visual cliff behavior on 1-year-olds. Developmental Psychology, 21, 195-200. Walden, T. & Ogan, T. (1988). The development of social referencing. ChildDevelopment, 52, 1230-1240.Table 1.Correlations of infant affect and behavior before parent cues within and across stimuli group. Group Affect BehaviorWithin Social .33 .62 Robot .37 .59 Animal .41 .50Across .18 .36