' I. “gr“ - h. ' 'bDI‘II-‘i' KM" , I‘I-I I I . " 3:31“ 'VIIAI')‘ 9' .I ‘l| - - .35..“ I I)?“ ‘ 15...}? -\AA‘ ‘ Fflkulfi‘g' “U .—”I I I“ I ‘ . "‘ 'l HI'.‘. I‘.‘.’ 'I' ”.‘I I.. ”I ‘ II“ , I 'T 53". _V }. It'gl r‘I'? VIII‘" nae]. ‘I‘I 'I‘ 43.- _,.; i. -1“. I .. ‘ ._ .I O "‘I- I I .‘ ‘ I‘- ‘.| ‘v' r‘:‘.II' . 5' ' I'I-' I“_;‘ ' ‘ . I I - . H.I‘u— ' I_.v. ‘z'V'l .JF . . ~. . ‘ 'I'iu' Hf. , ‘ . . u . .l . " a K. :3: ' ‘.I ' H I I'J-"ll :‘K'i‘ . _ III ‘ ' ‘ ‘ I": ' J." (I \\ I.“ I' . .‘J‘ilg‘ -' . “ ' E; II. I... I . , ~ ' _. ; :G' .: .',. ’ ‘ . ’f“ . " i ‘4' -' \‘I V I I .: ‘ . ' ' ' I‘ 4".1S.',’V1'l ‘2'" k .‘f «5%. o "'."~“ It” " ?'§; ‘:E:S:::?2§$‘ ._ ‘ ,' , ‘ .I .I I - . - "1'1 .. “K? ski . (I: ‘gfiu‘fi‘i‘vfid 2. . - ' ~ .. a SWS‘W n—orv-\~ .. Iv...“ ' ”" 'I‘:‘l.‘ .l'I .‘ ;Y“V‘I~I I 'I 1"} as. :i'bvl? <0: , I I‘. I 'I‘ I I I I I I ”r, I ' I .l I I. k..' I. I r V jug-6:3- I> , _ . I I ' __ . . ‘ ‘ ‘ ‘ ‘I ,I".' _ ,l "j. “II-mt I W . ‘ I II II . . .I I , I; '1" .- g' ,: . ‘ , 1W ,m: .1! r'I ,IIIIfImJI... I "9* . ’I '; 14m“IIIIMHHI.,3.» "WWI, a» W l) \\\\\\\\\\\\\\\\\\\ ~ LIBRARY 40 _. 283 37 Michigan State University This is to certify that the thesis entitled Factors Influencing the Regulation of Gazing, Smiling, and Vocalizing Between Parents and Their 3-Month-Old Infants presented by Katherine A. Hildebrandt has been accepted towards fulfillment of the requirements for Ph: nv degree in Jaxehology Major 91%”! }/ 95/pr Date EAT/7’? 0-7639 FACTORS INFLUENCING THE REGULATION OF GAZING, SMILING, AND VOCALIZING BETWEEN PARENTS AND THEIR 3-MONTH-OLD INFANTS BY Katherine Ann Hildebrandt A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1978 C13 (.7 (7 ABSTRACT FACTORS INFLUENCING THE REGULATION OF GAZING SMILING, AND VOCALIZING BETWEEN PARENTS AND THEIR 3 MONTH-OLD INFANTS By Katherine Ann Hildebrandt Visual behavior plays an important role in social interaction at all ages. It is a particularly informative behavior in early parent— infant interactions, since it is one of the few social behaviors which matures relatively quickly and is thus available to both parents and infants. Recent research has revealed a number of factors which may influence the amount and patterning of parent—infant social gaze. The purpose of the present study was to investigate some of these factors in a standardized laboratory play interaction. Two groups of 12 3—month-old infants (half boys, half girls) were studied. Group 1 consisted of first-born infants tested with both parents. Group 2 consisted of pacifier-using infants tested with their mothers. Two 2-minute segments of face-to-face interaction followed by a 1-minute interval during which the parent silently and motionlessly gazed at the infant were recorded for each infant. Infants in Group 1 were observed once with their mothers and once with their fathers. Infants in Group 2 were tested once while sucking on their pacifiers and once without them. Behaviors recorded during the interactions were parent gaze, parent smile, parent vocalize, infant gaze, infant smile, and infant vocalize. Following the interactions, parents completed a series of questionnaires and the infants' responses to several visual and auditory stimuli (Infant Test) were recorded. Multivariate analyses of variance revealed that the mothers and fathers in Group 1 did not differ in their interactive behaviors. Neither were the parent behaviors measured influenced by infant sex or the order in which the parents interacted with their infants. There was a sex difference in the amount of mutual gaze between parents and infants in Group 1. With male infants, more mutual gaze occurred during the first interaction than during the second, whereas with female infants, more mutual gaze occurred during the second inter- action than during the first. The interactive behaviors of the infants in Group 2 were influenced by the pacifier condition and by the parent behavior cessation condition. Infants smiled more and engaged in more frequent mutual gaze when they did not have their pacifiers than when they did. They gazed more and smiled longer when their mothers were interacting with them than when they weren't. Groups 1 and 2 were combined for further analyses. A sec-by-sec analysis of interactive behaviors revealed that mothers and infants both tended to smile more during infant gaze than their base rates of smiling would have predicted. Mothers' ratings of their infants' physical cuteness were higher than, but correlated with, the average cuteness ratings made by four college student observers. Cuter infants were rated by their mothers as being cute in behavior and dissimilar to their fathers in appearance. Cuteness ratings were not related to any of the mothers' interactive behaviors. Infants rated difficult in temperament vocalized more and smiled less during the Infant Test. Infants who smiled frequently during the face-to-face interaction also smiled frequently during the Infant Test. Although infant gaze during the interaction was not related to looking at visual stimuli during the Infant Test, the percent of mutual gaze during the interaction was positively correlated with the total time the infant spent looking at a photograph during the Infant Test. Mothers who had positive attitudes toward caring for their infants looked longer at their infants during the interaction. More total mutual gaze was associated with positive maternal attitudes for mothers of male infants but not for mothers of female infants. More frequent mutual gaze was associated with negative maternal attitudes for both sexes. Results are discussed in terms of cognitive and arousal effects on the regulation of behaviors during parentwinfant face-to-face inter- action. ACKNOWLEDGEMENTS A number of people aided in the completion of this dissertation. I would particularly like to thank my advisor, Hi Fitzgerald, for his support and assistance throughout my graduate career. The rest of the members of my committee, Bill Crano, Lauren Harris, Helen Benedict, and Gary Stollak also deserve thanks. The study could not have been done without the many students and friends who served reliably as research assistants, and the families who participated in the study. I would like to express my appreciation to them all. Finally, several close friends provided both direct and indirect support and assistance during the past year, and I would like them to know that I am grateful. ii CHAPTER 1. CHAPTER 2. CHAPTER 3. CHAPTER u TABLE OF CONTENTS INTRODUCTION. FACTORS WHICH INFLUENCE GAZE INTERACTIONS: LITERATURE REVIEW AND HYPOTHESES. Infant Characteristics. SEX. AGE. BIRTH CONDITION AND TEMPERAMENT. PREDISPOSITION TOWARD GAZE AVERSION. PHYSICAL ATTRACTIVENESS. AROUSAL AND INFORMATION PROCESSING. Parent Characteristics. SEX. PARENT ATTITUDES. Dyad Characteristics. INTERACTION BETWEEN PARENT SEX AND INFANT SEX. CONTEXT. PREVIOUS EXPERIENCES OF THE DYAD. Summary of Purpose and Hypotheses. METHOD. SUBJECTS. PROCEDURE. DEPENDENT MEASURES. DATA REDUCTION. RESULTS. Interactive Behaviors — MANOVAs. GROUP 1. GROUP 2. iii 0\ \OCDO‘ 12 114 21 21 23 2n 2” 26 27 35 35 37 M0 #2 NH M4 M5 A7 All Subjects — Part A. 50 INTERACTIVE BEHAVIORS ~ CONDITIONAL PROBABILITIES. 50 Other Measures. 52 Infant Test Behavior. 55 Correlates of Infant Interactive Behaviors. 57 Correlates of Mothers’ Gaze Behavior. 59 Group 1 Comparison of Mothers and Fathers. 6“ Summary of Results. 71 CHAPTER 5. DISCUSSION. 7” Tests of Hypotheses. 7A Other Findings. 82 EFFECTS OF INFANT FATIGUE. 82 INFANT TEST BEHAVIOR. 83 PHYSICAL APPEARANCE. 8M PARENT DIFFERENCES. 8H Conclusion. 85 APPENDIX A. 86 APPENDIX B. 102 APPENDIX C. 127 REFERENCE NOTE. 136 REFERENCES. 137 iv 0‘ U'l-t'UJN—h q 11. 12. 13. 1”. 15. 16. 17. 18. 19. 20. N ._A N N . I. ll. CD-NIO‘U'I-HWN—A nausunqanauiuau: LIST OF TABLES Background Information (Means and SDs). Background Information (Frequencies). Interactive Behaviors - Simple and Conditional Probabilities. Observer Attractiveness Ratings. Correlations Between Observer Attractiveness Ratings and Mothers Questionaire Responses. Multiple Regression: Temperament Ratings. Multiple Cuteness. Multiple Regression: Multiple Multiple Regression Regression Regression Multiple Regression: Frequency - Female Infants. Multiple Regression: Percent - Male Infants. Multiple Regression: Female Infants. Group 1 Parents Responses to Questionnaires. Attractiveness Ratings - Group 1. Correlations - Group 1. Mother Gaze Frequency and Infant Percent Observer Physical Multiple Cuteness Multiple Cuteness Multiple Cuteness Multiple Cuteness Multiple MANOVA MANOVA MANOVA MANOVA MANOVA MANOVA MANOVA MANOVA Appearance Regression: - Group 1. Regression: - Group 1. Regression: - Group 1. Regression - Group 1. Regression: Looking - Group 1. Multiple REgression: Looking - Group 1. Group Group Group Group Group Group Group Group ‘édd-‘ddd I Infant Test Measures and Mothers Mother Gaze Frequency and Infant Mother Gaze Percent and Infant Cuteness. Mothers Attitudes and Percent Looking. Mothers' Attitudes and Mutual Gaze Frequency - Male Infants. Mothers' Attitudes and Mutual Gaze Mothers' Attitudes and Mutual Gaze Mothers' Attitudes and Mutual Gaze Mother Gaze Percent and Infant Father Gaze Frequency and Infant Father Gaze Percent and Infant Mothers' Fathers' Parents Parents Parents Parents Infants Infants Infants Infants Attitudes and Percent Attitudes and Percent Frequency. Percent. Mean. Interval. Frequency. Percent. Mean. Interval. V 36 36 51 55 55 57 60 60 61 62 62 63 63 65 67 68 68 69 69 69 70 70 102 102 102 103 103 10“ 105 105 B-9 ANOVA - Group 1 - Mutual Gaze - Frequency. 106 B-10 ANOVA - Group 1 - Mutual Gaze - Percent. 106 8-11 ANOVA - Group 1 - Mutual Gaze - Mean. 106 B—12 ANOVA - Group 1 - Mutual Gaze - Interval. 107 3-13 MANOVA - Group 2 - Mothers - Frequency. 107 8-14 MANOVA - Group 2 - Mothers - Percent. 108 8-15 MANOVA - Group 2 - Mothers - Mean. 108 B-16 MANOVA - Group 2 - Mothers - Interval. 108 B-17 MANOVA - Group 2 - Infants - Frequency. 109 8-18 MANOVA — Group 2 - Infants - Percent. 110 B-19 MANOVA - Group 2 - Infants - Mean. 111 B-20 MANOVA — Group 2 — Infants - Interval. 112 8-21 ANOVA - Group 2 - Mutual Gaze - Frequency. 113 B-22 ANOVA - Group 2 - Mutual Gaze — Percent. 113 B-23 ANOVA - Group 2 - Mutual Gaze - Mean. 113 B-2u ANOVA - Group 2 - Mutual Gaze — Interval. 11H B-25 Means for MANOVA — Sex - Frequencies of Infant Behaviors - Group 1. 11A B-26 Means for ANOVA - Sex x Part - Frequencies of Mutual Gaze - Group 1. 11A B-27 Means for MANOVA - Sex x Order x Parent - Percent of Infant Behaviors - Group 1. 115 B-28 Means for ANOVA - Sex x Order x Parent - Mutual Gaze Percent - Group 1. 115 B—29 Means for MANOVA - Sex x Condition x Parent — Percent of Infant Behaviors - Group 1. 116 B—30 Means for MANOVA - Sex x Order x Condition x Parent - Percent of Infant Behaviors ~ Group 1. 117 B-31 Means for MANOVA - Sex - Frequency of Parent Behaviors Group 2. 118 B-32 Means for MANOVA - Sex x Order - Frequency of Parent Behaviors - Group 2. 118 Bu33 Means for MANOVA - Sex x Order - Frequency of Infant Behaviors - Group 2. 118 8-3” Means for MANOVA — Order x Part — Frequency of Infant Behaviors - Group 2. 119 B—35 Means for MANOVA - Part - Percent of Infant Behaviors Group 2. 119 B-36 Means for MANOVA - Order x Part - Percent of Infant Behavior - Group 2. 119 B-37 Means for MANOVA — Condition - Percent of Infant Behaviors - Group 2. 120 8-38 Means for MANOVA - Sex - Mean Length of Infant Behaviors - Group 2 120 B-39 Means for MANOVA Sex x Order - Mean Length of Infant Behavior - Group 2. 120 B-HO Means for MANOVA - Condition - Mean Length of Infant Behavior - Group 2. 121 B-M1 Means for MANOVA - Sex x Order - Intervals between Infant Behavior - Group 2. 121 B—A2 Means for MANOVA - Part - Intervals between Infant Behaviors - Group 2. 121 3-43 Means for ANOVA - Part - Mutual Gaze Frequency - Group 2. 121 vi E-uu B-MS B-A6 B-u7 8-48 B-u9 3-50 8—51 Simple Effects Tests Percent - Group 1. Simple Effects Tests Group 1. Simple Effects Tests Percent - Group 1. Simple Effects Tests Group 2. Simple Effects Tests Frequency - Group 2. Simple Effects Tests Percent - Group 2. Simple Effects Tests Group 2. Simple Effects Tests Group 2. Part A — Sex x Group Frequency. Part A - Sex x Group Percent. Part A — Sex x Group Mean. Part A - Sex x Group Interval. X X Sex x Sex x Sex x Sex x Order Order Sex x Sex x Order Order Order Order Order x Parents - Infant Gaze Parent - Mutual Gaze Frequency - Order x Parent - Mutual Gaze Order - Infant Smile Frequency - x Part - Infant Vocalize x Part — Infant Vocalize Order - Infant Smile Mean - Order - Infant Smile Interval - Mothers and Infants - Mothers and Infants - Mothers and Infants — Mothers and Infants - Mothers' Responses to Questionnaire Items. Intercorrelations between Parent Questionnaire Items - All Mothers. Infant Test Means. Infant Test Intercorrelations. Correlations between Infant and Mother Interactive Behaviors - All Infants - Part A. Correlates of Interactive Behaviors - All Infants - Part A Behavior. Correlations Between Infant Test Behavior and Interactive Behaviors. Intercorrelations Between Parent Questionnaire Items - Group 1 Mothers. Intercorrelations Between Parent Questionnaire Items - All Fathers. vii 122 123 12A 125 125 125 126 126 127 127 127 128 128 129 130 131 131 132 133 13“ 135 1. LIST OF FIGURES Sex x Order x Parent Interaction for Mutual Gaze Percent. viii H8 CHAPTER 1 INTRODUCTION The important role that visual behavior plays in social inter— action (Argyle & Cook, 1976) becomes particularly evident when one member of an interacting dyad either lacks vision or does not use visual signals appropriately (Adamson, Als, Tronick, & Brazelton, 1977; Fraiberg, 197A; Hutt & Ounsted, 1966; Thoman, 1975). Although many of the social behaviors of the adult are not present (or are present only in immature form) in the infant, visual behaviors are unique in that they tend to mature relatively early and thus are shared by parents and infants from an early point in their relationship (Robson, 1967). Within a few days after birth, most infants can fixate and follow a slowly moving object with their eyes, and, unlike many other newborn reflexive behaviors, visual fixation and pursuit do not drop out or become replaced by other behaviors (Greenman, 1963). Instead, they improve steadily with age, and by three months, most visual abilities reach adult levels (e.g., Haynes, White, & Held, 1965). Thus, the infant can use vision to contact, explore, and follow objects (including peOple) at a distance long before being capable of locomotion (Walters & Parke, 1965). Vision is also one of the first behaviors to come under voluntary control (Stern, 197Ab). Not only can the infant choose where he looks, but he also can choose whether or not he looks, thereby using vision as an intentional on-off system (Robson, 1967). Although the degree of intentionality of infant visual behavior may vary widely 1 2 across situations, the fact that vision can be mediated by cognitive factors makes it a particularly informative infant behavior for both researchers (Cohen & Salapatek, 1975) and parents (Fraiberg, 1974; Robson, 1967; Wolff, 1963). Researchers have used selective infant visual behavior to study such processes as discrimination, preference, schema formation, and memory. Parents often infer meaning from their infants' visual behavior, such as attentiveness, arousal level, orienta- tion to objects, discrimination, recognition, preference, and evaluation (Fraiberg, 197A; Robson, 1967). Particularly in early infancy, parents may attribute meaning and intention beyond the infant's capacity to behaviors which occur during social interaction (Anderson, 1977). Because of these inferences and because of the conversational patterning of many adult and infant behaviors (e.g., Jaffe, Stern, & Peery, 1973), social interactions take on the quality of being communicative (Bateson, 1975) and meaningful for the adult. Although early social interaction generally has been considered an important precursor of parent-infant social attachment, until recently little stress was placed on the visual component of these interactions. Bowlby's theory of attachment (1969) considers crying, smiling, following, clinging, and sucking to be the infant behaviors which express attachment to an adult. Robson (1967) added eye-to-eye contact to the list of attachment behaviors because of its role in establishing and maintaining contact between a mother and infant even in the early months of life. Robson also pointed out that eye—to—eye contact con- tributes to the mother's attachment to the infant as well as the infant's attachment to the mother, as do Bowlby's five biobehavioral components of attachment. Being looked at by her baby is rewarding for 3 most mothers, and often serves to facilitate other interactive behaviors (Wolff, 1963). Since Robson's (1967) paper, a number of studies have appeared which describe early social visual interaction. Moreover, looking behaviors have been included as variables in most studies of general caregiver-infant interaction. Many of these studies are reviewed below. On the basis of both direct and indirect evidence, a number of factors are believed to influence different aspects of social gaze between parents and young infants. The purpose of this dissertation was to summarize what is currently known about these factors, and to conduct a study to improve our understanding of the factors that regulate social gaze in face-to-face interactions between parents (both mothers and fathers) and young infants. For the purposes of the following literature review, factors which influence parent-infant gaze interactions have been divided into infant characteristics, parent characteristics, and dyad characteristics. None of these factors can be assumed to have simple or unidirectional effects on behavior. Even if a factor has its most direct influence on one member of the interacting dyad, it will almost inevitably influence the other member as well. This secondary influence can be mediated cogni- tively or can be the direct influence of the affected behavior on the other member of the dyad. Infant characteristics include sex, age, parity, birth condition, temperament, predisposition to gaze aversion, physical attractiveness, perceptual/cognitive capacities, and periodic changes in arousal level. Parent characteristics include sex, attitudes toward the infant and toward parenthood, and responsiveness to infant behaviors. Dyad charac- u teristics include the interaction between parent sex and infant sex, quantity and quality of previous contact between parent and infant, the context of the interaction, and the communicative nature of early adult-infant interaction. According to Stern (1974a), the goal of mother-infant play is to assure the "mutual maintenance of a level of attention and arousal within some optimal range." He argues that infant gaze alternation serves to control the infant's level of attention and arousal. If Stern is correct, gaze aversions should be more frequent when the infant becomes excited or overloaded with information, and less frequent when the infant is at a very low level of arousal or is provided with very little visual information. This hypothesized regulator of infant gaze patterns was indirectly tested in the present study. Rather than try to directly measure the infant's level of arousal, arousal level was manipulated by placing a pacifier in the infant's mouth. The literature reviewed below indicates that the availability of a pacifier (here defined as the presence of a pacifier in the infant's mouth, whether or not nonnutritive sucking is actually occurring) serves to maintain the infant at a moderate level of arousal. Thus, less gaze aversion was expected during pacifier availability than during interactions when the infant was not supplied with a pacifier. In addition to gaze, smiling and vocalizing by both infant and parent were recorded on a seconduby-second basis. These behaviors may elicit or maintain gaze. They also may have arousing properties. Thus, the temporal relationships among these behaviors are important to consider in any attempt to explain gaze patterns. Nevertheless, since the emphasis of the study was on gaze, other behaviors were not recorded 5 in so much detail as has been done in some previous studies of face—to-face interaction. In summary, the present study tested the effects of infant sex, parent sex, and pacifier use on parent-infant gaze interaction. Further analyses were conducted to describe the second-by—second regulation of parent-infant gaze. The literature review below summarizes what is currently known about factors influencing gaze interactions. Hypotheses related to this study are presented in the context of the appropriate literature and are summarized at the end of the review. The literature review is organized in three major sections: infant characteristics, parent characteristics, and dyad characteristics. CHAPTER 2 FACTORS WHICH INFLUENCE GAZE INTERACTIONS: LITERATURE REVIEW AND HYPOTHESES Infant Characteristics SEX Studies of infant visual behavior in response to different types of photographs have variously reported no sex differences or sex differ- ences favoring either girls or boys (cf. Maccoby & Jacklin, 197A). Maturational differences between the sexes are often cited as explana- tions for sex differences when they do occur, although it is unclear at this point which visual behaviors are more "mature." Therefore, general statements about sex differences in infant visual behavior are unwarranted given the present state of the literature. Studies of parent-infant face«to-face interaction by Brazelton and colleagues (Als, 1977; Brazelton, Koslowski, & Main, 1974; Brazelton, Tronick, Adamson, Als, & Wise, 1975; Tronick, Als, & Brazelton, 1977; Yogman, 1977) and Stern and colleagues (Jaffe et a1, 1973; Peery & Stern, 1975, 1976; Stern, 197Ab) do not consider or report information on the effects of infant sex. Some of these studies used very small samples, making sex comparisons unfeasible. Three other studies which used similar procedures and larger sample sizes included infant sex as a factor in their analyses, but found no significant main effects on infant, parent, or dyadic behaviors (Blehar, Lieberman, & Ainsworth, 7 1977; Field, 1977; Friedman, Thompson, Crawley, Criticos, Drake, Iacobbo, Rogers, & Richardson, 1976). In contrast to studies of gaze behavior in face-to-face inter- actions, effects related to infant sex have been reported in a study of gaze behavior in the home which included, but was not restricted to, face-to-face interactions (Moss & Robson, 1968). These differences were in the correlates of mutual gaze between mothers and their 3-month-old infants rather than in the gaze behaviors themselves. No sex differ- ences were reported by Moss (1967) in the amount of time infants spent looking at their mothers or in the amount of time their mothers spent looking at them, and Moss and Robson (1968) found no sex differences in the amount of mutual visual regard between mothers and infants. However, the amount of mutual visual regard was significantly correlated with ratings of the mothers' attitudes toward infants assessed during a pregnancy interview for mothers of girls but not for mothers of boys. More mutual visual regard between mothers and daughters (maternal looking and infant looking were not recorded separately in this study) was related to higher ratings for the mother on the "Degree to Which the Baby is Seen in a Positive Sense" and "Interest in Affectionate Contact with Infants." More mutual visual regard between mothers and sons was related only to the mother's concurrent concern over the baby's well- being. Moss and Robson (1968) suggest that factors such as the mother's uncertainty about how to interact with a boy and the greater irrita— bility of male infants (Moss, 1967) may attenuate the relationship between attitudes during pregnancy and later behavior for mothers of boys. In sum, the literature suggests that there are no direct effects of 8 the infant's sex on gaze interactions between mothers and infants. However, the infant's sex, or behaviors associated with the infant's sex, may interact with parental attitudes to influence the determinants of gaze behaviors. Thus, the factors influencing individual differences in gaze behavior may differ for the sexes, even though there are no group sex differences in such measures as absolute amount of looking. In light of these findings and speculations, no overall sex differ- ences in infant gaze behaviors were expected. However, it was hypoth- esized that there would be more mutual gaze between mothers and daughters when the mothers had positive attitudes toward their infants, whereas no such relationship was expected to hold for mothers and sons. The possibility remained that infant sex would influence father-infant gaze differently from mother-infant gaze. The interaction between parent and infant sex is discussed below. AGE Newborn infants spend very little time in eye-to-eye contact with their mothers (Als, 1977). Although the newborn may appear to be engaging in eye-to-eye contact at times, when the mother moves her head the infant often continues staring off in the original direction (Brazelton et a1., 1974). Wolff (1963) observed a change in the quality of face gazing around 3 1/2 weeks which he interpreted as the beginning of true eye-to-eye contact. Several of the mothers in his study also noticed this transition in their infant's behavior and began taking more interest in playing with them. Infant looking at mothers has been found to continue to increase from 3 weeks to 3 months whereas mother looking at infants remains the same (Moss, 1967). Thus, mutual visual regard 9 also increases across this age span (Moss & Robson, 1968). At older ages (”-8 months), the amount of mutual visual regard with the mother declines (Friedman et a1., 1976). Stern (197Aa) suggests that older infants, when given a choice, shift interest from their mother to objects that can be manipulated. Interest in visual interaction seems to be highest around 3 to A months, when infants' visual skills are relatively mature but locomotion and detailed manipulation of objects are not yet possible. Observations of visual interactions between parents and infants at this age are likely to be informative, but their deve10pmental significance may be limited if, as Stern (1974a) suggests, the function of gaze at the mother changes with the age of the infant. Eventually, longitudinal studies extending beyond 3 or 4 months of age will be needed. Since all the infants in the study to be reported here were as close as possible to the same age (3 months), no age differences in infant interactive behaviors were expected. BIRTH CONDITION AND TEMPERAMENT Most studies of parent-infant gaze have not considered the birth condition of the infant other than attempting to select infants who had experienced relatively "normal" births. A few studies, however, indicate that gaze interactions differ according to the infant's birth condition. For example, Osofsky (1976) found a positive correlation between 5 minute Apgar scores and the amount of time newborns looked at their mothers during feeding. A similar finding occurs with older infants, with asphyxiated infants at both 3 and 8 months looking less than control infants (Campbell, 1977). Infants labeled at risk due to 10 prematurity or postmaturity also have been found to gaze at their mothers less than normal gestation infants at 3 months of age (Field, 1977). Interpretation of these findings requires consideration of maternal behavior and perceptions as well as temperamental differences among infants of varying birth conditions. For example, Osofsky (1976) found that the more alert newborns in her sample received more visual stimula- tion from their mothers. It may be that these infants looked more at their mothers because their mothers elicited more looking from them. In the Field (1977) and Campbell (1977) studies, the mother's behavior also differed according to the infant's birth condition. Campbell (1977) found that mothers of asphyxiated infants played with their infants less than mothers of control infants. In contrast, Field (1977) found that mothers of premature and postmature infants were more active in face-to-face interactions with their babies. These differences in findings may be due to the different contexts in which interactions were observed in these two studies. Perhaps risk infants are played with less in the natural home environment because of their lower responsive- ness (e.g., Campbell, 1977). However, in a structured laboratory situa- tion where mothers are told to interact with their infants (e.g., Field, 1977), mothers of risk infants may feel they need to try harder to get their infants' attention because of their infants' general lower level of responsiveness. However, when mothers increase their activity levels in an attempt to elicit and maintain infant attention, infant gaze decreases (Field, 1977). Thus, higher maternal activity level may be caused by perception of their risk infant as unresponsive. In turn, higher activity levels may then produce less infant looking, thereby 11 confirming the mother's initial perception that her infant is unrespon- sive. The generally lower social responsiveness of newborns at risk suggests that infant characteristics are initially responsible for this cycle of low gaze interaction. The mother's perceptions and behaviors may then maintain the cycle. The infant's temperament is another factor often confounded with birth condition. In Field's study, more of the risk than control infants were rated as "difficult" by their mothers using the Carey Infant Temperament Survey (1972). Thus the "difficult" temperaments of the risk infants could have been the critical influence on their gaze interactions, rather than, or in combination with, their risk status. Campbell's study (1977) included a comparison of non-risk difficult infants with control infants and found a tendency for the mothers of difficult infants to look less at their infants than mothers of controls. She also found no difference in temperament ratings of the asphyxiated and control infants. These results indicate that both birth condition and temperament (at least Carey's "difficulty" dimension) can affect gaze interactions. The separate influences of these two factors cannot be discerned in the Field (1977) study. The present study was limited to infants of "normal" birth condition. However, infant temperament (and parents' perceptions of infant temperament) was expected to vary within this population and potentially to affect parent-infant gaze interaction. Thus, some questions related to infant temperament (taken from Carey's Survey) were asked of the parents. Based on the literature reviewed, it was predicted that there would be less mutual gaze when the parent perceived the infant as temperamentally difficult. Infants perceived as difficult 12 should also show more negative responses during gaze interactions, such as fussing and gaze aversion. PREDISPOSITION TOWARD GAZE AVERSION Some infants seem to show a predisposition for gaze aversion (Robson, 1967; Thoman, 1975), and extreme gaze aversion is frequently considered to be a sign of early infantile autism (Hutt & Ounsted, 1966). However, adequate social relationships between parents and infants can be established with very little or no social gaze (Adamson et a1., 1977; Fraiberg, 197A; Thoman, 1975), so gaze aversion cannot necessarily be considered a cause or indicator of psychopathology. Parental response to low levels of social gaze in their infant is an important determinant of the extent of influence of such infant behavior on social attachment. For example, although it is often difficult and awkward for adults to interact with blind infants, teaching parents to use other sensory modalities can result in a relatively normal social relationship (Fraiberg, 197“). On the other hand, Stern (1971) describes a disruptive social interaction between a mother and one of her three-month-old fraternal twin sons, who averted gaze frequently, and a smoothly running interaction between the same mother and her other twin, who tended to gaze for longer periods. Although Stern (1971) suggests a number of influences on these two interactions, the mother's different behavior may have been partly the result of differences in the twins' predispositions toward gaze aversion. A measure of infants' general degree of interest in social gaze was obtained in the present study through parental report. Although many 13 factors influence infant gaze, it was hypothesized that those infants whose parents have noticed their lack of interest enough to report it would show low levels of looking during the structured observations. PHYSICAL ATTRACTIVENESS Much is known about the effects of physical attractiveness on adult-adult, adult-child, and child-child social interactions (Berscheid & Walster, 1974). In general, physically attractive individuals receive more positive evaluations and are treated more positively than less physically attractive individuals. Although there has been little research on the topic, there is increasing evidence that physical attractiveness can affect adults' attitudes and behaviors toward infants. For example, Corter, Trehub, Boukydis, Ford, Celhoffer, & Minde (1976) asked nurses to rank order photographs of premature infants on the basis of physical attractiveness and intellectual prognosis and found the orderings to be highly correlated. In addition, Hildebrandt and Fitzgerald (1978) found that college students spent more time looking at photographs of infants whom they ranked higher in cuteness. The same effect held for mothers looking at infant photographs (Hildebrandt & Fitzgerald, Note 1). As might be expected, however, the mothers' perceptions of the cuteness of their own infants' photographs were higher than the cuteness ratings given the photographs by college students. Even though most of the mothers in this study gave their infant's photograph the highest possible cuteness score, others did not. Although all the mothers looked longest at their own infant when photographs were presented in pairs, the difference in looking at own and other tended to be larger 1A for mothers who gave their infants the highest cuteness score. The present study tested the hypothesis that parents who perceive their infants as cuter will look longer at their infants during social interaction than parents who perceive their infants as less cute. Several different questions about infant cuteness were asked in an attempt to elicit a range of cuteness ratings. Parents were asked to rate their infants' physical and behavioral cuteness, to predict how college students would rate their infants' cuteness, and to compare their infants' cuteness with that of a preselected group of infant photographs. AROUSAL AND INFORMATION PROCESSING The infant characteristics discussed so far are expected to be related primarily to the amount of social gaze occuring during face-to-face interactions. Other factors must be taken into account to explain the patterning of social gaze. A common observation in mother-infant interaction studies is that mothers look at their infant most of the time during a face-to—face interaction, whereas infants look substantially less. Thus, mutual gaze patterns are determined in large part by the infant's gaze alternation. For example, Stern (1974b) found that infants initiate and terminate 94% of the mutual gazes during play. This does not mean, however, that the mother's looking behavior is irrelevant to the infant's looking behavior. Stern (197hb) also noted that the infant was more likely to initiate a gaze and less likely to terminate a gaze when the mother was looking than when she was not. In addition, the actual initiation and termination of face-to-face interaction in the home is generally deter- 15 mined by the mother (Blehar et a1., 1977), although most infant initia- tions are responded to. Parent-infant face-to-face interactions are cyclical in character, with infant cycles of attention and nonattention occurring at an average rate of “.4 per minute (Brazelton et a1., 197”). In a smoothly flowing interaction, the infant's excitement builds to a peak which is followed by a period of gaze aversion, and the mother's behaviors are synchro- nized with the infant's in that she increases her behaviors as the baby's excitement increases and then decreases her activity during the infant's nonattention. Infants look longer, look away for shorter periods and become less upset when their mother's behavioral cycles are synchronized with their own (Brazelton et a1., 1974; Field, 1977). Both Stern (197Ha) and Brazelton et al. (197“) have proposed that infant gaze alternations serve a homeostatic function in regulating arousal and information intake during social interaction. It is assumed that face-to-face interaction with an adult is both arousing and infor— mative. This assumption is supported by studies of adult gaze (reviewed by Argyle & Cook, 1976): direct social gaze produces increased arousal, there is less looking at highly arousing stimuli, and there is less looking when information load is high. The assumption that face-to-face interaction with an adult provides the infant with a relatively high amount of information can easily be accepted. An interacting adult is likely to be one of the most complex, novel and continuously changing stimuli in an infant's perceptual world. The assumption that face-to-face interaction is arousing seems intu- itively correct from the descriptions of infant behavior during these interactions (e.g., Brazelton et a1., 1975). However, the reasons for 16 the arousing properties of face-to—face interaction have not been clearly established. The hypothesis that infants alternate their gaze to modulate arousal level seems to be based largely on observations of changes in motor activity during social interaction. None of the studies of parent-infant face-to-face interactions have measured physiological responses, although both Brazelton et al. (1975) and Stern (l97ub) have suggested this approach. The present study attempted to test the gaze/arousal hypothesis in a more indirect fashion, by manipulating arousal level and observing the effects on gaze behavior. The manipula- tion used was to provide the infant with a pacifier (a blind nipple). Nonnutritive sucking is one method of lowering the arousal level of an excited or upset infant. It also has been suggested as a buffering system, or means of protecting the infant from overwhelming stimulation (Bruner, 1973). Sleep, crying, and gaze aversion are also buffering systems. These systems may be hierarchically ordered, with the functioning of one system precluding the functioning of another. It can be hypothesized that the availability of the nonnutritive sucking buffering system will reduce the activity of the gaze aversion buffering system. Some support for this hypothesis is provided by Bruner's (1973) study of infant visual behavior with and without a pacifier. Saccadic movements were reduced by sucking, and they were reduced more for an unconventional film (achromatic animation of varied geometric shapes, stripes and textures) than for a more conventional film (a mother playing with her young child). Although not statistically significant, there was also more looking at the conventional film during pacifier use than without the pacifier. 17 Other studies of nonnutritive sucking also support the hypothesis that sucking can serve a buffering function. Brazelton (1967) inter— viewed mothers about their infants' use of thumb and finger sucking and found that one of the times it was most likely to occur was during mild frustration. Likewise, Lester, Bierbrauser, Selfridge and Gomeringer (1976) reported that children who seem to have difficulty averting their attention from environmental stimulation engage in more nonnutritive sucking in the natural environment than more easily distractible children, presumably as a means of escaping information overload. Most studies of nonnutritive sucking have taken place in labora- tories or hospitals, and have involved newborns or (less frequently) young infants. Again, a number of these studies support the notion of sucking as a buffering device. The most general finding is that sucking reduces arousal (Kaye, 1967; Kessen & Leutzendorff, 1963; Kessen, Leutzendorff, & Stoutsenberger, 1967). However, additional studies have shown that the direction and size of change in arousal produced by sucking is related to presucking arousal level. Both Kent (reported by Levin & Kaye, 1966) and Rovee and Levin (1966) found that sucking lowered the activity level of highly active infants, raised the activity level of quiescent infants, and did not change the activity level of moderately active infants. Thus, all infants were at approximately the same arousal level while sucking, regardless of their pre-sucking level. The arousal level of a sucking infant also seems to be more difficult to change than the arousal level of a nonsucking infant. Birns (reported by Bridger, 1962) observed the responses of infants with and without a pacifier to cold water applied to the foot. The sucking infants showed a slight increase in arousal and an increase in sucking. The nonsucking 18 infants showed a much larger increase in arousal. Bridger concluded that sucking served to protect the infants from, or make them less sensitive to, a stressful stimulus. In contrast to Birns' study, infants also have been observed to decrease their sucking in response to an external stimulus (e.g., Bronshtein & Petrova, 1967). Crook, Burke, and Kittner (1977) suggest that the sucking response depends on the intensity of the stimulus, with moderate stimulation eliciting an orienting response which includes sucking suppression, and with more intense stimulation eliciting a defensive reaction which includes sucking potentiation. Again, this hypothesis is consistent with the notion that sucking acts as a buffering system. Little direct information is available on the influence of sucking on looking behavior. Only Bruner's (1973) study has compared visual behavior during pacifier use and non-use. Mendelson and Haith (1976) compared newborn looking during bursts and pauses in sucking with a continually available pacifier and found no evidence for coordination of the two systems. Older infants (9 weeks) seem to alternate sucking and looking (Bruner & Bruner, 1968), and infants as young as 5 weeks can learn to suck as a means of focusing a moving picture (Kalnins & Bruner, 1973). It appears that the relevant variable may be pacifier availability rather than the actual sucking response. In Bruner's (1973) study, differences in saccadic movements were reported for pacifier availa- bility vs. pacifier nonavailability rather than sucking vs. nonsucking. Similarly, infants in Rovee and Levin's (1966) study showed shifts in arousal level when given a pacifier regardless of their actual sucking 19 responses. In the present study, infant gaze behavior was observed both with and without a pacifier. If the suggestion that infants avert gaze to modulate arousal is correct, then mean gaze length ought to be longer and mean gaze aversion shorter during pacifier availability. Including pacifier availability as a manipulation in the present study required consideration of several other factors. Since hungry infants are more aroused and suck more than satiated infants (Bridger, 1962; Kaye, 1967), initial hunger was kept as equivalent across infants as possible, by asking the parents to feed the infant within one hour preceding arrival at the laboratory. Other potential influences on infant nonnutritive sucking behavior were assessed through parental report. For example, whether the infant is breast or bottle fed has been shown to affect nonnutritive sucking (Dubignon & Campbell, 1968). Also, the amount of infants' experience with pacifiers can be expected to influence the ease with which a pacifier is accepted and retained. Another consideration in this study was the effect of infant pacifier use on parental interactive behaviors and infant nongaze behaviors. For example, infant smiling and vocalizing were expected to occur less frequently during pacifier use. Also, infant activity level was expected to be generally lower during pacifier use. These changes in infant behavior may have produced concomitant changes in parental behavior. If infant gaze alternations are related to information processing as well as, or instead of, changes in arousal level, individual differ- ences in infant social gaze should be related to individual differences in performance on cognitively-based tasks. Several studies have found 20 such relationships, although the measures of social gaze have been related to quantity rather than patterning, and thus do not directly address the question of the role of information processing in the regulation of infant gaze. Moss and Robson (1968) tested their sample of 3-month-old infants in the laboratory one week after observing their mutual gaze with their mothers at home. The tests involved the measurement of total fixation time to photographs of social (faces) and geometric (checkerboards) stimuli. Mutual gaze was significantly correlated with total fixation of the social stimuli for females only. Mutual gaze and total fixation of the geometric stimuli were not correlated for either sex. The authors suggest that the males were functioning at an earlier develop- mental level than the females as far as facial schema development was concerned, and this prevented their fixation times from being related to mutual gaze. Noll (1971) also tested the relationship between maternal looking and laboratory visual behavior. The number of times a mother looked at her infant during a 2-minute waiting period was used to divide the infants into high and low maternal visual contact groups. The infants in the high group were found to look longer at a series of pictures than the infants in the low group. No sex differences were reported. Along this same line, Beckwith, Cohen, Kopp, Parmalee, and Marcy (1976) found that more mutual gazing between mothers and 8-month—old preterm infants was associated with higher infant performance on a developmental test and a sensorimotor test at 9 months. Moreover, better cognitive performance at 9 months was associated with more positive social interactions at 1 and 3 months. One correlation, 21 however, did not fit this pattern. Higher scores on the develOpmental test were associated with less mutual gaze during feeding at 3 months. No explanation of this finding was offered by the authors, and insufficient information is available in their report to speculate on the factors underlying this correlation. Relationships between parent and infant interactive behaviors and infant responses to nonsocial (and one social) stimuli were investigated in the present study. Parent Characteristics SEX Over the years, most studies of parent-infant interaction have been concerned only with the mother. However, during the past five years considerable research interest has been directed toward the father's role during infancy (Parke, in press). Not only are fathers and infants interested in one another, but they become attached to one another in much the same way that mothers and infants become attached. In addition, fathers are competent caregivers, even as early as the newborn period. Qualitative characteristics of father-infant and mother-infant interactions seem to differ, with fathers engaging more in play and mothers engaging more in caregiving. This finding implies that fathers and mothers play different roles in the infant's social world (Lamb, 1975). Thus far, three studies have compared mothers and fathers in face-to-face interactions with their young infants. Parke and Sawin (1977) observed mothers and fathers separately during toy play with their 3-month-old infants. Although in many ways the behavior of 22 mothers and fathers in this context was similar, there were some slight differences. Fathers imitated their infants more and were slightly less stimulating (visually and auditorily) than mothers. Based on the results of Field's (1977) study, which found that infants gazed more when their mothers imitated them than when their mothers tried actively to get their attention, it might be expected that these fathers were more effective than the mothers in maintaining their infant's visual attention. However, no data on the infants' visual behavior were presented. In addition, the work of Brazelton et a1. (1974) suggests that the synchrony between parent and infant behaviors is also an important influence on infant attention. Since most fathers spend less time with their infants than mothers do, they may be less responsive to infant cues and less experienced at synchronizing their behaviors with the infants' attentional cycles. This could attentuate the positive effects of the fathers' level of stimulation. In addition, although fathers engage in slightly less auditory and visual stimulation of their infants than mothers do, they seem to be more inclined than mothers to engage their infants in rhythmic physical "games" (Field, 1978; Yogman, 1977). Mothers were observed to be more inclined than fathers to use soft, repetitive, burst-pause vocalizing (Yogman, 1977) and to hold their infant's limbs (Field, in press). These results suggest that parents differ in the types of stimualation they use with their infants. Because of procedural variations in these three studies, it would be presumptuous to use them as a basis for specific hypotheses about differences in maternal and paternal behavior in face-to—face inter- action with their infants. Parke & Sawin (1977) reported parent-infant 23 play with a toy, Yogman (1977) studied only five families longitudi- nally, and Field's (1978) study included fathers who were primary caregivers as well as fathers who were secondary caregivers. It does seem reasonable to hypothesize that father-infant and mother-infant interaction will differ. How or how much they will differ cannot be predicted, although differences in modality of stimulation may be more likely than other differences. Since Field (1978) found differences between fathers who were primary caregivers and those who were secondary caregivers, it also seems reasonable to hypothesize that the amount of experience a father has had with his infant will influence their inter— action. In Field's study, fathers and mothers who were primary caregivers were more similar to one another in their behavior than to fathers who were secondary caregivers. More experienced fathers thus should have been more similar to mothers than less experienced fathers in the present study. PARENT ATTITUDES Many studies of parent-infant interaction have considered only the directly observable behaviors of parents and infants. Parke (in press) suggests that more attention needs to be paid to "parental cognitions, perceptions, attitudes and knowledge" as mediators of observable behaviors. Research on gaze interactions between adults provides support for this approach, since the amount of interpersonal gaze has been found to be related to personality dimensions and attitudes toward interactive partners. Argyle and Cook (1976) conclude that "liking" is the most important determinant of the amount of social gaze by adult. Very likely, parents differ in how much they like their infant, for any 2A of a number of reasons. The possible influences of such infant charac- teristics as temperament and physical attractiveness on gaze inter- actions (discussed earlier) are probably due to their effects on the parents' liking of their infants. In general, more gaze should be associated with more positive parental attitudes toward their infant. However, it was suggested (in the section on infant sex) that this relationship will hold for mothers of girls but not for mothers of boys. A similar effect may exist for fathers, although there is no evidence. Parents' attitudes toward their infants were assessed in the present study by questionnaire. Individual differences in parental looking may be related to general personality characteristics as well. Robson (1967) suggested that some mothers are predisposed to gaze aversion which results in less mutual gaze between them and their infants. Mothers also differ in their responsiveness to infant cues (Blehar et a1., 1977) and in their interpretations of infant behaviors (Lewis & Rosenblum, 197A). Assessment of parental personality dimensions is beyond the scope of the proposed study. Dyad Characteristics INTERACTION BETWEEN PARENT SEX AND INFANT SEX So far, the effects of parent sex and infant sex have been discussed separately. There is evidence, however, that mothers and fathers differ in their responses to male and female infants, and that male and female infants differ in their responses to mothers and fathers. In particular, same-sex relationships seem to differ from different-sex relationships. Even in the newborn period, mothers and 25 fathers differ in their ratings of sons and daughters. Rubin, Provenzano, and Luria (197%) found that fathers rated daughters as cuddlier than sons whereas mothers rated sons as cuddlier than daughters. This cross-sex physical contact preference is maintained during the infant's first 3 months, with Parke and Sawin (1977) reporting that during play interactions, fathers held daughters close \j>fif/ more than sons, and mothers held sons close more than daughters. Other 8 types of stimulation, however, seem to be directed more to the same-sex infant. Fathers and mothers in the Parke and Sawin (1977) study looked at the same-sex infant more, touched them more, and stimulated their feeding more. Rebelsky and Hanks (1971) also found that fathers talked more to 3-month-old sons than daughters. This study was based on a very small sample which showed very low levels of talking, and so this result must be interpreted cautiously. Observations of older infants in inter- action with their parents confirms this trend; fathers tend to be more involved with sons and mothers more with daughters (cf. Parke, in press). Differences between male and female infants in how they respond to their parents is evident in some studies of attachment and social behavior in older infants (Parke, in press). For example, Spelke, Zelazo, Kagan, and Kotelchuck (1973) found that one-year-old girls spent more time near and vocalized more to their mothers while boys behaved similarly toward their fathers. Whether or not young (pre-attachment) male and female infants also differ in their responses to mothers and fathers is not known. It is known that young infants can discriminate their mother from their father (Yogman, 1977), and that mothers and fathers behave differently toward male and female infants. These 26 differences in parental behavior may produce differences in infant behavior. However, it is unlikely that young infants have sufficient knowledge of their own sex to recognize which parent is the same sex and which is the different sex from themselves (see Lewis & Brooks, 1975, for a discussion of the development of gender identity). It was hypothesized that parental behavior, but not infant behavior, would be influenced by the interaction between parent sex and infant sex. Specifically, it was expected that fathers would look more at and be more active with sons, and mothers would look more at and be more active with daughters. CONTEXT Contextual factors that affect parent-infant gaze interactions include the familiarity of the environment, the purpose of the inter- action, and other stimuli present in the environment. Because of contextual differences, gaze interactions in a structured laboratory setting are likely to differ from gaze interactions in the natural environment of the home (Belsky, 1977). In the home, mothers have been found to look at their infants 27% of the time overall (Lewis & Lee-Painter, 1974) and 70% of the time during play, spoon feeding, and bottle feeding (Peery A Stern, 1975). Although the laboratory studies of parent-infant face—to-face interaction reviewed thus far do not report exact values for parental looking, the impression from their descriptions is that parents look at their infants most of the time in this setting. Brazelton et a1. (197A) report that mothers tended to perceive the laboratory interaction as a period for heightened attention-getting activity. Likewise, the infants' behavior differed in 27 some ways from that commonly observed in the home, indicating that they too were aware of the unfamiliarity of the situation. The reason for face-to-face interaction also influences the gaze interaction. Peery and Stern (1975) found subtle differences in mother and infant gaze behaviors during play, bottle feeding, and spoon feeding. For example, infants looked away for longer periods during feeding than during play, and mothers' looks were much shorter during spoon feeding than during bottle feeding or play. These differences probably were the result of the different purposes of these inter- actions. Presumably mothers provide different types of attention- getting stimulation during feeding, and they must look away more frequently during spoon feeding in order to get the food. The interactions in the present study were limited to face-to-face play in an unfamiliar laboratory setting. Although such behavior is unlikely to be strictly representative of naturally occurring parent- infant interactions, it may be an example of parental behavior in certain types of stressful situations to which the infant must learn to respond (Brazelton et a1., 1974). Ideally, laboratory observations should be compared with home observations to discover commonalities and differences between the two contexts. Conclusions from the present research can not be automatically generalized to other contexts until such commonalities are demonstrated (Belsky, 1977). PREVIOUS EXPERIENCES OF THE DYAD Both the quantity and quality of the previous interactions of a dyad may be related to their interaction in a laboratory setting. For example, some of the differences between mother-infant and father-infant 28 dyads may be a function of different experience rather than the parents' sex (recall the comparison of primary and secondary caregiver fathers by Field, 1978). Although effects of the father's availability and amount of contact are evident in attachment relations (reviewed by Parke, in press), little information is available on how previous experience affects interactions between fathers (or mothers) and young (pre-attachment) infants. Many observers of early adult-infant interactions have noted the conversational quality of these interactions. The adult and infant seem to be communicating with one another. Describing the nature of these communicative exchanges requires at the least a fairly broad definition of communication. Within an ethological framework (Smith, 1965), the presence of a communicator, a signal, and a recipient are essential to communication and communicative events occur when two individuals establish a social relationship by use of a signal. This is true of adult-infant interchanges, since young infants can produce a number of signals that have meaning for adults. The young infant's communication skills are limited, however, in that signals are fairly gross (e.g., cries are only moderately specific) and there is little evidence that the young infant's signal production is intentional (Bates, Camaioni, & Volterra, 1975). However, the infant's behaviors do affect the adult, and as such are labeled communicative by some investigators (Als, 1977; Bakeman & Brown, 1977; Costello, 1976; Meier, 1975; Trevarthan, 19?”). Likewise, adult behaviors which (potentially) affect the infant's behavior are considered communicative. The communicative exchanges between adults and infants are necessarily nonverbal, which suggests the problem of determining the meaning of communicative acts. Smith (1965) 29 points out that the message of a communicative signal (a behavior which encodes some aspect of the CNS of the communicator) may not be the same as its meaning (the recipient's choice of interpretations of the signal). This mismatch may be particularly common in adult-infant exchanges. Although the code for some signals may be held in common through genetic mechanisms (cry=distress, for example), many other signals may have different meanings for the adult and infant. In fact, parents seem to infer meaning and intentionality from many infant behaviors during social interaction (Anderson, 1970; Bates et a1., 1975). The inferred meaning may determine the adult's response, and this consistency of response to what may actually be random behaviors in the infant may contribute to the infant's learning of the meaning of nonverbal (and eventually verbal) communicative acts (Anderson, 1970). The locking behavior of young infants serves as a communicative signal in adult-infant interaction. Mothers normally become less active when their infants look away from them (Field, 1977), which implies that mothers perceive and respond appropriately to this signal. If, as Stern has suggested (1974a), the infant looks away to reduce his level of arousal, then the message and the meaning match. There may be cases, however, where the message and meaning do not match. Stern (1971) described a mother who responded to her infant's gaze aversions by increasing her attempts to elicit attention. The interaction tended to be unsatisfying and upsetting to both individuals. Also, infants who do not alternate gaze appropriately (e.g., blind infants, Fraiberg, 197A; autistic children, Hutt & Ounsted, 1966; a gaze-avoiding infant, Thoman, 1975) but instead look away most of the time, are generally perceived as rejecting and uncommunicative. Mothers may differ as well in their 30 interpretation of gaze away behavior (Lewis and Rosenblum, 1974). An "optimistic” mother may interpret her infant's gaze aversion as a temporary interest in something else whereas a more "pessimistic” mother may interpret it as a personal rejection. It seems pertinent within the framework of the proposed study to ask parents why they think their infants look at and away from them during play. The meaning given to these behaviors should be considered in the interpretation of their message value. One reason why adult-infant interaction is so frequently labeled "communicative" is that it follows a conversational, turn-taking pattern. Body movements in response to speech (Condon & Sander, 1974), mutual gaze patterns (Jaffe et a1., 1973) and mutual vocalization patterns (Bateson, 1975; Stern, Jaffe, Beebe, & Bennett, 1975) are very similar to those observed in adults during conversation. It has been proposed that these rhythms and tendencies toward alternation are innate and form the basis for communicative interchange (Trevarthan, 1974). If nothing else, such behavior may contribute to the adult's impression that the infant is responding to him. Considering the importance generally attributed to these early interactions (e.g., Meier, 1975), establishing them as a pleasant experience can be considered an important goal in itself (Bates et a1., 1975). Brazelton et a1. (1974) describe mother-infant interaction as a time of "rule-learning" for both individuals. Each must learn what behavior patterns to expect from the other. They suggest that the most important rule the mother must learn is to be "sensitive" to her infant's capacity for attention and nonattention. Mothers who are more ”sensitive" (who synchronize their interactive behaviors with their 31 infant's attention and nonattention) receive more attention from their infant and participate in a more smoothly running, positive interaction. Evidence that infants also learn what kind of behavior to expect from their mothers comes from two observations of infant responses to mothers who were behaving "abnormally." Trevarthan (1974) observed infants' responses to an image of their mother interacting with an adult. The infants appeared puzzled, some seemed to try to elicit her attention, and then became dejected and withdrawn. Similar patterns of behavior were observed when mothers were asked to simply sit in front of their infants without responding or changing their facial expression (Brazelton et a1., 1975). It was not feasible to directly assess the nature of previous interactions between the parents and infants in the present study. However, a measure of the amount of previous interactive experience was obtained through parental report. In addition, the existence of infants' expectations for interactive behavior from their parents was tested by observing their responses to the cessation of their parents' behavior. Parents were asked to stop interacting for approximately one minute during the laboratory observa- tion. Based on Brazelton et al.'s (1975) description, infants were expected to first increase and then decrease their interactive behaviors. The amount of change in their behavior from when the parent was interacting to when the parent was not should have reflected how much the infant's cognitive expectancies related to the interaction were violated. It was hypothesized that low interactive dyads would differ from high interactive dyads, with the infant's expectancies being less well established in the low interactive dyads. 32 Summary Of Purpose And Hypotheses The purpose of the present study was to investigate some of the processes believed to underlie variations in the quantity and patterning of social gaze between parents and young infants. Specifically, the influences on gaze interactions of the infant's sex, temperament, physical attractiveness, changes in arousal, and cognitive expectancies, the parent's sex and attitudes, and such dyadic factors as the inter- action between parent sex and infant sex, the context of the inter- action, and previous experience of the members of the dyad with one another were assessed. Hypotheses about these factors are listed below: Infant characteristics Sex 1. There will be no sex differences in infant gaze behaviors. 2. There will be more mutual gaze between mothers and daughters when the mothers have positive attitudes toward their infants. This relationship will not hold for mothers of boys. Birth condition and temperament 3. There will be less mutual gaze when the parent perceives the infant as temperament- ally difficult. 4. Infants perceived by their parents as difficult will engage in less gazing and smiling and more fussing. 33 Predisposition toward gaze aversion 5. There will be less looking by infants whose parents report they have a low level of interest in social gaze. Physical attractiveness 6. Parents who perceive their infants as cuter will look longer at their infants. Arousal and information processing 7. Mean gaze length will be longer during paci- fier availability. 8. Mean gaze aversion will be shorter during pacifier availability. 9. Less infant smiling and vocalizing will occur during pacifier availability. Parent characteristics Sex 10. Father-infant and mother-infant interactions will differ. 11. More experienced fathers will be more similar to mothers than less experienced fathers. Attitudes 12. More parent gaze will be associated with more positive attitudes toward their infants. 34 Dyad characteristics Interaction between parent sex and infant sex 13. 14. Fathers will look more at and be more active with sons, and mothers will look more at and be more active with daughters. There will be no effect of this interaction on infant behaviors. Previous experience 15. Infants will respond to the cessation of their parents' behavior by first increasing and then decreasing their behaviors. Infants in dyads with low interactive exper- ience will change their behavior less in response to parent behavior cessation than infants in dyads with high interactive experience. CHAPTER 3 METHOD SUBJECTS The Ingham County (Michigan) birth records were used to obtain the names of parents of 3-month-old infants in the Lansing metropolitan area. Parents whose addresses were available (either in the birth records or in the telephone book) were sent a letter describing the study and a map to the research building (see Appendix A, A-1). Interested parents telephoned the experimenter to make an appointment for testing. In addition, many of the parents were contacted by telephone after first receiving the letter. At this time the parents were asked about their willingness to participate in the study, and their questions about it were answered. A total of 41 infants were tested. Of these, 17 did not complete the study due to fussiness. Of the remaining 24, 12 were first-born infants tested with both parents (Group 1) and 12 were pacifier-using infants tested with their mothers (Group 2). Half the infants in each group were girls and half were boys. Characteristics of the 24 families who completed the study are summarized in Tables 1 and 2, divided by group and by sex. All of the fathers and 12 of 23 mothers were employed in jobs ranging from skilled labor to professional. Over half the mothers had attended college and over half the fathers had completed college. 35 36 Table 1 Background Information (Means and SDs) Infant's age (days) Infant's birth wt. (02.) l ( Mother's age (yrs) Father's age (yrs) Years married Pregnancy complications yes no Labor complications yes no Delivery complications yes no Labor medication yes no Delivery medication yes no Infant illnesses since birth yes no Pacifier use yes Infants Group 1 Group 2 92.25 92.08 (6.48) (6.36) 120.92 119.25 (8.62) (16.18) 26.00 26.36 (3.52) (2.77) 28.17 28.91 (4.15) (3 59) 4.67 5.46 (2.93) (3.24) Group 1 Group 2 Male Female Infants 90.25 94.08 (6.77) (5.35) 17.50 122.67 12.97) (12.43) 26.36 26.00 (4.06) (2.09) 28.55 28.50 (4.11) (3.73) 5.46 4.67 (3.70) (2.39) Table 2 Background Information (Frequencies) Male Female Infants Infants 2 4 10 8 0 2 12 9 2 4 10 8 3 8 9 3 7 8 5 4 2 2 10 10 9 6 3 6 no 423(1) U'ION \OUJ [—4 . CID-l: Tot al Group 92. (6. 120 O (12 26. (3. 28 (3. 5. (3. 17 28) 08 .70) 17 11) .52 82) 04 04) Total Group 37 Type of feeding breast 6 4 5 10 bottle 3 7 5 5 10 combination 3 1 2 2 4 Birth order first 7 8 12 3 15 second 3 3 0 6 6 third 1 l 0 1 l fourth l O 0 2 2 Mother's education less than HS completion 0 l 0 l 1 HS completion 4 3 3 4 7 some college 3 6 6 3 9 college completion 2 l l 2 3 some grad school 1 1 1 l 2 master's degree 1 0 l 0 l doctorate O 0 O 0 0 Father's education less than HS completion 0 1 0 l 1 HS completion 2 1 2 l 3 some college 3 2 1 4 5 college completion 3 3 2 4 6 some grad schmol 2 3 4 l 5 master's degree 1 2 2 0 2 doctorate 0 O 1 0 l PROCEDURE In Group 1, the parents, infant, and experimenter first were seated in the experimental room. The testing procedure was explained, and signed consent obtained from both parents (see Appendix A, A-2). For half the infants of each sex (Order 1) the mother remained in the exper- imental room and the father left the room with the experimenter. The mother was seated facing the infant, who was seated in an infant seat placed on a table top. The experimental room had one-way mirrors around three sides; the mother and infant were placed so that each of their faces could be seen clearly by two of the observers. The mother was asked to interact with the infant as naturally as possible, without removing the infant from the infant seat or blocking the observers' views of the infant. She had been told that both her own 38 and her infant's behavior would be recorded, but that the major concern of the study was the infant's behavior. This unstructured interaction continued for approximately 4 minutes, with the first 2 minutes allowed for adaptation to the situation, and the second 2 minutes for recording. After two complete minutes of interaction were recorded, the mother was signalled by the experimenter (a light onset behind the infant's head) to stop interacting and to simply sit and gaze at the infant's face. This is labeled the "Stop" condition. The infant's behaviors during this condition were recorded for 1 minute. If the infant started to cry during the interaction segment of the test, the mother was instructed to do whatever she felt was necessary to calm her infant, including removing the infant from the infant seat, feeding or changing the infant, or calling in the father. Testing resumed at the discretion of the mother and continued until 2 minutes of interaction were recorded. If the infant started to cry during the final minute when the mother was not interacting, that part of the experiment was terminated and further testing resumed only after the infant was calm. Only one such termination occurred. After the interaction with the mother (Part A) was completed, the experimenter and father reentered the experimental room. The testing procedure was reviewed for the father, and the experimenter and mother then left the room. The interaction between the infant and father (Part B) then proceeded. Infants assigned to Order 2 interacted first with their father and then with their mother. The procedure for Group 2 was the same as for Group 1, except that the infant interacted only with the mother. Part A was the same as Part A for Group 1, whereas in Part B the infant's pacifier was provided by 39 the mother. Again, the order of Parts A and B were counterbalanced across sex of infant. Following the two interaction segments, parents were asked to complete a series of questionnaires (see Appendix A, A-3 to A-6). The first questionnaire (A-3, Today's Schedule and Trip to the Laboratory) was designed to determine the infant's schedule on the day of testing. The second questionnaire (A-4, Background Information) solicited back- ground information about the infant and parents. The third question- naire (A-5, Parent Questionnaire) was concerned with the parents' attitudes toward their infant and parenthood. Only responses to questions 1-17 and 31 and 32 were analyzed. For items with scale check-offs, the distance from the left scale end-point to the parent's check was converted to a percentage of the distance between the two end-points. The final questionnaire (A-6, Survey of Temperamental Characteristics) consisted of the 46 items from the Carey scale (1972) which are used to assess "difficulty" of infants' temperament. Temperament ratings were calculated by first assigning a value of 1 (low difficulty), 2 (medium difficulty), or 3 (high difficulty) to the response to each item according to Carey's scoring scheme (Carey, 1972). The average score for each of the scales of Rhythmicity (questions 1, 2, 5, 6, and 18), Adaptability (questions 4, 7, 15, 16, 22, 27, 28, 30, 32, 36, 42, and 44), Approach (questions 3, 12, 26, 29, 37, 41, and 43), Intensity (questions 9, 11, 14, 17, 21, 23, 25, 33, 35, 38, 40, and 46), and Mood (questions 3, 8, 10, 13, 17, 19, 20, 22, 24, 31, 32, 34, 37, 39, 41, and 45) was calculated. Unanswered questions were not included in these averages. Finally, the temperament (difficulty) rating was determined by averaging these scale scores. 40 While the parent(s) completed the questionnaires, the infants' responses to several visual and auditory stimuli were recorded (Infant Test). First, the infant was shown a color photograph (containing an array of patterned pottery) for 30 sec. Total amount and frequency of looking, smiling, and vocalizing were recorded. Then, a pair of black and white checkerboards (one 4 x 4 checks, the other 12 x 12 checks) was presented twice, for 30 see each time, with side of presentation reversed on the second presentation. Total amount and frequency of looking at each checkerboard were recorded. The infant was then held before the one-way mirror for 30 sec. Again, total amount and frequency of looking, smiling, and vocalizing in response to the mirror image were recorded. The infant then was replaced in the infant seat. A ring of colored plastic beads was dangled in front of the infant's face and moved first in one direction and then the other. Then, a set of keys was jiggled once on each side of the infant's head (out of the infant's eyesight). For each of these trials, eye movement and head movement in the direction of the beads or keys were recorded as being present or absent. DEPENDENT MEASURES All interactive behaviors were recorded continuously as they occurred. Observers were positioned behind the one-way mirrors on each side of the experimental room so that they had a clear view of either the parent or infant. There were two observers for the infant and two for the parent. Behaviors were recorded by pressing buttons attached to an event recorder. One observer recorded infant gaze and another recorded parent gaze. Gaze at the partner's face was recorded rather 41 than eye contact, since eye contact is less reliably discerned at a distance (Argyle & Cook, 1976). The other two observers recorded the presence or absence of smiling and vocalizing. A smile was defined as a widening and/or upturning of the corners of the lips which was judged to reflect positive affect. Any vocal sound produced by the infant was considered a vocalization. Although an attempt was made to stop recording whenever the infant became upset, recorded vocalizations included those of both a positive and negative nature. Adult vocaliza- tions were recorded as present or absent and defined as any vocally— produced sound (whether or not a word). Inter-observer reliabilities were assessed before the study was begun and periodically throughout the course of data collection. Each of 15 observers was tested at least once before being allowed to record usable data. The two-minute period preceding the two-minute recordings were frequently used to re-test observer reliabilities. Since not all of these were systematically scored, but rather checked by eye, the reliabilities reported here are based on the initial tests. The relia- bility measure used was the percent agreement on the presence of a behavior during a particular second for a pair of observers. Agreement for gaze averaged 85% (range 80% to 96%), and for smile and vocalize 75% (range 68% to 94%) and 78% (range 70% to 94%) respectively. Although the reliabilities for smile and vocalize were lower than expected, further practice did not seem to improve them. The differences between observers seemed to be the result of differences in reaction time and the small unit of analysis (one see). This may bias the sec-by—sec analyses, but not the analyses of the frequency, percent, mean, and interval of behaviors. 42 At the end of each experimental session, the four observers were asked to rate the attractiveness of the infant, mother, and father (if present) on a 5-point continuous scale (see Appendix A, A-7). Ratings were transformed to a percentage of the total scale with higher numbers indicating higher attractiveness ratings. DATA REDUCTION The output of the event recorder was scored second by second. Each sec of each interaction was scored as 0 (behavior absent), 1 (behavior present), 2 (behavior both present and absent, one transition), 3 (behavior both present and absent, two transitions), or 4 (behavior both present and absent, three transitions) for parent gaze, parent smile, parent vocalize, infant gaze, infant smile, and infant vocalize. In addition, each sec was scored for the presence (1) or absence (0) of mutual gaze. During the Stop condition, only infant behaviors were scored. For each behavior within a part (A or B) and condition (Interacting or Stop), four summary variables were produced: 1)the frequency of the behavior (doubled for the Stop condition since it was half as long as the Interacting condition), 2)the percent of total time spent engaged in the behavior (for Secs coded 2, 3, or 4, one half see was added to the sum), 3)the mean length of behavior occurrence, and 4)the mean length of interval between behaviors. These summary variables were treated in multivariate analyses of variance, with simple effects tests performed on significant univariate interactions. The simple probabilities of occurrence of each interactive behavior were compared with the conditional probabilities of occurrence during parent or infant gaze, and preceding infant gaze onset or offset. Pearson 43 product-moment correlations between interactive behaviors, observer attractiveness ratings, parent questionnaire responses, and Infant Test items were calculated for the entire group of mothers and infants, and for Group 1 mothers, fathers, and infants. Multiple regression analyses also were used to test some of the hypotheses presented in the Introduction. CHAPTER 4 RESULTS The results are presented in several sections. First, the inter- active behaviors of parents and infants were analyzed separately for each group using multivariate analyses of variance (MANOVA). Then, the two groups were combined and detailed analyses of mother and infant interactive behavior (Part A) and relationships between interactive behaviors, performance on the Infant Test and the mother's responses to the questionnaires were performed. A set of Sex of Infant x Group x Order MANOVAs (Tables C-1 to C-4, infant and parent interactive behaviors analyzed concurrently) revealed no differences between the groups in mothers' or infants' interactive behaviors, thereby justifying combining the groups. Some analyses were repeated using just Group 1, for the purpose of comparing mothers and fathers. Interactive Behaviors - MANOVAs These analyses were performed separately for each group and, within groups, separately for parents and infants. Parent MANOVAs were of the form: Sex of Infant (2) x Order (2) x Part (2). Infant MANOVAs included an additional factor, Condition (2). Part and Condition were within subjects factors. For Group 1, Part A was the mother interaction and Part B was the father interaction; for Group 2, Part.A was the non-pacifier interaction and Part B was the pacifier interaction. Condition refers to the Interaction and Stop conditions. MANOVAs were 44 45 performed separately for frequency, percent, mean, and interval with the three variables, gaze, smile, and vocalize, analyzed concurrently. MANOVA tables are presented in Appendix B, Tables B~1 to B—24. Univariate ANOVA results and standardized discriminant coefficients are reported for significant multivariate effects. Means and standard deviations for reported univariate effects (whether or not significant) are given in Appendix B, Tables B-25 to B-43. Simple effects tests for significant interactions are shown in Tables B-44 to B—51. GROUP 1 The analyses of parents' behavior in Group 1 (Tables B-1 to B—4) revealed no significant effects. Thus, this group of mothers and fathers did not differ from one another in their interactive behavior with their infants in a structured laboratory situation. Neither were the behaviors measured influenced by infant sex or the order in which parents interacted with their infants. In contrast, there were several significant influences of the independent variables on infants' interactive behaviors. These effects were on the frequency and percent of behaviors; there were no effects on mean or interval. The multivariate test of the effects of Sex of Infant on the frequency of infant behaviors (gaze, smile, vocalize) was significant, but none of the univariate tests were (Table B-5). Although the effect for gaze was nearest significance of the three (p<.06), the standardized discriminant coefficients were nearly equal, indicating that all three behaviors were influenced equally by the multivariate effect. There were three significant interactions involving the percent of 46 infant interactive behaviors (Table B—6). The Sex x Order x Parent interaction was accounted for primarily by gaze, E(1,8)=14.31,.p<.01. Simple effects tests (Table B—44) revealed that the only significant component of the interaction was the simple two-way interaction for Sex x Order during Part B, E(1,8)=6.38, nfi.05. Since the simple simple main effects contained in this interaction were not significant, no further interpretation of the interaction could be made. The multivariate test of the interaction between Sex, Condition, and Parent was also significant, but none of the univariate tests even approached significance. Therefore, no attempt was made to explain this interaction. The relevant means are presented in Table B-29. The four-way interaction between Sex, Order, Condition and Parent was significant for infant vocalization, E(1,8)=7.61,.p<.03. Inspection of the cell means (Table B-30) revealed that one cell was substantially different from the others. The amount of vocalization by male infants in Order 2 with their mothers in the Stop condition was over four times as great as the largest amount of vocalization in any of the other cells. A Newman-Keuls test confirmed that the mean of this cell was significantly different from the means of all the other cells, with none of the other cells differing from one another. Checking the experi— mental records of the three infants in the condition with the extreme mean revealed that two of these infants had become fussy during the Mother-Stop condition. Since fusses were recorded as vocalizations, a large mean was obtained for this one cell. There were two significant effects in the analyses of mutual gaze (Tables B-9 to B-12). First, there was a Sex x Parent interaction for the frequency of mutual gaze, E(1,8)=5.92, n<.05. None of the simple 47 main effects were significant (Table B-45). The Sex x Order x Parent interaction for percent of mutual gaze also was significant, E(1,8)=11.19, p<.02. The means for this inter- action are presented in Table Bu28. All but one of the simple simple main effects are significant (Table B-46). The most apparent summariza- tion of this interaction is that more mutual gaze occurred with the first parent for male infants and with the second parent for female infants (see Figure 1). The pattern of means for this interaction corresponded closely to that for the same interaction with infant gaze. However, more of the simple effects were significant in this case. It seems that mutual gaze is determined largely by infant gaze, probably because parents gazed most of the time whereas infants looked away more frequently, but there was less variability in mutual gaze than infant gaze. GROUP 2 Although both Sex and Sex x Order multivariate effects were signi- ficant for the frequency of mothers' behavior in Group 2, none of the univariate effects were significant (Table B-13). The standardized discriminant coefficients indicate that the effects were strongest for smile, fairly strong for gaze, and minimal for vocalize. The effect of Sex was also significant for the mean length of infant smiles (Table B-19), E(1,8)=7.94, p<.03. However, Sex interacted with Order, again for smilin8,.E(1,8)=12.94,.p<.01. The Sex x Order interaction also was significant for the frequency of smiling (Table B-17),.§(1,8)=79.29,.p<.01, and the interval between smiles (Table B-20), £(1,8)=43.75,.p<.01. None of the simple main effects was signi- 48 50 l I . Female Infants - Order 1 I l 1 Female Infants - Order 2 I I 40 I I I 1 Male Infants - Order 2 I I Percent 30 I of time I spent I in mutual l gaze I 20 1 Male Infants - Order 1 I E I I l 10 I I I I I I I I 0 1 First Parent Second Parent Figure 1. Sex x Order x Parent Interaction for Mutual Gaze Percent 49 ficant (Tables B-47, B—50, and B-51). According to the main effect of sex, boys' smiles were longer than girls' (Table B-38). The interaction between Order and Part was significant for the frequency and percent of vocalizing,.E(1,8)=10.90, pfl.02, E(1,8)=13.69, .p<.01, respectively. The simple effects tests showed that none of the simple main effects for frequency were significant, but two out of four were significant for percent (Tables B-48 and B-49). More vocalizing occurred during the no-pacifier interaction than the pacifier inter- action when the no-pacifier interaction followed the pacifier inter- action, whereas there was no difference when the no-pacifier interaction preceded the pacifier interaction (means presented in Table B—36). In addition, more vocalizing occurred during the pacifier interaction when it came second than when it came first. In general, there seemed to be more vocalizing during the second interaction than during the first. Even though the simple effects for frequency of vocalization were not significant, the pattern of means corresponded with that for percent (Table B-34). The effect of Part was significant for smiling percent, E11,8)=5.68,.n<.05, and interval, E(1,8)=8.54,.p<.02 (Tables B—18 and B-20). There were more smiling and shorter intervals between smiles during the no-pacifier interaction than during the pacifier interaction (Tables B-35 and B-42). Part also influenced the frequency of mutual gaze (Table B—21), IK1,8)=5.37, pfi.05. Mutual gaze was more frequent in the no-pacifier interaction than during the pacifier interaction (Table B-43). Condition (Interaction vs. Stop) significantly influenced the percent of gaze (Table B-18), E(1,8)=7.86, n<.03, and the mean length of 50 smile (Table B-19), F(1,8)=8.87, p<.02. There were more gaze and longer smiles during the Interaction condition than during the Stop condition (Tables B-37 and B-40). All subjects - Part A INTERACTIVE BEHAVIORS - CONDITIONAL PROBABILITIES For each interactive behavior, first the simple probability of occurrence during any 1 sec interval was determined. The simultaneous conditional probabilities of the occurrence of infant smiling and vocalizing, and parent gazing, smiling, and vocalizing, given the occurrence of infant gaze during any particular second were calculated, as well as the conditional probabilities of the occurrence of behaviors other than parent gaze given the occurrence of parent gaze. Finally, transition conditional probabilities for behaviors other than infant gaze, given infant gaze onset and offset, were determined. An infant gaze onset was defined as a sec coded l, 2, 3 or 4 (gaze present) preceded by a sec coded O (gaze absent). Thus, only those gaze onsets preceded by at least 1 sec of gaze aversion were included. Similarly, gaze offset was defined as a sec coded O (gaze absent) preceded by a sec coded 1, 2, 3 or 4 (gaze present). The probabilities of occurrence of behaviors other than infant gaze in the sec preceding infant gaze onset or offset were defined as transitional probabilities. The mean simple, simultaneous, and transitional probabilities for all mothers and infants (Part A) are displayed in Table 3. Binomial tests (Siegal, 1956) were performed to compare the simple and conditional probabilities for each dyad. The test could not be used in cases where the simple probability was either 1.0 or 0.0 (the behavior 51 occurred either all of the time or none of the time), and the result was meaningless when the conditional probability was 1.0. Thus the table reports both the number of subjects for whom the simple and conditional probabilities differ significantly (p<.05) and the number of subjects for whom the test could be used meaningfully. As can be seen in the table, none of the comparisons were significant for more than half the subjects. Table 3 Interactive Behaviors - Simple and Conditional Probabilities Infant. .Parent Gaze Smile Voc Gaze Smile Voc Simple p .461 .115 .094 .850 .363 .757 Conditional p/given BG -- .156 .105 .848 .422 .790 Conditional p/given PG .463 .114 .097 -- .369 .759 Transitional p/ -- .074 .046 .799 .266 .687 preceding BG onset Transitional p/ -- .085 .090 .828 .410 .744 preceding BG offset Simple vs. conditional p/ given BG #binomial test -- 8/21 3/18 1/12 9/23 4/20 &sign test -- l6/l9' 8/17 9/16 19/21* 11/22 Simple vs. conditional p/ given PG #binomial test 1/23 3/21 0/18 -- 0/23 0/23 &sign test 8/15 6/10 8/10 -- 13/16* 9/14 Simple vs. transitional p/ preceding BG onset #binomial test -- 5/20 1/17 1/7 2/22 2/18 &sign test —— 6/20 3/18' 10/16 4/23' 10/23 52 Table 3 (cont'd) Simple vs. transitional p/ preceding BG offset #binomial test —- 4/20 4/17 0/16 3/20 2/17 &sign test -- 8/20 8/18 12/16 14/22 13/23 #For binomial tests: Number of subjects for whom test is significant (p<.05)/ Number of subjects for whom meaningful test could be done. &For sign tests Number of subjects with higher conditional p than simple p/ Number of subjects with conditional p different from simple p. *Significant at pfl.05 (twOotailed). The most consistent effects were for infant and mother smiling given infant gaze. There appeared to be a tendency for both infants and mothers to smile more while the infant was gazing than their base rates of smiling would predict. Sign tests were performed to determine whether any of the differences between simple and conditional probabilities were consis~ tently in one direction (see Table 3). Again, both infant and mother smiling were more likely during infant gaze. Mothers' smiling is more likely during mothers gaze as well. In addition, infant gaze onsets tend to be preceded by less infant vocalizing and less parent smiling than would be predicted by the base rates. These results must be inter- preted cautiously, since in all cases the differences between the simple and conditional probabilities are significant for only a few mother- infant pairs. Other measures Mothers‘ responses to the Parent Questionnaire and Survey of Temperamental Characteristics (see Appendix A, A-5 and A~6) are 53 summarized in Table C—5. Intercorrelations among the questionnaire items are reported in Table C-6. (Degrees of freedom for all correla- tions are 22; all tests are two-tailed with p<.05 unless otherwise indicated.) Although all the means and correlations are presented, only some were of particular interest. The mothers' temperament ratings of the infants were correlated with ratings of the infants' behavioral cuteness (3;.45), satisfaction with the infants' temperament (IF-.55), and similarity of the infants' temperament to their own (2;.44). More difficult infants were rated as being less cute in behavior, having less satisfactory temperaments and being less similar to the mother in temperament. The intercorrelations among the questions about physical appearance (questions 13, 14, 15, 16, 31, and 32) were also of interest. Of the six correlations between questions about the infant's cuteness (questions l3, 14, 15, and 16), three were significant (13 and 14, .r=.60; 14 and 16, n;-.47, and 15 and 16, r; 66) (note that low numbers indicate high cuteness on questions 13 and 14 but the opposite is true for questions 15 and 16). Correlations were highest for questions using the same scale (13-14 and 15-16), even though the questions in these pairs differ in content. It is interesting to note that the cuteness ratings on question 13 were not significantly correlated with the cuteness ratings on question 16, even though the questions are essen- tially the same. Both of these ratings were significantly correlated with the rating of behavioral cuteness (question 14), which indicates that the mothers took behavior as well as physical appearance into account in their cuteness ratings. They also may have considered the similarity of their infants' appearance to their own and their 54 husbands', since one of the cuteness ratings (question 16) was signifi- cantly correlated with the mothers' rating of how much the infant looks like their husbands (question 32, n;.49). Infants who looked less like their fathers were given higher cuteness ratings. In addition, questions 31 and 32 were significantly correlated (n:-.48), indicating that infants who were perceived as looking more like their fathers were perceived as looking less like their mothers. These correlations imply that mothers should perceive infants who look like them as cuter. The appropriate correlation (questions 16 and 31) approached significance (L=-.36, .05n)n>—sc>n>c>_._.n) .07 .AI .96 .61 .57 .17 .27 .20 .25 .06 .31 .03 .15 .68 .5A .13 .21 .18 .81 .51 .2A .63 .61 p< .21 .33 .23 .55 .66 .A0 .19 .19 .19 .01* .01* .87 .08 .15 .OA* .32 .31 .12 .08 .02* .6A .2A .03* Standardized Discriminant Coefficient ~I.A7 Source Sex (S) Order (0) S x 0 Condition (C) S x C O x C S x O x C Parent (P) x UJOUJOMOUJ NXNKNN OOO'UO'U'U 105 Table B-7 MANOVA Group 1 - Infants - Mean df d O\O\O\O\O\O\O\C\O‘\O\O\O\O\O\O\ ‘ ‘ ‘0 *0 ‘0 \O N. ‘0 0 ‘ ‘4 t! O wwwwwwwwwwwwwww \. Table B—8 MANOVA dNO-bw—A—AO—AO—‘ON—BUJ F .50 .02 .08 .A6 .68 .70 .0A .87 .58 .A7 .67 .36 .57 .IA .02 p< .09 .A5 .21 .73 .27 .59 .A5 .51 .30 .32 .09 .35 .66 .20 .A5 Group 1 - Infants - Interval Source Sex (8) Order (0) S x 0 Condition (C) S x C O x C S x O x C Parent (P) x MOI/JOUJOUJ xxxxxx OOO'UO'U'U df wwwwwwwwwwwwwww oxcnoxo‘owoxCho‘aoIcnoxoxoxm no so w v w v v ‘0 O a \o O u c Q OOOOOONO—INNLUOO-I F .144 .39 .85 .86 .13 .08 .02 .69 .05 .73 .35 .82 .148 .uu .149 106 TABLE B-9 ANOVA Group 1 - Mutual Gaze - Frequency Source Sex (8) Order (0) S x 0 Error Parent (P) S x P 0 x P S x O x P Error Source Sex (S) Order (0) S x 0 Error Parent (P) S x P O x P S x O x P Error Source Sex (S) Order (0) S x 0 Error Parent (P) S x P 0 x P S x O x P Error MS 3 8 18 1 1 12 MS A7. 397. 3A2. 2355. .07 162. 229. 5AA8. A86. 0 .00 .33 12. .83 .33 75. 3. .33 .67 00 00 00 60 90 A0 10 07 69 5A 72 df mad—sdmddd Table B-10 ANOVA df mAddAmddd Table B-11 ANOVA F .16 000 .c: .1: :6A .11 .92 .2A .11 OOU'IO F 0.02 .17 .15 OO .00 1 .33 .A7 .19 dOOO Group 1 - Mutual Gaze - Mean MS A5. 221 78. 17A. 60. 128. 81 9O .02 5A 30 71 .6A 513. 157. 52 58 df m—bdddmA—b—l F .32 .27 .A5 O-so .38 .82 .52 .26 WOOD p< .71 .53 .A5 .76 .05' .6A .76 Group 1 - Mutual Gaze - Percent p< .90 .70 .72 .00 .58 .52 .02‘ p< .59 .30 .53 .56 .A0 .50 .11 107 Table B-12 ANOVA Group 1 — Mutual Gaze — Interval Source MS df F p< Sex (3) 2.08 1 0.01 .9u Order (0) 3.85 1 0.01 .92 S x O 30.08 1 0.10 .77 Error 312.82 8 Parent (P) 9.01 1 0.09 .77 S x P A01.36 1 A.19 .08 O x P 5.33 1 0.06 .82 S x O x P A83.87 1 5.06 .06 Error 95.58 8 Table B-13 MANOVA Group 2 - Mothers - Frequency Standardized Discriminant Source MS df F p< Coefficient Sex (S) -- 3,6 8.7A .02“ Gaze 27.00 1 0 35 .58 1 86 Error 78.33 8 Smile 290.08 1 A.93 .06 2.25 Error 58.83 8 Vocalize 385.33 1 0.30 .61 0.30 Error 1298.92 8 Order (0) -- 3,6 1.6A .28 Gaze 1A7.00 1 1.88 .21 1.93 Error 75.33 8 Smile 30.08 1 0.51 .50 -2.19 Error 58.83 8 Vocalize 396.75 1 0.31 .60 O.A2 Error 1298.92 8 Part (P) -- 3,6 0.33 .81 SK P *- 3,6 0.76 076 Source Sex (S) Order (0) Source Sex (3) Order (0) S x 0 Part (P) S x P 0 x P S x 0 x P Source Sex (S) Order (0) S x 0 Part (P) S x P O x P S x O x P 108 Table B—IA MANOVA Group 2 « Mothers — Percent C)l'\.>—-||han>ha of of of of of of of of of of of of MS 1097. .15 .29 65. .53 .68 .23 .18 267. .81 .92 .18 18 3 728 128 6 76 650 181 87 98 38 Main Effects Part Part Part Part Part Part Part Part Order Order Order Order MS 880. 1085. 82. 2000 68 AA 93 .57 .01 2202 1307. 907. 538. 570. 1512. 750. .63 15 7A 68 38 09 63 I-‘N-EO #ONH ~8u3h'h' F OI-‘OOOOI—‘OOOON .80 .18 .09 .10 .00 .51 .68 .86 .09 .17 .10 .58 WWWOO‘N OOCDOUOKD'U NU'INNCD‘I .25 .08 .00 .18 .01 .A9 .25 .01 .55 .16 .33 .29 For Sex For Sex For Part For Part Sex x At level 1 At level 2 At level 1 At level 2 Simple Effects Tests 123 Table B-A5 Parent - Mutual Gaze Frequency Group 1 Simple Main Effects of Part of Part of Sex of Sex MS 28.20 18.00 27.00 12.00 For For For For For For For For For For For For For For For For For For For At At At At For At At At At For At At At At Sex Sex Sex Sex Order Order Sex Sex Sex Sex Order Order Order Order Part Part Part Part Sex level level level level Order level level level level Part level level level level NNI—‘H NNI—‘H NNI—‘H P P P At At At At At At At At At At At At of of of of of of of of of of of of Order Order Part At At At art art art level level level level level level level level level level level level At At At NI—‘NHNI—‘NI—‘NI—‘NH Order Order Order Order Sex Sex Sex Sex Sex Sex Sex Sex Group 1 Simple 2—Way Interactions MS F level 1 of Part 1071.13 88. level 2 of Part 2088.10 161. level 1 of Order 1289.09 101. level 2 of Order 1183.86 90 level 1 of Sex 1139.29 89. level 2 of Sex 1852.05 186. Simple Main Effects MS of Order 369.63 29 of Order .52 0 of Part 38.88 3 of Part 82.30 3 of Sex 33.67 2 of Sex 161.81 12 of Part 59.81 8 of Part 55.38 8 of Sex 96.85 7 of Sex 8.52 0 of Order 307.85 28 of Order 5.75 0 Simple Simple Main Effects MS F And level 1 of Part 702.83 55. And level 2 of Part 1208.73 95 And level 1 of Part 159.18 12. And level 2 of Part 935.25 73. And level 1 of Part 51.16 A And level 2 of Part 1179.36 93 And level 1 of Part 1021.82 80 And level 2 of Part 957.86 75 And level 1 of Order 632.88 89. And level 2 of Order 266.93 21. And level 1 of Order 1300.07 102. And level 2 of Order 721.39 56. 128 Table B-A6 Simple Effects Tests Sex x Order x Parent - Mutual Gaze Percent 58* 65" 78* .25” 92' 18* .17' .08 .07 .38 .66 .78' .69 .37 .61' .67 .30' .85 88' .09' 56* 82* .08 .08' .65* .60’ 95' 07' 61* 98* For Sex For Sex For Order For Order For For For For For For For For Part Part Order Order Part Part Order Order 125 Table B-A7 Simple Effects Tests Sex x Order - Infant Smile Frequency Group 2 MS At level 1 of Order 18.38 At level 2 of Order 119.35 At level 1 of Sex 23.05 At level 2 of Sex 108.38 Table B-88 Simple Effects Tests Group 2 MS At level 1 of Order 96.11 At level 2 of Order 28.03 At level 1 of Part 30.15 At level 1 of Part 108.00 Table B-89 Simple Effects Tests Group 2 MS At level 1 of Order 10.72 At level 2 of Order 30.38 At level 1 of Part 9.72 At level 2 of Part 32.08 F 0.79 5.12 0.99 8.65 Order x Part - Infant Vocalize Frequency F 3.61 0.90 1.13 8.06 Order x Part - Infant Vocalize Percent F 1.90 5.37' 1.72 5.68* For For For For For For For For Sex Sex Order Order Sex Sex Order Order 126 Table B-50 Simple Effects Tests Sex x Order - Infant Smile Mean Group 2 MS At level 1 of Order 1.03 1 At level 2 of Order 0.08 0 At level 1 of Sex 1.19 1 At level 2 of Sex 0.02 0 Table B-51 Simple Effects Tests Sex x Order - Infant Smile Interval Group 2 MS At level 1 of Order 1538.88 0 At level 2 of Order 8139.10 2 At level 1 of Sex 3988.56 2 At level 2 of Sex 1635.88 0 F .11 .05 .28 .02 .78 .11 .03 .83 APPENDIX C Source Sex (S) Group (G) Order (0) S x G S x O G x O S x G x 0 Error Source Sex (S) Group (G) Order (0) S x G S x O G x 0 S x G x 0 Error Source Sex (S) Group (G) Order (0) S x G S x O G x O S x G x 0 Error 127 Table C-I Part A - Sex x Group x Order Mothers and Infants - Frequency df F p< 6 1.03 .86 6 1.96 .16 6 1.37 .31 6 2.70 .08 6 1.83 .19 6 0.55 .77 6 0.75 .63 11 Table C-2 Part A - Sex x Group x Order Mothers and Infants — Percent df F p< 6 0.39 .88 6 2.86 .10 6 0.78 .60 6 0.65 .70 6 1.08 .83 6 0.81 .59 6 0.51 .79 11 Table 0-3 Part A - Sex x Group x Order Mothers and Infants - Mean df F p< 6 0.90 .58 6 0.95 .50 6 0.72 .65 6 0.87 .55 6 0.69 .67 6 0.50 .80 6 0.91 .53 11 128 Table C-A Part A - Sex x Group x Order Mothers and Infants - Interval Source df F p< Sex (S) 6 0.09 1.00 Group (G) 6 0.28 .96 Order (0) 6 1.87 .28 S x G 6 2.16 .13 S x O 6 1.28 .35 G x O 6 0.39 .87 S x G x O 6 0.72 .65 Error 11 Table C-5 Mothers' Responses to Questionnaire Items Question Mean SD 1 16.17 28.27 2 18.67 22.26 3 6.92 6.75 8 32.58 33.86 5 50.06 38.20 6 8.83 8.98 7 55.89 35.96 8 23.75 20.23 9 7.68 12.58 10 8.71 9.31 11 30.73 27.83 12 5.08 3.66 13 23.37 16.98 18 23.83 15.57 15 78.73 17.85 16 87.97 15.28 17# 26.70 12.73 31 60.35 22.02 32 30.10 25.78 Temperament 1.60 1.97 Satisfaction with infant temperament 13.73 16.72 Similarity to mother in temperament 29.83 18.31 Similarity to father in temperament 37.61 27.28 #Father participation l 1 _ 2 82* 3 30 8 l8 5 ~38 6 ~16 7 08 8 ~05 9 ~01 10 83' ll ~28 12 22 13 03 18 Ol 15 26 16 09 17 16 31 09 32 ~18 Temp ~06 Sat ~09 Sim1# ~00 Sim2$ 26 12 13 09 18 22 15 23 16 28 17 13 31 -03 32 -03 Temp ~18 Sat 07 Sim1# ~11 Sim2$ 18 #Similarity $Similarity ~10 -23 21 ~27 17 ~10 ~26 16 10 20 01 ~15 21 O3 22 27 ~81 ~20 ~18 17 ~22 13 60* -22 ~38 ~26 30 ~03 25 28 08 ~05 to mother in temperament to father in temperament 28 ~18 ~05 -13 ~02 73* 89* 80 51* -02 86' 21 15 26 10 05 27 22 27 05 18 ~27 ~87! -55‘ 18 ~07 85' 33 80 ~09 Intercorrelations between 129 Table C~6 Parent Questionnaire Items ~18 26 00 13 35 ~0l 06 12 ~07 09 18 10 ~08 18 ~07 ~25 ~07 02 ~19 15 66’ 81' ~22 12 ~18 15 09 ~09 All Mothers 5 17 25 ~33 ~19 02 33 02 15 ~02 ~21 ~19 -118! ~07 ~20 17 ~01 10 ~26 16 58* ~36 89* ~37 -03 ~38 05 6 17 ~02 08 02 ~17 ~ll 03 ~09 ~18 26 06 -32 68* ~08 09 ~83! -25 17 ~18 37 -36 ~10 -33 ~08 ~19 ~12 ~16 ~18 05 15 23 -38 ~29 -36 ~05 19 07 ~11 05 31 ~88! 06 ~07 85" ~15 ~05 21 08 -23 ~08 ~26 03 l9 17 85* ~08 03 ~27 01 20 32 07 30 -35 ~10 06 35 88* ~18 37 ~08 18 ~08 02 27 22 35 18 02 Temp -55! 88* 20 10 22 83' 38 07 15 17 08 30 ~09 07 ~02 ~08 19 Sat 27 16 ll 80 07 33 —1o -17 -32 30 ~80 09 03 83* ~06 Sim1# ~15 Photograph Look total (PHLKT) Look freq. (PHLKF) Smile total (PHSMT) Smile freq. (PHSMF) Voc total (PHVOCT) Voc freq. (PHVOCF) Mirror Look total (MIRLKT) Look freq. (MIRLKF) Smile total (MIRSMT) Smile freq. (MIRSMF) Voc total (MIRVOCT) Voc freq. (MIRVOCF) Checkerboards (complex minus noncomplex) (CHECKS) Beads (total number of eye and head movements out of 8 possible) (BEADS) Keys (total number of eye and head move~ ments out of A possible) (KEYS) Male Infants Mean 18.91 7. 3.82 1 0.00 O. 0.00 O. 1.02 1 1.08 1 19.13 9. 2.67 1 1.02 1 0.50 O 2.10 3. 1.82 2. 12.13 25. 1.75 1. 0.50 0 Infant Test Means SD 36 .51 00 00 .59 .31 29 .88 .81 .80 60 31 18 18 .67 130 Table C~7 Female Infants Mean 13.01 3.67 1.61 0.33 0.36 0.50 19.78 12.99 2.75 0.67 SD .22 .A3 .88 .89 .71 .00 .77 .78 .58 .23 .88 .38 .20 .66 .23 All Infants Mean 15.96 8. 3.58 1 0.80 3. 0.17 0. 0.69 1 0.79 1. 19.86 8. 3.08 1 3.39 5. 1.29 1 1.70 2. 1.25 1 12.56 18 2.25 1 0.58 0. SD 70 .98 85 38 .25 18 38 .88 30 .83 83 .87 .29 .88 97 —I OOCDNO‘U'I-zWN—t d-A-h-s-b-s mmsz—a PHLKT PHLKF PHSMT PHSMF PHVOCT PHVOCF MIRLKT MIRLKF MIRSMT MIRSMF MIRVOCT MIRVOCF CHECKS BEADS KEYS SEX TEMP —5 4 11 12 13 18 15 16 17 MIRVOCT MIRVOCF CHECKS BEADS KEYS SEX TEMP Mother Gaze freq Gaze % Smile freq SMle$ Voc freq Voc $ 18 ~25 21 ~16 03 08 ~02 ~31 ~10 22 07 ~12 ~10 31 ~35 17 10 ~31 ~30 00 ~18 ~12 88* -38 2 ~18 -01 —29 ~15 03 25 08 ~03 ~26 ~10 07 17 19 06 20 11 88* ~86! ~33 ~08 ~15 38 3 53* ~13 ~16 ~10 13 28 10 ~13 ~15 ~08 27 ~18 28 ~02 12 ~06 .82; ~06 ~09 87* 131 Table C~8 Infant Test Intercorrelations A 5 6 -15 __- ~21 72* ~~~ ~01 ~10 ~05 23 ~27 ~19 80 ~33 ~37 83* ~31 ~35 ~08 30 88* 12 38 60* 08 17 10 23 ~12 -22 ~16 ~30 ~38 85* ~27 ~25 ~25 80 85' 13 18 15 12 ~-- ~18 ~05 ~~- 02 38 09 ~08 ~22 06 Table 0~9 7 8 ~85* ___ 35 ~15 29 06 ~18 ~22 ~05 ~27 27 ~26 ~27 22 08 19 08 23 ~13 ~26 16 17 -39 --- Correlations between Infant and Mother Interactive Behaviors All Infants ~ Part A Mutual gaze freq 72 Mutual gaze 1 Infant Gaze .__Smile___ ._chalizs_ Freq 1 Freq % Freq $ 10 ~08 ~10 12 ~35 ~38 ~12 20 25 25 18 ~01 ~38 85 27 69 02 00 ~11 80 20 23 20 26 ~16 ~03 ~10 27 ~22 ~21 18 19 32 18 25 16 ~13 05 ~23 ~17 ~28 05 93 69 79 50 38 9 60* ~20 ~16 03 02 -13 86* -55! Mutual_saze Freq % ~05 ~11 26 87 ~29 88 10 36 01 06 ~03 20 ~~~ 08 08 ~~~ 132 Table C~10 Correlates of Interactive Behaviors All Infants ~ Part A Behavior Jan—MA Freq 1 Freq 1 Freq % Question 11 ~ 36 ~02 amount infant looks Question 12 ~ 19 ~26 amount infant likes to look Temperament rating ~08 13 ~10 16 05 18 Time since last feeding 23 ~28 ~22 ~26 ~07 03 Time since last nap 31 81* 26 13 70' 68* Length of last nap 10 ~03 ~09 ~02 ~28 ~19 Differences in today's routine 05 28 17 18 28 18 Baby's mood ~28 ~27 ~27 ~22 ~13 00 Baby's age 37 ~08 ~06 ~18 ~17 ~18 Early/late birth ~08 OA 25 12 11 01 Birth weight ~01 21 3O 11 O6 05 Pregnancy and delivery complications ~22 16 ~03 ~02 06 08 Birth medication ~11 80 82* 55* 32 20 Corrected age 15 ~01 18 O3 00 ~06 Baby illnesses ~39 ~01 06 16 ~05 03 Pacifier use ~10 ~08 00 11 18 20 Birth order 16 ~03 ~01 ~08 ~20 ~28 133 Table C-11 Correlations Between Infant Test Behavior and Interactive Behaviors Infant _Qaz_e___5_mils___lcc_l’lutual_gaz_e Freq % Freq % Freq % Freq % PHLKT 28 80 83* 38 81* 27 28 88* PHLKF ~06 ~06 ~02 13 17 15 ~13 ~06 PHSMT 39 ~05 21 ~02 ~15 ~16 83* 07 PHSMF 39 19 71* 18 29 08 81' 35 PHVOCT ~18 ~25 ~23 ~26 ~08 ~01 ~09 ~28 PHVOCF ~22 ~06 ~18 ~18 16 36 ~30 ~18 MIRLKT 00 38 23 33 83* 88* 02 33 MIRLKF 35 ~20 ~01 ~12 ~09 ~12 11 ~16 MIRSMT ~06 ~15 26 01 05 ~03 08 ~05 MIRSMF 21 13 60* 18 26 12 08 22 MIRVOCT 08 ~06 ~06 ~08 ~18 ~06 03 ~09 MIRVOCF ~18 ~03 ~10 00 ~03 09 ~19 ~08 CHECKS ~37 17 01 06 17 20 ~27 11 BEADS 10 ~26 ~18 ~25 ~15 ~17 28 ~20 KEYS 17 18 00 29 ~21 ~18 28 23 —I CCCDNGU’IRWN-fi 8'3 gwwggggggg 'om—sqoxmsz-a SAT SIM1# SIM2$ 12 13 18 15 16 17 31 32 Temp SAT SIM1# SIM2$ 82 83 19 ~28 ~17 -25 -29 07 6o -25 13 ~06 01 30 18 13 36 ~26 ~16 ~08 27 -09 12 19 68' 37 88 ~27 08 ~10 22 12 18 ~28 138 Table C-12 Intercorrelations Between Parent Questionnaire Items Group 1 Mothers 2 3 8 5 6 7 -19 - ~18 39 _ 33 ~16 ~18 — ~15 88 12 27 ~ 01 01 28 25 00 — —03 ~19 ~55 05 ~03 12 ~30 73* 89 ~28 20 17 18 59* 08 07 82 ~10 21 32 81 86 60* 37 10 59* 59* 32 88 22 ~17 20 08 12 76* ~25 ~80 76* 57 ~09 75* ~03 33 28 16 ~10 81 ~83 33 21 ~13 37 80 15 80 ~80 ~02 -37 ~53 ~83 88 ~07 10 ~19 ~39 16 ~56 13 ~16 ~38 26 ~28 ~15 ~23 ~81 ~03 ~16 27 ~23 33 20 -32 39 -08 25 19 80 00 07 ~19 ~50 ~01 ~25 05 05 20 13 18 15 16 17 31 50 ~ 81 29 ~ 10 08 55 - ~22 ~51 87 ~07 ~ ~22 ~09 09 ~26 88 ~ 17 21 36 33 11 ~50 08 ~82 ~06 36 12 ~29 38 36 80 22 11 26 03 11 60' 16 52 ~88 09 ~06 ~50 08 ~53 06 #Similarity to mother in temperament $Similarity to father in temperament ~81 21 ~11 ~56 02 ~39 ~10 -07 09 88 09 32 ~17 03 ~01 32 61 10 32 39 ~09 22 58 -21 62* 09 18 ~22 ~36 30 -20 58 22 07 10 18 18 52 38 28 11 -29 38 ~22 ~18 ~06 19 ~02 11 73* 33 55 21 19 -31 38 ~35 -32 -02 08 ~87 Temp SAT SIM1# 21 ~35 86 33 28 l 1 _ 2 AA 3 ~02 A 16 5 ~11 6 78* 7 16 8 ~39 9 A6 10 37 11 ~27 12 00 13 ~02 18 ~26 15 38 16 ~03 17 ~08 31 ~33 32 ~09 Temp ~36 Sat ~02 Sim1# 09 Sim1$ ~05 12 12 ~ 13 18 18 02 15 ~18 16 ~13 17 A7 31 58 32 03 Temp 18 Sat 31 Sim1# 09 Sim2$ 15 #Similarity $Similarity 87* 60* 52 ~09 21 17 ~08 .38 ~03 15 -68! 82 10 ~20 19 31 18 ~12 21 ~18 13 ~17 -61! ~72 ~08 81 61* 21 -32 ~18 ~02 to mother in temperament to father in temperament 12 16 26 18 20 06 ~01 25 19 73* 08 -52! -76! 13 58 18 ~01 ~86 08 18 ~13 ~02 05 ~09 ~85 3A 59* ~29 ~l6 135 Table C-13 Intercorrelations Between Parent Questionnaire Items All Fathers 8 5 6 7 8 9 10 11 62* - 32 30 ~ 11 15 05 ~ 38 62* ~07 ~07 ~ 20 03 80 16 ~31 - 03 ~02 ~02 58 ~18 70* ~ ~03 ~12 ~18 58* 18 ~05 02 - ~05 32 ~01 29 58 ~26 13 13 20 80 31 20 07 ~13 ~15 ~03 ~38 ~28 ~09 31 02 17 ll 58* 37 05 07 09 ~22 32 26 ~22 16 ~16 ~81 11 ~06 ~09 16 ~06 ~19 12 ~21 ~05 28 18 87 ~83 29 58 05 10 56 ~26 ~33 83 23 32 15 Ol 13 ~70* ~65* ~03 89 38 ~03 38 80 ~28 ~30 81 00 10 18 ~02 33 13 ~05 12 ~12 26 23 ~08 08 ~37 ~83 ~05 ~17 18 26 ~12 28 53 22 15 15 16 17 31 32 Temp Sat Sim1# 798 _ 03 03 ~ ~33 ~86 ~08 - ~30 ~23 ~88 37 ~ 16 22 ~18 85 38 ~ 32 17 35 16 ~38 87 ~ 12 08 ~80 10 37 ~13 ~09 ~ 83 ~23 ~37 12 ~03 ~66* ~13 ~09 REFERENCES REFERENCE NOTE 1. Hildebrandt, K. A., & Fitzgerald, H. E. Mothers' and toddlers‘ responses to infant physical appearance. Unpublished manuscript Michigan State University, 1978. 136 REFERENCES Adamson, L., Als, H., Tronick, E., & Brazelton. T. B. The development of social reciprocity between a sighted infant and her blind parents: a case study. ' 281911121211. 1977. 16. 198 207. Als, H. The newborn communicates. .1gunnal_gf_ggmmunigatign, 1977,.21, 66-73. Anderson, B. J. The emergence of conversational behavior. Journal_gfi Communication. 1977. 21, 85-91. Argyle, M., & Cook, M. .Qagg_an14mutual_gaze. Cambridge: Cambridge University Press, 1976. Bakeman, R., & Brown, J. Behavioral dialogues: An approach to the assessment of mother-infant interaction. .Qnild_neyelgpment, 1977, 8.8. 195-203. Bar-Tal, D. & Saxe, L. Physical attractiveness and its relationship to sex role stereotyping. ,§§x_figle§, 1976, 2, 123 133. Bates, E., Camaioni, L., & Volterra, V. The acquisition of performa- tives prior to speech. MW. 1975, 21. 205-226. Bateson, M. C. Mother-infant exchanges: The epigenesis of conversa~ tional interaction. WW. 1975. .263. 101-113. Beckwith, L., Cohen, S. E., Kopp, C. B., Parmalee, A. H., & Marcy, T. G. Caregiver-infant interaction and early cognitive develOpment in preterm infants. when. 1976. .81. 579—587. Belsky, J. Mother-infant interaction at home and in the laboratory: The effect of context. Paper presented at the biennial meeting of the Society for Research in Child Development, New Orleans, 1977. Berscheid, E., & Walster, E. Physical attractiveness. In L. Berkowitz (Ed.). WW (1101- 7)- New York: Academic Press, 1978. Blehar, M. C., Lieberman, A. F., & Ainsworth, M D. S. Early face-to- face interaction and its relation to later infant-mother attach- ment. Wain-gent. 1977. 18. 182-198. 137 138 Bowlby. J. Wat. New York: Basic Books, 1969. Brazelton, T. B. Sucking in infancy. In Y. Brackbill & G. Thompson (Eda). WM readings. New York: Free Press, 1967. Brazelton, T. B., Koslowski, B., & Main, M. The origins of reciprocity. In M. Lewis & L Rosenblum (Eda). Wits canegiyen. New York: Wiley, 1978. Brazelton, T. B., Tronick, E., Adamson, L., Als, H., & Wise, 3. Early mother-infant reciprocity. In M. A. Hofer (Ed.), Parent;infant interaction, Amsterdam: Elsevier, 1975. Bridger, W. Sensory and autonomic function of the neonate. Journal_gfi the.AQadem1_9f_0hild_Esthiatnx. 1962..l. 67-82. Bronshtein, A. I., & Petrova, E. P. An investigation of the auditory analyser in neonates and young infants. In Y. Brackbill & G. Thompson (Eds. ), WWWK 9£_n§aging§. New York: Free Press, 1967 Bruner, J. S. Pacifier-produced visual buffering in human infants. WM. 1973. 6.. 85-51. Bruner, J. S., & Bruner, B. M. On voluntary action and its hierarchical structure. Wm. 1968. 3. 239-255. Campbell, S. B. Maternal and infant behavior in normal, high risk, and "difficult" infants. Paper presented at the biennial meeting of the Society for Research in Child Development, New Orleans, March, 1977. Carey, W. B. Measurement of infant temperament in pediatric practice. In J. c. Westman (Ed. ), Indifldualjifienencesjuhildm New York: Wiley, 1972. Cohen, L. B. A two process model of infant visual attention. MW. 1973.19. 157-180. Cohen, L. B., & Gelber, E. R. Infant visual memory. In L. B. Cohen & P. Salapatek (Eds.). WW: tion (Vol. 1). New York: Academic Press, 1975. Cohen. L. Bo. 8: Salapatek. P. (Eda). WWW ,tg_gggnitign_LaglglsLL. New York: Academic Press, 1975. Condon, W. S., & Sander, L. W. Speech: interactional participation and language acquisition. Science, 1978, 183. 99-101. Corter, C., Trehub, S., Boukydis, C. Ford, L. Celhoffer, 1., & Minde, K. Nurses' judgments of the attractiveness of premature infants. Unpublished manuscript, University of Toronto, 1976. 139 Costello, A. J. Pre-verbal communication. Jgunnal_g£_§h11d_£§ychglggy WW. 1976. .11. 351-353. Crook, C. K., Burke, P. M., & Kittner, S. Some effects of an extero- ceptive stimulus upon nutritive sucking of neonates. Dexelgpmental .Psxchclcsx. l977..13. 869-872. Dubignon, J., 8 Campbell, D. Sucking in the newborn in three conditions: non—nutritive, nutritive and a feed. Journal_cf WM. 1968. 6.. 335-350. Field, T. M. Effects of early separation, interactive deficits and experimental manipulations on infant-mother face-to-face inter- action. W. 1977. 5.8... 763-771. Field, T. Interaction patterns of primary versus secondary caretaker fathers. Dcxclcpmcntal_fisxchclcsx. 1978..18. 183-188. Fraiberg, S. Blind infants and their mothers: an examination of the sign system. In M. Lewis & L. A. Rosenblum (Eds.), Ihg_gfifect_gf Ithc_infant_cn_its_carcsixcn. New York: Wiley. 1978. Friedman, S., Thompson, M. A., Crawley, S., Criticos, A., Drake, D., Iacobbo, M., Rogers, P. P., & Richardson, L Mutual visual regard during mother-infant play. Perceptual_andJuctcn_Skills. 1976,.12. 827-831. Greenman, G. W. Visual behaviour of newborn infants. In A. Solnit & S. Provence (Eds.). Mcdcrn_cccspccti1cs_in_child_dexclcpment. New York: International Universities Press, 1963. Haynes, H. M., White, B. L., & Held, R. Visual accommodation in human infants. Science. 1965..188. 528-530. Hildebrandt, K. A., & Fitzgerald, H. E. Adults' responses to infants varying in perceived cuteness. .thayigunal_£ngcesses, 1978, 3. Hutt, C., & Ounsted, C. The biological significance of gaze aversion with particular reference to the syndrome of infantile autism. .Bchaxicucal_§cicncc. 1966..ll. 386-356. Jaffe, J., Stern, D. N., & Peery, J. C. "Conversational" coupling of gaze behavior in prelinguistic human development. .Jgunnal_g£ Esxchclinsuistic_ficscanch. 1973..2. 321-330. Kalnins, I. V., & Bruner, J. S. The coordination of visual observation and instrumental behavior in early infancy. Perception, 1973, 2, 307-318. Kaye, H. Infant sucking behavior and its modification. In L. P. Lipsitt & 0. C. Spiker (Eds.). Adxanccs_in_child_dc1clcnmcnt_and pghayign. Vol. 3. New York: Academic Press, 1967. 180 Kessen, W., & Leutzendorff, A. M. The effect of nonnutritive sucking on movement in the human newborn. W- W. 1963. 5.6. 69-72. Kessen, W. Leutzendorff, A. & Stoutsenberger, K. Age, food depriva- tion, non-nutritive sucking and movement in the human newborn. WWW.1967.63. 82 86. Lamb, M. E. Fathers: Forgotten contributors to child development. Wt. 1975. 18. 285-266. Lester, G., Bierbrauser, B., Selfridge, B., & Gomeringer, D. Distract- ability, intensity of reaction, and non-nutritive sucking. BuchclcsicaLRencnts. 1976. 39. 1212-1218. Levin, G. R. & Kaye, H. Work decrement and rest recovery during nonnutritive sucking in the human neonate. Journal_g£_Expenimental W. 1966. .1. 186-158. Lewis, M., & Brooks, J. Self, other, and fear: Infants' reactions to people. In M. Lewis a. L. Rosenblum (Eds.), WM Maximu- New York: Wiley. 1978. Lewis, M., & Lee-Painter, S. An interactional approach to the mother-infant dyad. In M. Lewis & L. A. Rosenblum (Eds.), Inc W New York: Wiley. 1978. Lewis, M., I: Rosenblum, L. A. (863.). W .canegiyen. New York: Wiley, 1978. Maccoby. E. E-. 8: Jacklin. G. N. W. Stanford: Stanford University Press, 1978. Meier, G. W. Behavioral development viewed in terms of conspecific communication. In N. R. Ellis (Ed.),.Achcnant_dc1clcnmcnt_in W. Hillsdale. N.J.= Lawrence Erlbaum, 1975. Mendelson, M. J., & Haith, M. M. The relation between nonnutritive sucking and visual information processing in the human newborn. Child—Denim. 1976. £16. 1025-1029. Moss, H. A. Sex, age, and state as determinants of mother-infant interaction. MW. 1957. .13. 19-36. Moss, H. A., & Robson, K. S. Maternal influences in early social visual behavior. ,Child_fl§1§lgpment, 1968, 39, 801-808. Noll, R. L. The relationship between visual behavior of mothers and infant attentiveness. Dissentatimcstcactuntcnnaticnal. 1971, 3211i). 250- 181 Osofsky, J. D. Neonatal characteristics and mother-infant interaction in two observational situations. §h11d_neyelopment, 1976, 81, 1138-1187. Parke, R. D. Perspectives on father-infant interaction. In J. D. Osofsky (Ed.), flangbggk_gf_1nfangy. New York: Wiley, in press. Parke, R. D., & Sawin, D. B. The family in early infancy: Social inter~ actional and attitudinal analyses. Paper presented at the biennial meeting of the Society for Research in Child Development, New Orleans, March, 1977. Peery, J. C., & Stern, D. N. Mother-infant gazing during play, bottle feeding. and spoon feeding. WM. 1975. 91. 207-213. Peery, J. C., & Stern, D. N. Gaze duration frequency distributions during mother-infant interaction. Jgnnnal_g£_fienetic_£sychglggy, 1976, 123” 85-55. Rebelsky, F., & Hanks, C. Fathers' verbal interaction with infants in the first three months of life. Chilg_flexglgnment, 1971, 82, 63-68. Robson, K. S. The role of eye-to-eye contact in maternal-infant attach- ment. WW. 1967. .8. 13~25. Rovee, C. K., & Levin. G. R. Oral "pacification" and arousal in the human newborn. WM. 1966. .3. l~l7. Rubin, J. 2., Provenzano, F. J., & Luria, Z. The eye of the beholder: Parents' views on sex of newborns. American_lgnznal_gfi W. 1978. 13. 720-731. Siesa1.S. W New York: McGraw-Hill, 1956. Smith, W. J. Message, meaning and context in ethology. American Natunaliat. 1965..99. 805-809. Spelke, E., Zelazo, P., Kagan, J., & Kotelchuck, M. Father interaction and separation protest. .Dclclccmcntal.£sxchclcsx. 1973,.9. 83-90. Sroufe, L. A., & Waters, E. The ontogenesis of smiling and laughter: A perspective on the organization of development in infancy. £§!9h91931931_flfilicfl. 1976..§3. 173-189. Stern, D. N. A micro-analysis of mother-infant interaction: Behavior regulating social contact between a mother and her 3 l/2-month-old twins. W. 1971. 103 501-517. 182 Stern, D. N. The goal and structure of mother-infant play. ,Jggnnal_g£ WWW. 1978.13. 802-— 821 (a) Stern, D. N. Mother and infant at play: The dyadic interaction involving facial, vocal, and gaze behaviors. In M. Lewis & L. A. Rosenblum (Eds. ), W New York. Wiley, 1978. (b) Stern, D. N., Jaffe, J., Beebe, B., & Bennett, S L. Vocalizing in unison and in alternation: Two modes of communication within the mother-inf ant dyad. WW. 1975. 263. 89~100. Thoman, E. B. How a rejecting baby affects mother-infant synchrony. In M. A. Hofer (Ed.), Wham. Amsterdam: Elsevier, 1975. Trevarthan, C. Conversations with a two-month-old. fl§w_fig1§nt1§t, 1978. M87. 230-235. Tronick, E. D., Als, H., & Brazelton, T. B. Mutuality in mother-infant interaction. WM. 1977. 27.. 78~79. Walters, R. H., & Parke, R. D. The role of the distance receptors in the development of social responsiveness. In L. Lipsitt & C. Spiker (Eds. ) . W. New York: Academic Press, 1965. Wolff, P. H. Observations on the early development of smiling. In B. M. Foss (Ed.), WWII. New York: Wiley, 1963. Yogman, M. J. The goals and structure of face-to-face interaction between infants and fathers. Paper presented at the biennial meeting of the Society for Research in Child Development, New Orleans, March, 1977. "Ilffllllflllllflilllliflr