UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Cry and facial behavior during induced pain in neonates Grunau, Ruth Veronica Elizabeth 1985

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1985_A1 G78.pdf [ 5.76MB ]
JSON: 831-1.0076790.json
JSON-LD: 831-1.0076790-ld.json
RDF/XML (Pretty): 831-1.0076790-rdf.xml
RDF/JSON: 831-1.0076790-rdf.json
Turtle: 831-1.0076790-turtle.txt
N-Triples: 831-1.0076790-rdf-ntriples.txt
Original Record: 831-1.0076790-source.json
Full Text

Full Text

CRY AND FACIAL BEHAVIOR DURING ' INDUCED PAIN IN NEONATES BY RUTH V.E. GRUNAU B.A., The University of Sydney, Australia, 1967 M.A., The University of British Columbia, 1969 Ed.D., the University of British Columbia, 1975 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES INTERDISCIPLINARY STUDIES PSYCHOLOGY/PAEDIATRICS/AUDIOLOGY and SPEECH SCIENCES We accept this thesis as conforming to the required-standard. THE UNIVERSITY OF BRITISH COLUMBIA , June 1985 © Ruth V.E. Grunau, 1985 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by h i s or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of I n t e r d i s c i p l i n a r y Studies The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date September 5, 1985 np.-fi i " * / $ m ii ABSTRACT Pain behavior of neonates was compared across sleep/waking states and sex. From Gate-Control Theory (Melzack and Wall, 1982) it was hypothesized that pain behavior would vary depending on the ongoing functional state of the infant, in contrast with Specificity Theory (Mountcastle, 1980), from which one would expect neonatal pain expression to be solely a function of degree of tissue damage. The findings of facial action variation across sleep/waking state was interpreted as consistent with Gate-Control Theory. Awake alert infants responded with the most facial activity, which supported Brazelton's (1973) view that infants in this state are most receptive to environmental stimulation. Fundamental frequency of cry was not related to sleep/waking state. This suggested that findings from the cry literature on pain cry as a reflection of nervous system "stress", in unwell newborns, do not generalize directly to healthy infants under varying degrees of stress as a function of state. Sex differences were apparent in speed of response, with boys showing shorter time to cry and facial action following heel-lance. Issues raised by the study include the importance of using measurement techniques which are independent of pre-conceived categories of affective response, and the surprising degree of responsivity of the neonate to ongoing events. iii Exploratory analyses suggested obstetric factors were related to overall facial action. Caution was expressed in this interpretation due to the great complexity of the inter-relationships of medical, physiological and maternal variables which go far beyond the scope of this study. It was concluded that obstetric features such as mode of delivery should be considered in sample selection for neonatal pain studies, in contrast to current practise which has been to assume healthy newborns form an homogeneous population. It was clear from these findings that the issues are multifaceted, and the optimal way to proceed with research in the area of neonatal pain is with an interdisciplinary team format. iv TABLE OF CONTENTS Page No. ABSTRACT i LIST OF TABLES vi LIST OF FIGURES vii LIST OF APPENDICES viii ACKNOWLEDGEMENTS ix INTRODUCTION 1 LITERATURE REVIEW 5 Pediatric Pain 5 Developmental Changes in Pain Expression 5 Pain Behaviour in Neonates 10 Sex Differences 13 THEORETICAL BACKGROUND 15 PHYSICAL MATURATION AT BIRTH 17 Nervous System 17 Development of Cutaneous Senses 18 Pain Receptors 19 Neonatal Behavioural States 20 PAIN MEASUREMENT IN CHILDREN 25 Overview 25 Vocalization 26 Facial Movement 29 RATIONALE 32 Statement of the Problem 32 HYPOTHESES 3* V TABLE OF CONTENTS ( cont'd ) Page No. METHOD 37 Subjects 37 Apparatus 39 Procedure 39 Measures 40 Acoustic Analysis 40 Video Analysis 41 Perinatal Variables 43 RESULTS 45 Preliminary Analysis 45 Subjects Excluded Due to No Cry 45 Reliability 45 Face Coding 45 Number of Cry Cycles 46 State 46 Rate of Occurrence of Each Facial Action 47 Technician Effects 47 Time Since Feeding 50 Hypothesis Testing : State and Sex 50 Facial Action 50 Heel Rub 60 Heel Lance 60 Latency to Face Movement 61 Cry Measures 61 Cry Cycles, Latency and Duration of First Cry to Heel Lance 61 Fundamental Frequency 68 Pitch-Plot Pattern 68 Facial Actions and Cry Measures 74 Interrelationships Among Facial Actions 74 Interrelationships Among Cry Measures 77 Relations Between Cry Measures and Face Movement 77 vi TABLE OF CONTENTS ( cont'd ) Page No. Exploratory Analyses 80 Relationship of Perinatal and Maternal Variables to Pain Expression 80 Cry Latency 80 Fundamental Frequency 81 Latency of Facial Response to Heel Lance 85 Face Movement 90 Heel Rub 90 Heel Lance 90 Race 97 DISCUSSION 102 State 102 Sex 108 Technician Effects 109 Perinatal and Material Correlates of Neonatal Pain Expression 110 Birthweight and Cry 110 Obstetric Medication 112 Race 117 Conclusions 118 Pain Expression . 121 REFERENCES 122 vii LIST OF TABLES TABLE 1 —-jyrequency and Percent of Each Facial Action Across 48 Segments 1 to 12 2 Mean Time Since Feeding by State 51 3 Mean, (standard deviations) and Duncan post-hoc 52 Comparisons for Facial Movement to Heel Rub and Heel Lance 4 Frequency of Taut Tongue Action to Hee Rub (Segments 1 54 and 2) and Heel Lance (Segments 3 and 4) by State 5 Frequency of Vertical Stretch Mouth to Heel Rub (Segments 1 55 and 2) and Heel Lance (Segments 3 and 4) by State 6 Frequency of Naso-Labial Furrow to Heel Rub (Segments 1 56 and 2) and Heel Lance (Segments 3 and 4) by State 7 Frequency of Eye Squeeze to Heel Rub (Segments 1 and 2) 57 and Heel Lance (Segments 3 and 4) by State 8 Frequency of Brow Action to Heel Rub (Segments 1 and 2) 58 and Heel Lance (Segments 3 and 4) by State 9 Frequency of Lip Part to Heel Rub (Segments 1 and 2) 59 and Heel Lance (Segments 3 and 4) by State 10 Means and Standard Deviations for Latency to 68 Facial Movement by State and Sex 11 Means and Standard Deviations for Cry Variables by 69 State and Sex 12 Means and Standard Deviations of Fundamental Frequency 72 of First Cry to Heel Lance 13 Frequencies and Percentages of Overall Cry Patterns 73 by State 14 Pearson Correlations Between Facial Actions Summed 75 Over Heel Rub and Heel Lance Segments 15 Pearson Correlations Between Total Facial Action 76 to Heel Lance and Latency to Facial Action to Heel Lance viii . LIST OF TABLES (cont'd) Page TABLE No. 16 Pearson Correlations Between Cry Measures 78 17 Pearson Correlations Between Cry Measures and Face 79 Actions to Heel Lance 18 Classification Results of Perinatal Measures to Short 82 ( 4z 499msec. ) and Long ( ^  500msec. ) Cry Latency 19 Log-Linear Analysis of Short and Long Cry Latency (C) 83 by State (S) and Sex (X) 20 Observed Frequency, Percent of Column Totals and Ratio 84 of the Log-Linear Parameter Estimates to Standard Error for Short and Long Cry Latency by State and Sex 21 Classification Results for Low (^599hz) or High 86 (i600hz) f 0 1 22 Mean Birthweight (gm) by Level of f 0 1 87 23 Mean Birthweight (gm) by Level of f 0 1 for 88 Smokers and Non-Smokers 24 Classification Results for Low (^599 hz) and High 89 (^ 600 hz) f Q 1 Excluding Infants of Smoking Mothers 25 Calssification Results for Low and High Facial 91 Movement to Heel Rub 26 Classification Results of Low and High Total Facial 92 Movement to Heel Lance 27 Mean Total Face Movement to Heel Lance 94 28 Mean Total Amount of Face Movement to Heel Lance 95 by Delivery Mode 29 Parity and Mode of Delivery (Frequency, Column Percentage, 98 and Adjusted Residuals) 30 Pearson Correlations Between Obstetric Medication 99 and Perinatal/Infant Variables 31 Method of Feeding by Race 100 32 Incidence of Vaginal and Cesarian Section Delivery by Race 101 LIST OF FIGURES •« "^e&stca^.--FIGURE 1 Percentage Taut Tongue to Heel-Rub (Segments 1 and 2 and Heel-Lance (Segments 3 to 7) by State 2 Percentage Vertical Stretch Mouth to Heel-Rub (Segments 1 and 2 and Heel-Lance (Segments 3 to 7) by State 3 Percentage Naso-Labial Furrow to Heel-Rub (Segments 1 and 2) and Heel-Lance (Segments 3 to 7) by State 4 Percentage Eye Squeeze to Heel-Rub (Segments 1 and 2) and Heel-Lance (Segments 3 to 7) by State 5 Percentage Brow Action to Heel-Rub (Segments 1 and 2) and Heel-Lance (Segments 3 to 7) by State 6 Percentage Lip Part to Heel-Rub (Segments 1 and 2) and Heel-Lance (Segments 3 to 7) by State 7 Mean Fundamental Frequency of the First Cry to Heel-Lance Over Time by State X LIST OF APPENDICES Page No. APPENDIX A Consent Form 138 B Characteristics of Infants Who Did Not Cry to Heel Lance 139 C Face Action Inter-Rater Reliability Per Subject 140 D Summaries of Univariate ANOVA of each Cry Variable 141 by Technician E ANOVA Summary for Time Since Feeding by State 142 F ANOVA Summary for Facial Movement to Heel Rub/Heel Lance 143 by State and Sex G ANOVA Summary for Latency to Facial Movement by State 144 and Sex H ANOVA Summary for Number of Cry Cycles by State and Sex 145 I ANOVA Summary for Latency to Cry by State and Sex 146 J ANOVA Summary for Duration of Cry by State and Sex 147 K ANOVA Summary for Fundamental Frequency of Cry Over 148 Time by State, and Sex L ANOVA Summary for Birthweight by f Q 1 Category and 149 Smoking Status M Frequency ( and Column % ) of Type of Obstetric 150 Medication by Mode of delivery N Frequency ( and Row % ) of Type of Obstetric 151 Medication by State XX ACKNOWLEDGEMENTS I wish to express my gratitude to Dr. Kenneth Craig for his invaluable contribution as committee chairman, to Dr. John Gilbert, Dr. Sydney Segal and Dr. Michaj^ lJSchulzer for their guidance throughout the dissertation research, andto Dr. Barry Munro and Joan Morison, for their very thoughtful advice as committee members. I would also like to extend my appreciation to the following : Dr. Philip Lieberman for providing expertise, and the opportunity to analyze the cry data at the Linguistics Laboratory, Brown University, and to John Mertus and Jack Ryalls for their assistance with the computer work ; Tom Armstrong for videotaping the infants ; Andy Gotowiec for his unflagging enthusiasm through the tedious videocoding and data entry phases of the study and for clerical assistance throughout ; Mary-Jo Wilson for reliability coding ; Dr. Malcolm Greig for assistance with statistical analysis ; Doug Lee for many helpful suggestions ; Brenda Gerhard for moral support ; Dr. Robert MacLean, Dr. H. McMorland, the laboratory technicians and nursing staff, as well as the parents of the infants, for their assistance while at Grace Hospital ; Constance Smith for preparation of the manuscript ; and Dr. Tony LePage who kindly provided the opportunity for educational leave from B.C. Children's Hospital. Most of all, I would like to thank my family : my husband, John Grunau, without whose encouragement this work would not have been possible, and our children Esther (age 16 years), Peter (13 years) and Anna (9 years) for their support. 1 INTRODUCTION A transition from reflexive patterns of response to noxious stimuli in the newborn, to the complex psychophysiological phenomenon of pain experience in adults, is observable during the course of human development. Pain in adults has been studied extensively and is recognized as involving cognitive, affective and sensory components acting synergistically (Bonica and Ventafridda, 1979 ; Melzack and Wall, 1982). In contrast, far less is known about parameters of pain in childhood, although assessment and management of pain are important in pediatric medicine. There are prevalent but unsupported beliefs that children do not experience pain with the same intensity as adults, recover more quickly, and may not remember the pain related events (Eland and Anderson, 1977 ; 3eans, 1983 ; Levine and Gordon, 1982 ; Owens, 1984 ; Varni, Katz and Dash, 1982). A beginning has been made in addressing questions related to measurement of pain in children, developmental changes in pain perception, behavioral and phenomenological aspects of pain experience in healthy and sick children, and special problems such as the hospitalized child and pain management. There is, however, a paucity of systematic studies addressing pain perception and basic pain mechanisms in infancy. While there appears to be consensus that normal neonates respond, often vigorously, to noxious stimuli in that they display vocal and behavioral distress and agitation, there also is disagreement as to the source of the reactions, that is whether "pain" is perceived (Bennett and Bowyer, 0 2 1982). Adults' concepts of pain reflect experience and physical maturation and it would be inappropriate to presume children respond in the same manner. In the infant neurophysiological immaturity must be taken into account. However, Swafford and Allen (1968) have stressed the contradictions inherent in concepts of reduced sensitivity to pain in the first week of life when, for example, considerable restraint is required for alert newborns undergoing simple surgical procedures without general anesthetic. On the other hand, Ritchie (1981) suggests that it is difficult to determine the extent to which the neonates' vigorous response to invasive procedures is due to pain per se, as opposed to dislike of restraint. These difficulties in interpretation, and potential confounds in systematic study, highlight the importance of, distinguishing between observation and inference in pain research. A framework based on social learning theory appears useful for studying developmental changes in pain behavior (Craig, 1978, 1980, 1983). This conceptual model allows for integration of sensory, cognitive, affective, and behavioral events, within specific familial and cultural contexts which may mediate the experience and behavioral responses to noxious stimuli. In this model, reflex patterns of response to noxious stimuli in early infancy would be viewed as the foundation for subsequent socialization influences. In order to examine the ways in which the sensory and behavioral aspects of pain may interrelate in infancy, it may be most productive to delineate these issues within the mainstream theoretical views which have developed in the study of adult pain. Explicit theoretical formulation appears 3 useful at this point ; one needs to look to the major pain theories which have evolved in adult research. The relationship between sensory and other psychological aspects of sensation are conceptualized quite differently depending on the pain theory involved. Evolution of pain theories may be broadly described as falling into two viewpoints. The traditional theory of pain, largely delineated by Von Frey (Boring, 1942) is known as 'specificity theory', and proposes that a specific pain system carries messages from pain receptors in the skin to a pain centre in the brain. This theory implies a direct invariant relationship between stimulation of a given skin receptor and psychological manifestations of pain based on a fixed direct line communication system (Mountcastle, 1980). The 'gate control' theory of pain (Melzack and Wall, 1965, 1982), on the other hand, proposes that there are functional interactions between ascending and descending systems which modulate input before pain is experienced. Implicit in the position of those such as Bennett and Bowyer (1982), who maintain that neonates cannot feel "pain", appears to be a specificty theory assumption that the psychological experience of pain requires a pain centre in the brain, and that due to the immaturity of cortical connections in the neonate, they cannot centrally "interpret" the sensory input as pain. From specificity theory and gate control theory quite different predictions regarding neonatal pain behavior ensue. Based on specificity theory one would expect neonates to respond to invasive stimulation as a function of degree of tissue damage. From gate control theory one would hypothesize that despite equal degree of tissue damage, response variation would occur as a function of the neonate's ongoing behavioral state. In the present study neonates' behavioral responses to noxious invasive stimulation were studied in relation to functional sleep/waking states and sex. 5 LITERATURE REVIEW Pediatric Pain The majority of studies on pain in children are concerned with pain perception and management relating to pediatric medical problems. Among the conditions examined has been recurrent abdominal pain (e.g. Apley, 1975 ; Barr and Feuerstein, 1983), arthritic pain associated with hemophilia (Varni, Gilbert and Dietrick, 1981), chronic pain (McGrath et al., 1985), headaches (e.g. Joffe, Bakal and Kaganov, 1982), burns (Varni, Bessman, Russo and Cataldo, 1980), and cancer (e.g. Katz, Shar, Kellerman, Marston, Hershman and Siegel, 1982). Katz and his associates have observed that -endorphin immuno-reactivity in cerebrospinal fluid of children with acute leukemia undergoing routine lumbar puncture was related to behavioral measures of distress. Varni, Katz and Dash (1982) have reviewed the literature on pediatric pain associated with physical injury, disease states, medical procedures and pain of unknown origin, integrating the few childhood studies with related reseach in adults, and providing a comprehensive overview of this area. Developmental Changes in Pain Expression Ontogenetic changes represent major constraints on the experience and expression of pediatric pain. The area provides a major challenge to the investigator since neurophysiological maturation is occurring concurrently with the accumulation of personal experience, compounded by individual 6 differences relating to constitutional factors and variation in the rate of central nervous system development. While there has been little systematic developmental research on the relationship between age and changes in pain experience and behavior, the belief that children feel less pain than adults has been prevalent (Eland and Anderson, 1977). It has been suggested that in the first week of life neonates are less responsive to pain as part of a system of protection from pain during the birth process (Bondy, 1980). Others (e.g. Bennett and Bowyer, 1982) view reduced sensitivity to pain as lasting to approximately one month or longer, due to immaturity of the central nervous system. In contrast, Volpe and Koenigsberger (1981) note that the premature of only 28 weeks gestation differentiates touch and pain. Haslam (1969) studied pain thresholds in children from 5 to 18 years of age, using pressure to the tibia until the child said it hurt. Haslam found an inverse relationship between age and reported sensation of pain, that is the younger the children the more susceptibility to pain, suggesting young children may perceive pain at lower thresholds than adults. Although this finding certainly calls into question assumptions about robust resistance to pain in young children, one major drawback to this study is that the children were to report when it "hurts". There is every reason to believe that children may be susceptible to impression management, response bias, and situational demand, as has been found with adults (Craig and Prkachin, 1980). 7 Systematic observation of reactions to innoculation in the first and seond years of life have shown changes in behavior (Craig, McMahon, Morison and Zaskow, 1984) and in facial expression alone (Izard et al., 1984). Craig et al. (1983) studied responses of infants aged 2 to 12 months compared with 13 to 24 months of age. A pain behavior rating scale was developed to encode categories of expressive behavior including verbal (e.g. language, crying, fear expressions) and nonverbal (e.g. activity in the face, torso and limbs) responses. Reactions of infants in the younger group were more spontaneous, global, and linked to the actual time of injection. The older group showed more anticipatory distress and voluntary, self-protective movements, plus descriptive language. Other studies of reactions to innoculation have shown little evidence of apprehensive behavior prior to 6 months of age (Levy, 1960). The findings also were consistent with an earlier study of sensory-motor and behavior responses to pinprick in infants and young children (McGraw, 1941). The motor reaction developed from a diffuse unlocalized response with reflexive withdrawal of the body site stimulated, to defensive reactions around 12 months of age. Anticipatory fussing and withdrawal reactions appeared between 7 to 12 months of age, and may indicate emergence of learning and memory processes. The following studies examined response to noxious stimuli other than innoculation, during the first year. Fisichelli, Karelitz, Fisichelli and Cooper (1974) studied pain sensitivity using cry responses to rubber band snap to the sole of the foot. Infants were tested at 5 hours, 2 days and 4 days, and 2, 4, 6, 8, 12, 16, 26, 39 and 52 weeks. This study was particularly interesting as both 8 longitudinal and cross-sectional samples were included, that is, groups of different infants tested at each level. This allowed observation of adaptation effects due to repeated stimulation at successive intervals. The results showed that at 2 days of age all infants responded by crying to the first snap to the sole. Up to and including 8 weeks of age 10% or less of the infants failed to respond with crying to the first stimulus. There was a marked decrease in cry responsivity for both the longitudinal and cross-sectional groups at and beyond 16 weeks of age. At 12 weeks the incidence of no cry response to the first snap was significantly greater in the longitudinal as compared with the cross-sectional group. This appeared to be adaptation to the noxious experience, however it was not a systematic effect as it was not found at 16 weeks or thereafter. While it was a common reaction not to cry in response to the first snap at 16 weeks and beyond ; grimaces and struggling to get free, as well as other motor reactions were observed. There was marked variability between subjects in all measures except response latency. Level of responsivity for each age group appeared to be similar from 2 days to 12 weeks in this study, suggesting no support for the position of Bondy (1980) and Bennett and Bowyer (1982), among others, of decreased sensitivity to pain in the neonatal period. Latency data showed slower response speed at 5 hours of age, but the authors note that at this time most of the infants were either asleep or sleepy. Several possibilities for the slower reactivity at 5 hours were suggested, including fatigue due to effort expended during birth, the transition from placental to pulmonary breathing, type of delivery, and effects of analgesics or anesthetics administered to the mother during delivery. Struggling to get free may be viewed as anticipatory behavior. In the 9 Fisischelli et al. (1974) sample these responses were observed beginning as early at 16 weeks, rather than 6 or 7 months as was found in earlier studies (e.g. Levy, 1960). Rieser, Green, McLaughlin, and Doxsey (1982) observed reactions of infants aged 2, 12, and 52 weeks to a heel stab procedure used in drawing blood during routine healthy baby visits to a pediatric clinic. Both the 2 week and 52 week old groups cried during more than 85% of the observation periods, whereas 12-week-old infants quieted the fastest during a 30 second observation phase following the heel stab procedure. Behavioural responses were diffuse activity at 2 weeks, with little increase in directed behaviors at 12 weeks. At 52 weeks all the babies repeatedly looked at their mothers, reached toward their mothers and grasped objects presented as distractors. It is interesting that in both the Fisichelli et al. (1974) and Rieser et al. (1982) studies many infants around 3 months of age showed less intense crying response to a noxious stimulus than either younger or older babies. It is conceivable that this may be indicative of a transition from more reflexive reaction to pain in the younger infants, to some more centrally mediated components of responding in the older groups. Thoden and Koivisto (1980) reviewed studies of infants' pain cry, and included summaries of the latency, duration and fundamental frequency data of the studies reported. They mention the difficulties inherent in comparing results across studies due to differences in definitions of the cry features and heterogeneity in the groups of infants included. Their own study examined 10 pain cries elicited by a pinch to the arm in 38 infants who met strict criteria for normality, and were seen longitudinally at 1 day, 5 days, 3 months and 6 months. The criteria for inclusion were : full term gestation, normal delivery and perinatal period (mean Apgar score 9.7), appropriate weight and leng~th for gestational age, normal development up to one year of age, repeatedly normal EEG and normal chromosomal karyotype. No statistically significant differences in latency, duration or maximum pitch were found in comparisons over age. The large within groups variability in all the studies of infant cry is noteworthy. There is evidence that newborns are sensitive to pain from birth onwards, although the central question of degree of responsivity remains unanswered. Barr (1983) suggests that the sensitivity for the stimulus may well be intact, and that changes over the course of development during infancy and early childhood demonstrate increasing efficiency in modulating reactions to noxious stimuli. Furthermore, he concludes that lack of anticipatory responses prior to about the middle of the first year does not necessarily imply the stimulus was not earlier felt and remembered. Pain Behaviour in Neonates Rich, Marshall and Volpe (1974) documented responses to a series of fine pinpricks of the lower limbs in 130 healthy newborns. They described the normal responses as movement of the upper and lower limbs, usually accompanied by facial grimace and/or cry. Tactile sensitivity has been measured in other ways : using an "aesthesiometer" which is a series of nylon 11 filaments graded in diameter (Bell, Weller and Waldrop, 1971), a jet of air to the infant's abdomen (Bell and Costello, 1964), and a cold disc (Birns, 1965). Acredolo and Hake (1982) discussed studies relating to tactile and pain responses in newborns, as part of a review of the area of infant perception. They concluded that the potential for pain responsivity has been demonstrated in neonates in studies of heart rate and motor changes to tactile stimulation (e.g. Lamper and Eisdorfer, 1971 ; Wolff, 1967 ; Yang and Douthill, 1974), and in reports of reactions following circumcision (Weiss, 1968). Changes in sleep patterns have been found following circumcision (Anders and Chalemian, 1974 ; Emde, Harmon, Metcalf and Wagonfeld, 1971 ; Gunnar et al., 1984). Studies of the behavioral and physiological effects of circumcision were recently reviewed by Dixon et al. (1984), with the conclusion drawn that this is a stressful and painful event. Marshall, Stratton, Moore and Boxerman (1980) found behavioral changes persisted when babies were evaluated four hours following circumcision. As in the cry studies, diversity of individual differences in response to circumcision was noted. Marshall et al. (1980) suggested that the variation in response to stressful events may be indicative of different coping styles evident from birth. A study by Lipsitt and Levy (1959) of electrodermal pain thresholds in neonates showed a decrease in response threshold over the first four days of life. The mean threshold decreased from 85 to 90 volts on day 1 to 70 volts on 12 day 4, using a repeated measures design. It is interesting that this study has been widely cited as evidence of reduced pain sensitivity in newborns whereas the authors themselves suggested their findings might reflect a slow recovery from effects of maternal anesthesia during the first 4 days of life, or possibly varying degrees of effects of anoxia. Gullickson and Crowell (1964) examined electrodermal thresholds in full term healthy neonates, reporting increased threshold (habituation effects) with repeated stimulation over 3 days. There have been two studies of neonatal reaction to heel-lance. Owens and Todt (1984) found heart rate and percentage of crying time increased following tactile stimulation and further increased following heel-lance relative to baseline recording. They noted wide baseline variability, in the two measures. McKeel and Saunders (1984) also examined heart rate responses to heel-lance. However, in this study, heart rate changes due to tactile response was confounded due to use of an audible voice signal, in the light of the fact that heart rate changes to auditory stimulation is well documented (Berg and Berg, 1979) There are difficulties in evaluating the findings of the neonatal studies due to the lack of attention to potentially important variables such as the type and amount of anesthetic adminstered to the mother during delivery, and the "state" of the infant at the time the noxious stimulus was applied, that is whether the baby was asleep, awake-alert, or in some other state at the time the experimental observation as carried out. These parameters have been 13 shown to be important in evaluating newborn reponses and need to be addressed before questions concerning reduced pain sensitivity in the first few days of life can be resolved. Sex Differences In a comprehensive review of the psychology of sex differences up to 1974 (Maccoby and 3acklin, 1974), cautions are expressed in drawing conclusions from the existing studies comparing male and female infants due to the number of reports which did not give their selection criteria in detail. Maccoby and 3acklin (1974) concluded that sex differences reported in measures in the neonatal period were likely to depend upon the prevalence of complications at delivery in each sample, as unscreened samples will have a higher proportion of males with problems. Findings on sex differences in the newborn period tend to be ambiguous. Studies of sleep-wake cycles (Ashton, 1971) frequency of startles, hand-mouth contacting, and responses to caretaker techniques among newborns have yielded inconsistent results regarding sex differences (Korner, 1983 ; Korner and Thoman, 1970, 1972). Of eight studies concerned with gender related neonatal tactile sensitivity, four reported sex differences, with girls found to be more sensitive than boys in each case. Females showed more movement to removal of a blanket, in response to an air jet to the abdomen (Bell and Costello, 1964 ; and Bell, Weller and Waldrop, 1971), and skin conductance (Lipsitt and Levy, 1959 ; Weller and Bell, 1965). Bell and his associates have taken perinatal 14 complications into account, whereas the Lipsitt and Levy study which has been widely cited as showing decreasing pain thresholds over the first few days of life, and greater sensitivity of girls, gave no description of their subject sample. Gullickson and Crowell (1964) found full term clincially healthy neonates showed no sex differences in response to electrotactuai stimulation. Also, no sex differences were apparent to heel-lance (Owens and Todt, 1984). There is another important factor which has not been considered in any of these studies, namely the circumcision status of the boys. It has been suggested that gender differences may reflect behavioral changes in boys following circumcision, rather than true sex differences in the response variable under study. That is, physical insult due to circumcision may be differentially affecting males, with behavioral and physiological consequences which have been overlooked by infant researchers (Richards, Bernal and Brackbill, 1976). More recently Brackbill and Schroder (1980) concluded that there is little evidence for neonatal sex differences in behavior regardless of circumcision status. To determine whether cultural sex-role sterotypes of females as more responsive and/or reactive to pain have any biological bases it is necessary to compare responses in early infancy before the overlay of socialization begins to take place. The question of neonatal sex differences in pain related behavior remains open. 15 THEORETICAL BACKGROUND Historically, to the ancient Greeks, pain was viewed as the opposite of pleasure ; Aristotle characterized pain as an affective feeling state. The sensory model emerged beginning with Descartes in the seventeenth century. He conceived of a pain system as a straight through channel from the skin to the brain. By the end of the nineteenth century the sensory model of pain had come to be placed in juxtaposition to the older emotion theory. There were others, such as Sherrington (1900) and Titchener (1909 - 1910) who viewed pain as comprised of both sensory and affective dimensions. With the development of basic medical science during the 19th and 20th centuries, the sensory-specificity model of pain moved to the forefront. Von Frey postulated that a single type of receptor was specific to each of four skin modalities (Von Frey 1895 : see Boring, 1942). The basis of specificity theory is a presumed specific pain system which carries messages from pain receptors, through the pain pathway in the spinal cord, to a pain centre in the brain. In this theory pain is equated to tissue damage, the sensory component is viewed as fundamental, and the cognitive and affective aspects treated as secondary reactions to the pain sensation. Numerous clinical findings have defied explanation in terms of a rigid straight through specific pain system (Melzack and Wall, 1982). Examples of these include : (a) phantom limb pain in amputees ; (b) severe pain in paraplegics in specific areas of the trunk or limbs below the level of a known spinal transection ; and (c) the lack of success in certain types of pain 16 syndromes of attempts to provide permanent relief through surgical lesions of the peripheral and central nervous system. Due to clincial evidence which appeared incompatible with specificity theory, gate control theory was fomulated (Melzack and Wall, 1965). There is no disagreement about physiological specialization ; it is a fact that skin receptors have specialized properties. The inadequacy of specificity theory lies in the assumption that experience of pain has a one to one match with tissue damage. It is the relationship between sensory and psychological aspects of sensation which are conceptualized quite differently depending on the pain theory involved. Specificity theory proposes a direct invariant relationship between sensation of pain and stimulation of a given skin receptor, based on a fixed direct-line communication system. Current statements of specificity theory are found in Mountcastle (1980) and Guyton (1982). Gate control theory (Melzack and Wall, 1982), on the other hand, proposes that there are functional interactions between ascending and descending systems which modulate input before pain is experienced. 17 PHYSICAL MATURATION AT BIRTH Nervous System There have been suggestions of reduced sensitivity to pain in the neonatal period (Bennett and Bowyer, 1982 ; Bondy, 1980 ; Mersky, 1971), and controversy over the legitimacy of this assertion (Barr, 1983). Neurophysiological immaturity is cited as the reason for limited pain perception. Behavioural perspectives on neonatal response to noxious stimuli reflect beliefs about the maturational status of the central nervous system at this age. Until fairly recently, neural functions of newborns were considered to be limited to reflexive reactions. According to this view, only with later development of the cortical interconnections do these reflexes become incorporated into more complex mechanisms. This view was questioned when healthy infants were studied in their own right and in more detail. Systematic observation of normal newborns showed complexity and organization of behavior which had been previously overlooked. Current conceptualizations (e.g. Emde and Robinson, 1979 ; Prechtl, 1981) stress qualitative transformations in properties of the young developing nervous system, as opposed to the static perspective of the infant brain as a "brainstem preparation". Newborns show evidence of very complex control systems regulating vital functions such as breathing, rooting, sucking and swallowing, crying, spatial orientation, sleeping and waking, which are incompatible with the notion of simple reflexive responses. 18 Prechtl (1981) maintains that systems analysis of the nervous system has refuted the concept of a simple progressive adding of higher structures to the more primitive lower brainstem and spinal cord structures. When infants are no longer viewed as deficient adults, the emphasis shifts from the apparently limited abilities of the neonate to their amazing competence to cope with a wide variety of specific demands of internal regulation, and adaptation and responsiveness to features of the environment (Brazelton, 1983). This approach has been conceptualized within an evolutionary framework (e.g. Anokhin, 1964). Biobehavioral perspectives on brain maturation are consistent with expectations of continuing transformations in pain experience and expression subsequent to birth. Using myelination patterns of sensory afferents as an indication of relative development of function, subcortically mediated sensory systems appear most advanced at birth, with neocortically mediated systems showing the greatest degree of postnatal change (Bronson, 1982 ; Trevartlen, 1978 ; Reinis and Goldman, 1980). Multiple feedback networks integrating activity between subcortical and cortical components of the nervous system are the least advanced at birth, and show the most protracted development. Development of Cutaneous Senses Reinis and Goldman (1980) summarized the embryological develop-ment of the cutaneous senses and their receptors as follows : 19 The sensory nerves approach the skin of the fetus in the eighth prenatal week, and are in contact with it in the ninth week. The basal cells of the skin epithelium are altered after contact with the nerve, although no receptors have been described at this early foetal stage. Sensitivity to pressure or contact develops in a cephalocaudal direction from the lips downward : end of seventh prenatal week lip tactile reflex responses may be evoked 10.5 prenatal weeks palms of the hands are responsive to light stroking with a hair 11 prenatal weeks face and all parts of the upper and lower extremities are sensitive to touch 13.5 to 14 prenatal weeks entire body surface (except for back and top of head) sensitive to pain. Pain Receptors The traditional view of morphologically specific receptors beneath each sensory spot on the skin assigned to each of the four modalities of touch, heat, cold and pain has not been supported by physiological studies (for a review see Melzack and Wall, 1982). .However, the assumption that skin receptors have specialized physiological properties remains valid, but not in the modality-specific sense. Rather, receptors may generate different temporal patterns of nerve impulses as a function of threshold to mechanical distortion, threshold to negative and positive temperature change, threshold to chemical change, rate of adaptation to stimulation, size of receptor field, 20 duration of after discharge, and possible interrelationships amongst these variables. Sensory fibers may be divided into three groups : A-beta or large myelinated, A-delta or small myelinated, and C or unmyelinated fibers. Approximately 60 to 70% of all sensory afferents are in the C group. The A-beta fibers appear to be very responsive to light pressure, whereas fibers in the A-delta and C groups may respond to only one or to more than one type of stimuli, for example there are C "polymodal fibers" that respond to pressure, heat and chemicals. However, any nerve fiber hit hard enough and fast enough will give off an injury discharge. The neurobiological study of pre- and post-natal development of pain pathways, including morphology and receptor pharmacology is very recent and confined to subprimate species (Fitzgerald et al., 1984). Neonatal Behavioural States The state concept refers to hypothetical processes within the organism which mediate its external and internal interactions with the environment. These states are assumed to reflect fundamental processes in the central nervous system which can be studied by behavioral and/or elec-trophysiological procedures (Prechtl and O'Brien, 1982). 21 The concept of relatively stable "states" and their possible implications regarding physiological and behavioral responsivity in neonates were derived from naturalistic observations carried out by Wolff (1959, 1966). This has been viewed as a breakthrough in infant research (Emde and Robinson, 1979 ; Prechtl and O'Brien, 1982). Until then, due to the wide range of unexplained inter- and intra-individual variability in neonatal responses, the view was held that inconsistency and moment to moment variability was an inherent characteristic of newborn functioning. In the neonate, state refers to constellations of functional patterns of physiological and behavioral variables which repeat, and appear to be relatively stable (Hutt, Lenard and Prechtl, 1969). Five states form the basis of the system used by Prechtl (1974). Simply stated in operationally defined behavioral terms they are : 1. eyes closed, no movement (except occasional startles) 2. eyes closed, movement 3. eyes open, no movement 4. eyes open, movement 5. crying. A sixth state between sleep and waking has been suggested (Brazelton, 1973) ; Prechtl views this as a transition between stages, rather than an additional functional pattern. 22 Prechtl and his associates at Groningen (Prechtl, 1974) and Korner and her associates at Stanford (Korner, 1972) have strongly rejected the view of sleep/awake states as a continuum of "arousal". Rather, neonatal states are viewed as reflecting distinct and qualitatively different modes of physiological activity (Prechtl, 1974). It is the pattern of physiological concomitants taken simultaneously which relates to state, not any single response measure. Thus it is argued that states do not reflect an underlying continuum, but rather a nominal scale. The view has been expressed that in neonates the rest-activity cycle which occurs within both sleeping and waking states is a more important key to ongoing changes than whether the infant appears to be asleep or awake (Sternman, 1972 ; Kleitman, 1973). Use of interpretative terminology for state descriptions such as "level of arousal", "vigilance", "level of consciousness", "deep and light sleep", has been viewed as a serious difficulty which is counterproductive for neonatal research (Robinson, 1969 ; Ashton, 1973 ; Prechtl, 1974). This is consistent with the strong inderdisciplinary emphasis current in infant research which asserts that propositions about behavioral regulation which are incompatible with what is known about neurophysiological regulation are not worthwhile (Emde and Robinson, 1979). The importance of state as a mediator of environmental input and response is emphasized by Brazelton in his 1973 monograph : 23 "Since an infant's reactions will be state-related it is vital that observations on this 'state' should be considered as a starting point from which all other observations are made. An infant's use of states as a framework for his reactions to the examiner may be most important as a part of the observation." (p. 13) Prechtl (1974) documented studies of reflex responses in relation to state 1, 2 or 3. Proprioceptive reflexes were maximal in state 1, whereas tactile and pressure responses appeared minimal in state 1. Nociceptive reflexes occurred in all three states, but with less intensity of response in state 1. Auditory response was minimal in state 1 and maximal in state 3 ; visual pursuit was, of course, absent in states 1 and 2 ; vestibular response was absent in state 1 but maximal in state 3. Apart from naturalistic studies of spontaneous behavior (Korner, 1969) the earlier work on state dependent changes in responsivity (reviewed by Graham and Jackson, 1970) tended to evaluate state at a gross level, simply differentiating sleep from wakefulness. More recently, studies mainly have examined the auditory modality in relation to state, as visual responsivity other than in state 3 is precluded. Studies of behavioral features of tactile responsivity have been sparse, and have generally been addressed to differential effects during sleep states 1 and 2 only (Schmidt and Birns, 1971 ; Dittrichova and Paul, 1974 ; Rose, Schmidt and Bridger, 1978). Rose, Schmidt and Bridger (1978) concluded it still remains unclear to what extent sleep states play a role in infants' behavioral responsivity, as results between studies have often been contradictory. State itself probably does not exert a uniform effect, but would depend not only on stimulus modality but also on stimulus features such as intensity. 24 State is a "low-level" concept which appears to be readily open to operational definition, has theoretical implications for the organization and sequencing of behavior and potentially allows 3 for prediction (Emde and Robinson, 1982). However, Stratton (1982) has emphasized that with our current state of knowledge, concentration on possible underlying mechanisms would be misplaced and highly speculative. In his view, at present systematic observation of functioning of the different rhythmic processes in the neonate is needed. State has frequently been viewed as a variable to be controlled in infant research, rather than studied (Korner, 1969). This has certainly been the case in the literature on stimulus-induced cry. There apprear to be no investigations of variation in cry features as a function of pre-stimulus state, or of responses to invasive procedures such as the heel-lance across state. 25 PAIN MEASUREMENT IN CHILDREN Overview Measurement issues in the study of pain in preverbal children are of central importance. Older children and adolescents may have facility in selecting verbal descriptors to provide observers with information about subjective experiences (Savedra, Gibbons, Tesler, Ward and Wagner, 1982). In contrast, young children do not have the facility with language, either conceptually or expressively, to describe their subjective experiences (Craig, 1980 ; Craig and Prkachin, 1985). Access to the phenomenological component of pain experience is therefore precluded in infants and very young children. There is no doubt that the lack of an adequate metric for pain has contributed to controversy regarding pain experience in infancy (Craig, 1980). Response to a pinprick is used in evaluating the integrity of the newborn's central nervous system (Volpe 6c Koenigsberger, 1981) and behavioral response patterns (Brazelton, 1981), but there has been limited precise clinical documentation of the range and characteristics of normal reponses to pinprick (and other noxious stimuli) in the neonatal period. Autonomic nervous system activity such as electrodermal and heart activity changes are associated with pain experience in adults, and may be useful with children, although they may be dissociated from adult verbal reports (Craig 6c Prkachin, 1978). Barr (1983) has observed that objective indices of physiologic distress such as heart rate, blood pressure, 26 electrodermal response and blood volume are potentially useful in highly controlled studies as indicative of biological mechanisms mediating pain episodes. Crowell, Davis, Chun and Spellacy (1965) showed that foot and leg electrodermal activity in neonates is responsive to a wide variety of sensory stimulation, including tactile (puff of nitrogen to abdomen) stimulation. Other autonomic responses to sensory stimuli have been observed in neonates (Harpin & Rutter, 1982 ; Lipton, Steinschneider & Richmond, 1961). A combination of physiological and behavioral measures has been suggested for evaluation of supplemental regional blocks for circumcision (Johnson, McGrath and Schillinger, 1983). Detailed analysis of cry features has been proposed as one possible parameter with preverbal children, as it has shown promise with infrahuman subjects (Leyine and Gordon, 1982). V o c a l i z a t i o n Infant cry as social communication of distress has received a great deal of attention, particularly its effects on caregivers (Murray, 1979). Acoustical properties of cry signals have been studied using sound spectography and computer analysis (Wolff, 1969 ; Lester and Zeskind, 1982). Ostwald (1972) described cries as intense penetrating noises which compress energy into a highly sensitive region of the auditory spectrum producing sounds ideally designed to alert a caregiver to arouse interest and hold attention. The adaptive value of the cry signal for survival, given the helplessness of infants, is clear. 27 There have been attempts to classify cries according to the type of distress they reflect, for example, to distinguish cries associated with pain, anger and hunger (Wazz-Hockert et al., 1968). Early studies suggested types of cries could be distinguished either acoustically or by human ear, with experienced caregivers more accurate than inexperienced (Wazz-Hockert, 1964). However, a study which controlled for the duration of each cry sample found mothers could not discriminate cry types (Muller et al., 1974). The consensus of current cry research based on spectographic and listener analyses is that acoustical properties do not discriminate between pain cries and cries for other reasons such as hunger (Murray, 1979 ; Lester and Zeskind, 1982 ; Owens, 1984). Rather, the listener uses length of cry and contextual cues to judge. Murray (1979) concluded that rather than unique differences of cry types based on what caused them, cries differ along a continuum of intensity according to the amount of discomfort experienced by the infant. She has viewed cries in the sociobiological context of a graded signal. Graded signals cannot be easily characterized by dissection into specific messages, and their meaning depends on the context of the event and upon the motivational circumstances of the listener. The majority of studies of acoustic properties of cry have been concerned with features which distinguish healthy neonates from sick infants. Sound spectrographic and computer studies of pain-induced crying in neonates, have shown promise of yielding signs of central nervous system stress in "at risk" infants with from a wide range of clinical problems (e.g., perinatal com-28 plications, differential fetal growth, sudden infant death syndrome, asphyxia, malnutrition). Reviews of this literature (Murry & Murry, 1980 ; Lester & Zeskind, 1982) have expressed concern about significant methodological problems leading to lack of comparability across studies due to differing definitions of cry features, methods used to extract these features, plus the heterogeneity of the infant samples studied and the lack of normative studies of neonatal cry. Recommendations have been made for standarization of procedures in measurement and analysis of cry characteristics (Lester 6c Zeskind, 1982 ; Sirvio & Michelsson, 1976 ; Thoden & Koivisto, 1980). Lester and Zeskind (1982) have emphasized that : ". . . there are virtually no normative studies of age or any other changes or sex differences. Variables that affect the initial state of the infant . . . are often not reported and may result in differential cry responses." (p. 150) Within this body of literature there appears to be consensus that cry originates in the central nervous system, and is indicative of the capacity of the nervous system to be activated, and to inhibit or modulate that activation (Parmelee, 1962 ; Golub and Corwin, 1979 ; Lester and Zeskind, 1982). Fundamental frequency is viewed as that aspect of cry which reflects compensatory and recovery capacities following stimulation. This is consistent with the expense-energy model in which maintenance of the homeostatic adaptation is viewed as the most crucial aspect for survival of the neonate. 29 Facial Movement The study of facial expression as nonverbal communication goes back to Darwin (1872). He postulated that facial expressions of human emotion are innate, genetically determined patterns of movement which are independent of cultural experience. This has been supported by subsequent research which has found certain emotions have the same facial expressions in widely differing cultures (Izard, 1977). It is only recently that reliable methods of empirically based study of facial expression have been developed. A reliable way to objectively describe facial expression is in terms of the facial muscles which produce it. The musculature underlying facial movements of the skin is well established (Darwin, 1872 ; Ekman & Friesen, 1978 ; Rinn, 1984). In humans, by the eighth week of prenatal development the muscles and their nerve tracts arrived at their final positions (Crelin, 1981). The muscles for mastication are on the side of the head ; muscles of expression around the mouth, nose and eyes. It is possible to divide the face into three regions based on the fact they they are largely motorically independent of each other : (a) the brows and forehead ; (b) the eyes, lids and root of the nose ; (c) the lower face, including the cheeks, mouth and lower nose and chin. 30 The lower and upper face differ in degree of fine motor control. The oral region is controlled by many small muscles and can be moved in all directions, whereas the brows are manipulated by fewer muscles and can move only up or down or be drawn together. Based on lesion studies with humans, there appears to be a duality in communication and its underlying mechanisms (Rinn, 1984 ; Murray, 1979). In humans, the facial nerve primarily arranges facial features in configurations specialized for communication. Subtle manipulation of these muscles is involved in both speech and non-verbal facial expression. Volitional face movements derive from impulses emanating from the cortex, whereas involuntary movement involves mostly subcortical systems. This voluntary/in voluntary distinction is well established ; the evidence from clincial neurology has been reviewed by Rinn (1984). Emphasis on empirically derived description rather than on inferences about meaning of expression or underlying emotion has resulted in development of objective measures with which to investigate similarities and differences in facial movement and vocalization regardless of inferences made about these responses. 31 Those who have maintained that neonates are incapable of pain experience have frequently based their arguments on the limited cortical connections evident at birth. However, as it appears that as neither face movement nor vocalization depend on elaborate cortical development, the study of these communication systems during noxious stimulation allows for the possibility of pain expression in neonates. Two reliable coding systems have been developed for studying facial expression. Izard (1978) has developed templates which are facial configurations associated with inferred emotional states, called the Maximally Discriminative Facial Movement Coding System (MAX). Another way to objectively describe facial expression is in terms of the particular facial muscles which underly skin movement (Ekman <5c Friesen, 1978), allowing for a detailed account of discrete movements. The Facial Action Coding System (FACS) was developed using action units which are anatomically based discrete movements. The reduction of facial expression to a discrete number of AU's has the advantage of providing a highly objective way of describing facial movement without recourse to preconceived categories of emotion. Although the MAX system is based on muscle movement as well, all possible facial movements are not considered, only those that discriminate between different emotions. Izard et al., (1979, 1982, 1983) have used the MAX system for study of infant emotion. Oster (Oster, 1978 ; Ekman and Oster, 1979) has begun an extension of the FACS system to infants (Baby FACS). 32 RATIONALE S t a t e m e n t o f t h e P r o b l e m The lack of systematic research into children's pain has serious implications for children since supposition rather than fact dictates assessment and treatment (Eland and Anderson, 1977 ; Jeans, 1983 ; Kauffman, 1980 ; Owens, 1984). Kauffman (1980) considers it a myth that children tolerate pain better than adults, and attributes this idea partly to the difficulty of assessing pain in infants and young children, and partly to the reluctance to administer potentially toxic drugs to this age group. A better understanding of children's pain would greatly improve decision making. While a social learning theoretical framework for pain study views developmental changes and the integration of cognitive, affective, sensory, and behavioral events within familial and cultural contexts (eg. Craig, 1983), study of the social influences is constrained by poorly understood biological predispositions. Examination of the earliest patterns of response should yield descriptions of the foundations for subsequent maturational and social influences. In contrast to characterizations of neonates as relatively unresponsive to pain (Bennett and Bowyer, 1982), empirical study of neonates has indicated that the pattern of response to noxious stimuli is most often vigorous and dramatic and reflects changes adults would interpret as pain in response to tissue damage (Owens and Todt, 1984). Apparent decreased sensitivity may be an artifactual consequence of the analgesia or anesthetics delivered to 33 the mother during delivery, fatigue due to effort expended during birth, the transition from placental to pulmonary breathing, or type of delivery. Monitoring these conditions and studying their covariation with individual differences in expressive behavior would contribute information useful in resolving the controversial understanding of neonatal pain perception. Accurate assessment of pain in infants and preverbal children is a significant problem in pediatrics, particularly for evaluating adequacy of analgesics. An evaluation of parameters of pain-induced vocalization and facial expression may contribute towards development of measurement techniques for this age group. The importance of infant sleep/awake states is recognized in measuring neonatal responses (Prechtl, 1974) ; yet there appear to be virtually no studies comparing responses of normal infants across states to invasive procedures. Similarly there are little data available on sex differences to induced pain. The aim of this study was to examine facial expression and vocalization to discomfort (heel-rub) and pain (heel-lance) as a function of neonatal sleep/wake state and sex. Perinatal variables which may affect pain related behavior in the first few days of life, such as type of delivery, and maternal obstetric medication, as well as concomitant factors such as time from feeding to the heel-lance event, which have not received any attention in the small literature on neonatal pain behavior, were considered in secondary exploratory analyses. 34 HYPOTHESES Adult pain theories may be broadly conceived as two positions : specificity theory which proposes experience of pain is a direct reflection of tissue damage (Mountcastle, 1980) and gate control theory which proposes that ascending and descending systems and their functional interactions modulate input prior to pain experience (Melzack and Wall, 1982). It was assumed that the heel-lance would provide comparable tissue damage for all groups varying in behavioral state. Hypothesis 1 : Gate Control Theory : Pain is modulated by the ongoing environmental circumstances and context of tissue damage. Behavioural response to heel-lance will differ by state. Specificity Theory : Pain is a function of amount of tissue damage. No differences will be found between groups across state in behavioral responses to heel-lance. 35 State organization is currently viewed as an indication of the infant's neurobehavioral status, reflecting the infant's ongoing internal organization and capacity for integrating environmental input. The following hypotheses were conceptualized for state differences as mediators of infant responsivity to Invasive stimulation : Hypothesis 2 : Sleep versus Wake Response patterns included by noxious events will differ depending on whether the infant is asleep or awake. 2.1 Variation will appear between infants in states 1 (quiet/sleep) and 2 (active/sleep) combined, and states 3 (quiet/awake) and k (active/awake) combined. Basic Rest - Activity Cycle The rest-activity cycle which occurs within both sleeping and waking states is a more important key to ongoing changes than whether the infant appears to be asleep or awake. 2.2 Responses will be more similar for states 1 (quiet/sleep) and 3 (quiet/awake) as compared with states 2 (active/sleep) and 4 (active/awake). 36 Alert - Awake State 3 (quiet/awake) has been viewed as reflecting optimal orientation to environmental input (Brazelton, 1973). 2.3 Response patterns in state 3 (quiet/awake) will differ from states 1 (quiet/sleep), 2 (active/sleep) and k (active/awake). Fundamental frequency of the first cry following pain is currently conceputalized as indicative of central nervous system capacity for response modulation (Golub, 1979 ; Lester and Zeskind, 1982). Hypothesis 3 : Higher fundamental frequency will be evident in those states which evoke the most facial reaction. 37 METHOD Subjects Data were collected during April and May 1984, on a continuous series of 174 infants from the well baby unit of Grace Hospital, a major metropolitan maternity hospital. Mean time from delivery to heel-lance was 43.05 hrs. (S.D. 7.06 hrs.). As the aim of this study included examining responses in relation to perinatal events in healthy babies, it was important that the investigator remain "blind" to the health status of the infant at the time of testing and recording. Thus, infants who did not meet criteria for inclusion were excluded after the data collection phase. The criteria for inclusion were : birthweight above 2500 gm - gestation of 38 to 42 weeks Apgar at 5 minutes of 8 to 10 - circumcision had not taken place. The following babies were excluded : - three due to low birthweight - one due to exceptionally high birthweight of 5100 gm (born to a diabetic mother) - nine of less than 38 weeks gestation four with 5 minute Apgar below 8 38 one due to circumcision - four who began crying prior to heel-rub two whose faces were obscured by their blanket - three for later equipment failures during tape editing Of the remaining 147 infants, seven were precluded from data analysis as they did not cry following heel-lance. Ultimately, 140 infants comprised the total study sample of 77 boys and 63 girls with mean birthweight of 3439 gm (range 2500 gm to 4940 gm) ; mean maternal age was 29.3 years (range 17 to 42 years) ; 89 (63%) of the mothers were white, 38 (27%) Oriental, 8 (6%) East Indian and the remaining 4% comprised 4 Philipino and 1 North American Indian. The mode of delivery was 72 (52%) spontaneous vaginal, 34 (24%) vaginal forceps, 21 (15%) planned Cesarean section, and 13 (9%) Cesarean section following labor. At the time of the heel-lance procedure the mean time since feeding was 2.63 hours (S.D. = 1.42). Characteristics of the seven infants excluded due to no cry are presented in Appendix B. There were no apparent systematic differences between these babies and the study sample on the perinatal variables examined. 39 A p p a r a t u s A Panasonic WV - 3900 color camera was used for video recording with 3/4" video tape. In addition to audio recording on the video tape, separate sound recordings were carried out on Ampex precision 1.5 ml polyester magnetic audio tape on a Sony TC377 reel to reel tape recorder using 1/4" track recording. VU levels were set prior to recording with a 1000 hz tone. No noise suppression was used, and VU levels were not adjusted during recording to account for peaking. An AKG D109 Lavalier microphone was suspended approximately 18 cm from the infant's mouth. To cue the heel-lance event, an inaudible tone of 1000 Hz was triggered using a shure mixer connected straight into the separate audio system, and to the audio portion of the video system. The original 3/4" videotapes were copied onto 1/2" VHS tapes with an RCA time-date generator used to superimpose a digital time display for coding purposes. An RCA VHS video selectavision tape recorder (model VET650) with remote control and a RCA 48 cm playback colour monitor (JD975WV) were used during video coding. P r o c e d u r e Informed consent was obtained from a parent, usually the mother (Appendix A). Testing was conducted in a quiet room near the nursery, between 7 a.m. and 9 a.m. The infant remained in the bassinet. 40 The sleep/waking state [(1) eyes closed, no facial movement ; (2) eyes closed, facial movement ; (3) eyes open, no facial movement ; (4) eyes open, facial movement] was recorded by the investigator in the nursery. Any infants who were already crying at the time sleep/waking state was recorded were excluded from the study. Video and sound recording were carried out prior to and during the routine heel-lance procedure, which was performed by a hospital laboratory technician for PKU screening of blood samples. The camera was zoomed close-up on the infant's face at all times, so the face filled the whole frame. The baby remained partially swaddled. The lab technician first checked the infant's identification band on either the wrist or ankle, then picked up the foot and rubbed it to disinfect the skin (heel-rub phase). The heel-lance procedure was done using a small disposable metal scalpel (4.9 mm long point microlance) for incision (heel-lance phase). The heel was then squeezed and the blood sample collected into four circles on an absorbent card. There were nine lab technicians involved on rotating shifts with the number of infants seen by each technician ranging from 4 to 43. Measures Acoustic Analysis Cry was defined as "audible vocalization". Acoustic analysis was carried out by the investigator, in the Department of Linguistics acoustic laboratory at Brown University, Rhode Island. Cries were transferred from 41 the Ampex magnetic tape to a PDP 11/34 digital computer, converted from analog to digital signal and sampled at 20,000 points per second. A 20 kHz sampling rate was used as it yields a high degree of resolution in the waveform. The digitized waveform was then displayed on a VF11 vector graphics terminal. Durational measurements of latency from the end of the tone (which signalled the heel-lance event) to the start of the first cry expiration (cry latency), and from the start to the end of the first cry expiration (cry duration), were taken directly from the VT - 11 vector graphics terminal which displayed the waveform. This was done using bit pad controlled cursors which were placed to demarcate portions of the cry signal. Fundamental frequency (perceived as pitch) is the first harmonic in a complex periodic sound wave (Borden and Harris, 1980). Pitch analysis was done by short-term autocorrelation (Mertus, 1984). Pitch plots of the first cry expiration were displayed on a Tektronix videoscope, then printed. Fundamental frequency was recorded for the first pitch point (f *), then at 2 3 100 msec, (f ) and 200 msec, (f ). o o The number of cry cycles from each breath inspiration to expiration over 30 sec. from heel-lance was counted from the audiotapes by a "blind" coder. 42 Video Analysis A coding system for facial movement was devised based on the FACS approach of coding action units representing discrete facial movements. The system represented an adaption of BabyFacs (Oster, 1978). Movement is the key to scoring these actions. Action Brow Bulge Eye Squeeze Naso-labial furrow Open Lips Stretch Mouth (Vertical) Stretch Mouth (Horizontal) Lip Purse Taut Tongue Description Bulging, creasing and vertical furrows above and between brows occurring as a result of the lowering and drawing together of the eyebrows. Identified by the squeezing or bulging of the eyelids. Bulging of the fatty pads about the infant's eyes is pronounced. Primarily evidenced by the pulling upwards and deepening of the naso-labial furrow (a line or wrinkle which begins adjacent to the nostril wings and runs down and outwards beyond the lip corners). Any parting of the lips is scored as open lips. Characterized by a tautness at the lip corners coupled with a pronounced downward pull on the jaw. Often stretch mouth is seen when an already wide open mouth is opened a fraction further by an extra pull at the jaw. This appears as a distinct horizontal pull at the corners of the mouth. The lips appear as if an "oo" sound is being pronounced. Characterized by a raised, cupped tongue with sharp tensed edges. The first occurrence of taut tongue is usually easy to see, often occuring with a wide open mouth. After this first occurrence, the mouth may close slightly. Taut tongue is still scorable on the basis of the still visible tongue edges. Chin Quiver An obvious high frequency up-down motion of the lower jaw. 43 Coding was carried out using a slow motion remote controlled stop-frame feedback system for precision. A coder who had been trained in the adult FACS system (Ekman and Friesen, 1978) scored the video tapes. He was blind to the purpose of the study and all information about the infants. A second blind coder who was not FACS trained carried out reliability coding on the state and cry cycle data. All reliability coding of the discrete face actions was done by the investigator. Thirty-six seconds of video tape, divided into three second segments, were scored for each subject. Segments 1 and 2 showed responses to heel-rub ; heel-lance took place in Segment 3 ; during segments 4 to 12, blood collection was taking place so these reflect not only the responses to heel-lance, but also to the heel squeezing which occurred in blood collection. Each face variable, namely brow, eye squeeze, naso-labial furrow, open lips, vertical stretch mouth, horizontal stretch mouth, lip purse, taut tongue and chin quiver was scored as 1 if the action occurred. This was done for each of the nine facial actions separately for each of the 12 segments. 44 Perinatal Variables The following information was taken from the hospital chart : type of delivery (spontaneous vaginal, vaginal forceps, planned cesarean section, cesarean section with labor), duration of labor, obstetric maternal medication, gestational age in weeks, birthweight, head circumference, time from last feeding to heel-lance, Apgar at one minute and five minutes, type of feeding (breast or bottle), maternal age, parity. The head of the Department of Anesthesiology at Grace Hospital, Dr. H. McMorland, ranked the analgesia/anesthesia on an ordinal scale following Lester et al. (1982). The scale was modified to reflect drug combinations used in Grace Hospital. Type of analgesia/anesthesia was ranked as follows : 1 none 2 inhalation (50% N ^  0  in oxygen) 3 = perineal infiltration of local anesthetic 4 = combination of two and three 5 = epidural 6 = combination of 2 and 5 or 3 and 5 7 combination of 2, 3, and 5 8 = narcotics 9 = combination of 2 and 8 or 3 and 8 or 2, 3 and 8 10 = combination of 5 and 8 or 3, 5 and 8 11 = general anesthesia None of these mothers were given sedatives or tranquilizers. 4 5 R E S U L T S P r e l i m i n a r y A n a l y s e s S u b j e c t s E x c l u d e d D u e t o N o C r y O n l y i n f a n t s w h o c r i e d i n r e s p o n s e t o h e e l - l a n c e w e r e i n c l u d e d f o r d a t a a n a l y s i s . O f t h e 147 s u b j e c t s w h o m e t t h e i n i t i a l c r i t e r i a f o r i n c l u s i o n a s t h e y w e r e h e a l t h y , f u l l b i r t h w e i g h t , f u l l t e r m i n f a n t s , s e v e n d i d n o t c r y f o l l o w i n g h e e l - l a n c e . I t w a s o f i n t e r e s t t o s e e w h e t h e r f a i l u r e t o c r y w a s r e l a t e d t o p r i o r s l e e p / w a k e s t a t e . W h e r e a s o n l y 3 3 % ( 4 9 / 1 4 7 ) o f t h o s e w h o c r i e d h a d b e e n i n s l e e p s t a t e 1 ( q u i e t s l e e p ) p r i o r t o i n t e r v e n t i o n , 7 1 % ( 5 o u t o f 7 ) w h o d i d n o t c r y h a d b e e n i n q u i e t s l e e p . T h e r e f o r e , f a i l u r e t o c r y d i d i n d e e d a p p e a r t o b e r e l a t e d t o p r i o r s t a t e . M e a n t o t a l i n c i d e n c e o f a n y f a c e m o v e m e n t t o h e e l - r u b w a s .71 f o r t h e s e v e n w h o d i d n o t c r y , a s c o m p a r e d w i t h 4 . 4 1 f o r t h e s t u d y s a m p l e ; t o h e e l - l a n c e 1 6 . 7 2 f o r t h e n o n - c r y g r o u p a n d 2 5 . 0 3 f o r t h e s t u d y s a m p l e . T h u s t h e s e s e v e n e x c l u d e d i n f a n t s d i d a p p e a r t o b e l e s s r e s p o n s i v e t h a n t h e s t u d y g r o u p . R e l i a b i l i t y F a c e C o d i n g I n t e r o b s e r v e r r e l i a b i l i t y w a s c o m p u t e d f o r a r a n d o m l y s e l e c t e d 2 0 % o f t h e s u b j e c t s , f o r e a c h f a c e v a r i a b l e i n e a c h s e g m e n t , u s i n g t h e c o n s e r v a t i v e F a c i a l A c t i o n C o d i n g S y s t e m r e l i a b i l i t y f o r m u l a ( E c k m a n a n d F r i e s e n , 1 9 7 8 ) . T h i s i n v o l v e s a r a t i o c a l c u l a t e d s e p a r a t e l y f o r e a c h f a c e a c t i o n o v e r s e g m e n t s : 46 C o e f f i c e n t = ( n u m b e r o f a c t i o n s o n w h i c h c o d e r 1 a n d c o d e r 2 a g r e e d ) x 2 o f n u m b e r o f a c t i o n s s c o r e d b y t h e t w o c o d e r s R e l i a b i l i t y O v e r a l l r e l i a b i l i t y w a s . 8 8 . R e l i a b i l i t y f o r e a c h s u b j e c t i s p r o v i d e d i n A p p e n d i x C . T h i s s y s t e m p r o d u c e s r e l i a b i l i t y s c o r e s m u c h l o w e r t h a n c o m m o n l y u s e d p e r c e n t a g e a g r e e m e n t , w h i c h g i v e s c r e d i t t o a g r e e m e n t o f t h e a b s e n c e o f a n e v e n t e q u a l t o a g r e e m e n t o f o c c u r r e n c e . N u m b e r o f C r y C y c l e s T w o i n d e p e n d e n t r a t e r s c o u n t e d t h e n u m b e r o f c r y c y c l e s f o r 3 0 s e c o n d s f o l l o w i n g h e e l - l a n c e . R a t e o f a g r e e m e n t w a s 8 5 % w i t h a n e r r o r m a r g i n o f - 1 c r y c y c l e . S t a t e I n o r d e r t o c h e c k o n t h e o b j e c t i v i t y o f t h e s t a t e r a t i n g , i t w a s s c o r e d b y t w o b l i n d c o d e r s u s i n g t h e f i r s t t h r e e s e c o n d s o f v i d e o t a p e , s c o r e d i n r e a l t i m e . I n t e r - r a t e r a g r e e m e n t b e t w e e n t h e t w o c o d e r s w a s 8 3 % . D i s t r i b u t i o n s o f D e p e n d e n t M e a s u r e s S k e w o f t h e d i s t r i b u t i o n o f e a c h d e p e n d e n t v a r i a b l e w a s e x a m i n e d . D u e t o d e p a r t u r e s f r o m n o r m a l i t y , l o g 10 t r a n s f o r m a t i o n s w e r e a p p l i e d t o c r y l a t e n c y , f a c e l a t e n c y , a n d c r y d u r a t i o n e a c h o f w h i c h s h o w e d p o s i t i v e s k e w , t h e r e b y i m p r o v i n g t h e a p p r o x i m a t i o n t o n o r m a l . 47 Rate of Occurrence of Each Facial Action Frequency and percent of occurrence of each of the nine face variables (Table 1) were examined prior to data analysis. Up to 99% of the infants showed brow bulge, eye squeeze, naso-labial furrow and open lips in some segment. Taut tongue was observed in up to 74% of infants, and vertical stretch mouth in 43%. In contrast, horizontal stretch mouth occurred in only 3 out of 140 (2%) of infants to heel-rub, and not at all to heel-lance ; pursed lips occurred in 2 of 140 to heel-rub and 1 of 140 to heel-lance. Thus these two face actions were not considered any further, and omitted from statistical analysis. Chin quiver occurred infrequently and mainly in later segments, so it was not analyzed individually, but was included in total amount of face movement when all actions were summed. Technician Effects Chi-square analyses were carried out on each facial action for each segment across the nine lab technicians. Of the heel-rub analyses, statistically significant technician effects at p^.01 were obtained for naso-labial furrow in 21.84, pss.005, and segment 2, "XT (8) = 23.52, p= .003, and for eye squeeze in segment 2, % . (8) = 20.53, p=r.008. Table 1 Frequency and Percent of Each Facial Action Across Segmen Segment Brow Eye Naso-Labial L i p Squeeze Furr Part V e r t i c a l S t r e t c h Mouth H o r i z o n t a l S t r e t c h Mouth 1 60 (43%) 60 (43%) 57 (41%) 98 (70%) 12 (9%) 3 (2%) 2 59 (42%) 56 (40%) 54 (39%) 101 (72%) 9 (6%) 3 (2%) 3 135 (96%) 135 (96%) 136 (97%) 137 (98%) 43 (31%) 0 (0%) 4 138 (99%) 138 (99%) 138 (99%) 139 (99%) 60 (43%) 0 (0%) 5 137 (98%) 138 (99%) 137 (98%) 139 (99%) 54 (39%) 0 (0%) 6 135 (97%) 134 (96%) 134 (96%) 138 (99%) 51 (36%) 0 (0%) 7 136 (97%) 135 (96%) 134 (96%) 137 (98%) 48 (34%) 0 (0%) 8 132 (94%) 129 (92%) 129 (92%) 132 (94%) 40 (29%) 0 (0%) 9 130 (93%) 128 (91%) 127 (91%) 133 (94%) 34 (24%) 0 (0%) 10 127 (91%) 126 (90%) 125 (89%) 132 (94%) 28 (20%) 0 (0%) 11 125 (89%) 124 (89%) 125 (89%) 135 (96%) 31 (22%) 0 (0%) 12 135 (89%) 123 (88%) 124 (89%) 134 (96%) 27 (19%) 0 (0%) 1 to 12 L i p Purse 2 (1%) 0 ( 0 % ) 0 ( 0 % ) 0 (0%) 1 (1%) 0 (0%) 0 ( 0 % ) 1 (1%) 2 (1%) 3 (2%) 3 (2%) 1 (1%) Taut Tongue Chin Quiver 23 (16%) 4 ( 3%) 22 (16%) 3 ( 2%) 95 (68%) 3 ( 2%) 98 (70%) 4 ( 3%) 104 (74%) 8 ( 6%) 96 (69%) 9 ( 6%) 94 (67%) 7 ( 5%) 91 (65%) 13 ( 9%) 86 (61%) 18 (13%) 84 (60%) 14 (10%) 80 (57%) 9 ( 6%) 79 (56%) 9 ( 6%) 49 Following heel-lance, which occurred in segment 3, there were no significant technician effects in segments 3 to 7, i.e. for 15 seconds after heel-lance. Of the remaining segments, which reflected heel squeezing to complete blood collection, the following significant technician effects were found : brow bulge segment 11, (8) = 28.43, p = .0004 and segment 12, OC.2(8) = 22.35,p= .0004, eye squeeze segment 10, "X.* (8) = 19.75, p .01, segment 11, (8) = 31.08, p= .0001, segment 12, "X. (8) = 26.91, p=.0007, 2 z naso-labial furrow segment 11,'X- (8) = 21.90, p=r .005, segment 12, "X- (8) = 21.51, pr .0009, lip part segment 8, "Xl (8) = 30.50, p= .0002, segment 10, *X.2(8) = 19.69, p=.01, segment 11, "X2, (8) = 26.87, p= .0007, vertical stretch 2 2. mouth segment 10, "X (8) = 19.58, .01, taut tongue segment 8, "X. (8) = 19.13, p = .01 . 1 2 Univariate one-way analysis of variance was carried out on f , f , 3 f , number of cry cycles, cry duration and latency to cry, across technicians. There was no statistically significant technician effect for any of the cry variables. The ANOVA summary tables are presented in Appendix D. There was no evidence infants in any particular state were assigned 2 disporportionately across technicians OC. (24) = 22.02, p = .58. Facial data from segments 3 to 7 only were included in analyses of heel-lance as segments 8 through 12 reflected the impact of heel squeezing rather than lancing. 50 Time Since Feeding Since heel-lance for PKU testing was carried out between 7 a.m. and 9 a.m. only, it was not possible to experimentally control for time from feeding to heel-lance. This was a potentially important confounding factor in the study, as infant behavioral state may vary depending on hunger. One-way analysis of variance was performed on time since feeding across sleep/wake states. Mean time since feeding for each state is shown in Table 2. There were no statistically significant differences between groups F (3,136)= 1.60, n.s. The ANOVA summary table is provided in Appendix E. Hypothesis Testing : State and Sex Facial Action Individual face actions were summed across two segments (3 seconds each) prior to heel-lance (during heel-rub) and separately for two segments (3 seconds each) following heel-lance. A three-way univariate analysis of variance (ANOVA) of summed facial movement 4 (state) x 2 (sex) x 2 (heel-rub/heel-lance) with repeated measures on the last factor was carried out. Main effects of state F (3,131 ) = 7.29, p=.0001 and heel-rub/heel-lance F (1,132 )= 295.35, p=.0001 were highly significant as was the state x heel-rub/lance interaction F ( 3,132) = 3.69, p=.01. There were no significant sex differences. The interaction was explored further with Duncan post hoc 51 Table 2 Mean Time Since Feeding by State Quiet Sleep Active Sleep Quiet Awake Active Awake n 49 45 20 26 M 2.42 2.99 2.38 2.55 SD 1.32 1.56 1.31 1.39 52 Table 3 Mean, (standard deviations) and Duncan post-hoc Comparisons for F a c i a l Movement to Heel Rub a and Heel Lance b Behavioural State 1 Quiet Sleep n 49 Active Sleep 45 3 Quiet Awake 20 Active- Duncan post-hoc Awake comparisons 26 (p 4s .05) Heel Rub M SD 2.9 ( 3.0) 4.3 ( 3.4) 6.7 (4.4) 5.6 ( 4.0) 1 < 3 , 1 < 4 , 2 < 3 Heel Lance SD 9.5 2.2 10.1 1.3 10.7 1.9 10.2 1.6 1 < 3 Note: f a c i a l actions summed across 2 segments (6 seconds) p r i o r to heel lance. f a c i a l actions summed across 2 segments (6 seconds) following heel lance. 53 comparisons separately across each level of the repeated measures factor, with error rate set at .05. The ANOVA summary table is presented in Appendix F ; and Duncan tests of differences in mean facial action in Table 3. Quiet awake (alert) infants responded with significantly more facial movement as compared with those who were in quiet sleep to both heel-rub and heel-lance stimulation. Additional state differences were apparent to a significant degree only in heel-rub, namely quiet/awake infants also showed more facial response than those active/sleep, and active/awake more than quiet/sleep. For a more fine-grained assessment of facial movement, log-linear analyses (Haberman, 1978) of each face action in each segment were carried out to examine the interrelationships of each face action with prior behavioral state and sex. Log-linear analyses showed only one statistically significant 2 effect for sex, namely brow bulge in segment 2, X (3) = 8.28, p .04. No significant sleep/wake state by sex interactions were found. As 42 log-linear analyses were executed, at a type I error rate of .04 one would expect 2/50 tests to be significant by chance, therefore, this sole sex effect was considered a chance finding and sex was eliminated from consideration in facial movement. The analyses were repeated as two-way contingency tables, using chi-square statistic for significance testing when frequency of occurrence was below 90%, and adjusted residuals to indicate which cells deviated from expected levels when overall significance was found. Frequency of each face action to heel-rub (segments 1 and 2) and heel-lance (segments 3 and 4) by prior behavioral state is shown in Tables 4 to 9. 54 Table 4 Frequency of Taut Tongue Action to Heel Rub (Segments 1 and 2) and Heel-Lance (Segments 3 and 4) by State State Quiet Active Quiet Active Sleep Sleep Awake Awake "X2 n = 49 n = 45 n = 20 n = 26 Heel Rub Seg. 1 NO 46 (94%) 42 (93%) 12 (60%) 17 (65%) 2.4 2.1 -3.1 -2.8 21.27 .0001 YES 3 (6%) -2.4 3 (7%) -2.1 8 (40%) 3.1 9 (35%) 2.8 Seg. 2 NO 45 (92%) 41 (91%) 12 (60%) 20 (77%) 1.8 1.5 -3.2 -1.1 13.55 .003 YES 4 (8%) -1.8 4 (9%) -1.5 8 (40%) 3.2 6 (23%) 1.1 Heel Lance Seg. 3 NO 21 (43%) 2.0 10 (22%) -1.7 • 3 (15%) -1.8 11 (42%) 1.2 8.54 .036 YES 28 (57%) -2.0 35 (78%) 1.7 17 (85%) 1.8 15 (58%) -1.2 Seg. 4 NO 20 (41%) 2.0 11 (24%) -1.0 3 (15%) -1.6 8 (31%) .1 5.54 n.s. YES 29 (59%) -2.0 34 (76%) 1.0 17 (85%) 1.6 18 (69%) - .1 55 Table 5 Frequency of Vertical Stretch Mouth to Heel Rub (Segments 1 and 2) and Heel Lance (Segments 3 and 4) by State State Quiet Active Quiet Active Sleep Sleep Awake Awake "X? n = 49 n = 45 n = 20 n = 26 Heel Rub Seg. 1 NO 47 (96%) 1.4 44 (98%) 1.8 15 (75%) -2.8 22 (85%) 01.4 12.00 .007 YES 2 ( 4 % ) -1.4 1 ( 2 % ) -1.8 5 (25%) 2.8 4 (15%) 1.4 Seg. 2 NO 49 (100%) 2.3 44 (98%) 1.4 14 (70%) -4.6 24 (92%) - .3 23.23 .0001 YES 0 -2.3 1 ( 2 % ) -1.4 6 (30%) 4.6 2 ( 8%) .3 Heel Lance Seg. 3 NO 38 (78%) 1.6 33 (73%) .7 9 (45%) -2.5 17 (65%) - .5 7.65 .05 YES 11 (22%) -1.6 12 (27%) - .7 11 (55%) 2.5 9 (35%) .5 Seg. 4 NO 31 (63%) 29 (64%) 7 (35%) 13 (50%) 1.1 1.2 -2.2 - .8 6.28 n.s, YES 18 (37%) 16 (36%) 13 (65%) 13 (50%) -1.1 -1.2 2.2 .8 56 Table 6 Frequency of Naso-Labial Furrow to Heel Rub (Segments 1 and 2) and Heel Lance (Segments 3 and 4) by State State Quiet Active Quiet Active Sleep Sleep Awake Awake X 2 n = 49 n = 45 n = 20 n = 26 Heel Rub Seg. 1 NO 37 (76%) 2.9 25 (56%) - .6 7 (35%) -2.4 14 (54%) - .6 10.81 .01 YES 12 (24%) -2.9 20 (44%) 13 (65%) .6 2.4 12 (46%) .6 Seg. 2 NO 40 (82%) 3.6 26 (58%) - .6 9 (45%) -1.6 11 (42%) -2.2 14.99 .002 YES 9 (18%) -3.6 19 (42%) 11 (55%) .6 1.6 15 (58%) 2.2 Heel Lance Seg. 3 NO 3 ( 6 % ) 0 1 ( 5 % ) 0 1.7 -1.4 .6 -1.0 YES 46 (94%) 45 (100%) 19 (95%) 26 (100%) -1.7 1.4 - .6 1.0 Seg. 4 NO 2 ( 4 % ) 0 0 0 1.9 -1.0 - .6 - .7 YES 47 (96%) -1.9 45 (100%) 20 (100%) 26 (100%) 1.0 .6 .7 57 Table 7 Frequency of Eye Squeeze to Heel Rub (Segments 1 and 2) and Heel Lance (Segments 3 and 4) by State P r i o r State Heel Rub Heel Lance Quiet Sleep Active Sleep Quiet Awake Active Awake n = 49 n = 45 n = 20 n = 26 Seg. 1 NO 35 (71%) 2.5 23 (51%) -1.0 7 (35%) -2.2 15 (58%) .1 8.76 .03 YES 14 (29%) -2.5 22 (49%) 13 (65%) 1.0 2.2 11 (42%) - .1 Seg. 2 NO 39 (80%) 3.5 25 (56%) - .7 9 (45%) -1.5 11 (42%) 02.0 13.47 .004 YES 10 (20%) -3.5 20 (44%) 11 (55%) .7 1.5 15 (58%) 2.0 Seg. 3 NO 3 ( 6%) 1.2 1 ( 2 % ) - .6 1 ( 5 % ) .4 0 -1.1 YES 46 (94%) -1.2 44 (98%) 19 (95%) .6 - .4 26 (100%) 1.1 Seg. 4 NO 2 ( 4%) 1.9 0 -1.0 0 - .6 0 - .7 YES 47 (96%) 45 (100%) 20 (100%) 26 (100%) -1.9 1.0 .6 .7 58 Table 8 Frequency of Brow Action to Heel Rub (Segments 1 and 2) and Heel Lance (Segments 3 and 4) by State State Heel Rub Heel Lance Quiet Sleep Active Sleep Quiet Awake Active Awake n = 49 n = 45 n = 20 n = 26 Seg. 1 NO 80 (57%) 33 (67%) 1.8 25 (56%) - .3 7 (35%) -2.2 15 (58%) .1 6.14 .10 YES 60 (43%) 16 (33%) -1.8 20 (44%) 13 (65%) + .3 2.2 11 (42%) - .1 Seg. 2 NO 81 (58%) 36 (74%) 2.7 25 (56%) - .4 9 (45%) -1.3 11 (42%) -1.8 8.93 .03 YES 59 (42%) 13 (27%) -1.7 20 (44%) 11 (55%) .2 .9 15 (58%) 1.2 Seg. 3 NO 5 (4%) 3 ( 6 % ) 1.2 1 ( 2 % ) - .6 1 ( 5 % ) .4 0 -1.1 YES 135 (96%) 46 (94%) -1.2 44 (98%) 19 (95%) .6 - .4 26 (100%) 1.1 Seg. 4 NO 2 (1%) 2 ( 4 % ) 1.9 0 -1.0 0 - .6 0 - .7 YES 47 (96%) 45 (100%) 20 (100%) 26 (100%) 138 (99%) -1.9 1.0 .6 .7 Table 9 Frequency of L i p Part to Heel Rub (Segments 1 and 2) and Heel Lance (Segments 3 and 4) by State State Quiet Sleep Active Sleep Quiet Awake Active Awake n = 49 n = 45 n = 20 n = 26 Heel Rub Seg. 1 NO 21 (43%) 2.4 14 (31%) .2 3 (15%) -1.6 4 (15%) -1.8 8.67 YES 28 (57%) -2.4 31 (69%) 17 (85%) - .2 1.6 22 (85%) 1.8 Seg. 2 NO 19 (39%) 2.1 12 (27%) - .2 4 (20%) - .8 4 (15%) -1.6 5.57 YES 30 (61%) -2.1 33 (73%) 16 (80%) .2 .8 22 (85%) 1.6 Heel Lance Seg. 3 NO 4 ( 4 % ) 0 1 ( 5 % ) 0 1.2 -1.2 1.0 - .8 YES 47 (96%) 45 (100%) 19 (95%) 26 (100%) -1.2 1.2 -1.0 .8 Seg. 4 NO 1 ( 2 % ) 0 0 0 2 ( 1 % ) 1.9 -1.0 - . 6 -.7 YES 138 (99%) 47 (96%) 1.4 45 (100%) 20 (100%) 26 (100%) .7 .4 .5 60 Heel Rub Facial reaction to heel-rub varied significantly in relation to prior sleep/wake state. Occurrence of taut tongue (Table 4), vertical stretch mouth (Table 5), and naso-labial furrow (Table 6) showed statistically significant differences across state in segments 1 and 2. In segment 1 infants who had been quiet/awake prior to heel-rub (state 3) showing significantly more occurrence of each facial action. Infants who had been in quiet/sleep (state 1) prior to heel-rub showed significantly fewer facial actions. Segment 2 showed the same pattern except for naso-labial furrow, where active/awake and quiet/ awake were about equal. In segment 1, eye squeeze (Table 7) and brow action (Table 8) showed more reaction for quiet/awake, (3) = 8.76, pa.03 and"^*(3) = 6.14, pi.10 respectively. Eye squeeze in segment 2 showed similar reactivity for both awake states, and significantly less for quiet/sleep. Lip part (Table 9) occurred frequently to heel-rub across all sleep/wake states (70% overall in segment 1 and 72% in segment 2). It was not significantly related to prior behavioral state in either segment 1 or 2, but there appeared to be a trend toward less reaction by those who had been asleep, with least occurrence following quiet/sleep. Heel Lance To heel-lance (segment 3) 96% of the infants showed brow action (Table 8), 69% eye squeeze (Table 7), 97% naso-labial furrow (Table 6), 98% lip part (Table 9), thus these face variables formed a constellation. In contrast, 61 only 31% showed vertical stretch mouth (Table 5) and 68% taut tongue (Table 4). More frequent occurrence of stretch mouth occurred following state 3, o£(3) - 7.65, pe.05, and similarly with taut tongue X?*(3) = 8.54, p=?.04 following state 3 ; lower rate of occurrence following sleep was apparent. The trend across state over segments 1 to 7 for each face variable is shown graphically as percent occurrence in Figures 1 to 6. Latency to Face Movement Two-way 4 (state) x 2 (sex) analysis of variance was carried out on the latency to face movement. There was a significant main effect for sex F (1,132) = 4.73, ps .03, (with shorter time to reaction for boys than girls, but not for state and no significant interaction. Means and standard deviations across state and sex are shown in Table 10, and the ANOVA summary table in Appendix G. Cry Measures Cry Cycles, Latency and Duration of First Cry to Heel Lance Number of cry cycles, latency, and duration of first cry to heel-lance were analyzed with univariate two-way 4 (state) x 2 (sex) analysis of variance. The means and standard deviations by sex and state are provided in Table 11 and ANOVA summary tables in Appendices H, I and 3. Boys cried significantly more cycles than girls, F (1,132) = 3.74, ps.05. The ANOVAs showed no significant differences across state and no sex by state 62 Fiqure 1. Percentage Taut Tongue to Heel Rub (Segments 1 to 2) and Heel Lance (Segments 3 to 7) by State. * — . S t a t e 4 State 3 •—•» — . — State 2 State 1 c. I: n — • • a • * i 3 SEGMENT 4 . o 7 _2, Percentage Vertical Stretch Mouth to Heel Rub (Segments 1 to Heel Lance (Segments 3 to 7) by State, 2) and 64 % 1 0 0 90 80 70 6 0 50 4 0 30 2 0 10 X State 4 State 3 • — • — • — State 2 State 1 _ j r 3 4 SEGMENT 6 F i g u r e 3 , P e r c e n t a g e N a s o - l a b i a l F u r r o w t o H e e l Rub ( S e q m e n t s 1 t o 2) a n d H e e l L a n c e ( S e g m e n t s 3 t o 7) by S t a t e , % 100 90 80 70 60 50 40 30 20 JO 3 4 SEGMENT * State 4 C3— • — • O State 3 State 2 State 1 Figure 4 Percentaae Eve Saueeze to Heel Rub CSeaments 1 to 2) and Heel Lance (Segments 3 to 7) by State. 66 % 100 90 80 70 60 50 40 30 20 10 I yi State 4 State 3 0— •- state 2 State 1 2 . 3 4 SEGMENT Fi qure 5 , Percentage Brow Action to Heel Rub (Segments 1 to 2) and Heel Lance (.Segments 3 to 7) by State, 67 % 100 90 80 70 60 50 40 30 A 20 10 4 J: J * State 4 State 3 Q ~ * — ' — • State 2 State 1 2 . 3 4 5 6 SEGMENT Figure 6 , Percentage Lip Part to Heel Rub (Segments 1 to 2) and Heel Lance (Seqments 3 to 7) bv State. 68 Table 10 Means and Standard Deviations for Latency to Facial Movement* by State and Sex Quiet Active Quiet Active Sleep Sleep Awake Awake Boys M 0.92 0.63 0.99 1.11 0.87 SD .54 .56 1.07 .94 .73 n_ 46 26 12 13 77 G i r l s M 1.37 1.22 0.82 0.90 1.16 SD .99 .74 .56 .69 .82 n 23 19 8 13 63 TOTAL M 1.13 0.88 0.92 1.01 SD .81 .70 .89 .81 'r» 49 45 20 26 * log 10 (minutes) 69 Cry Cycles Table 11 Means and Standard Deviations f o r Cry Variables by State and Sex Quite Sleep Active Sleep Quiet Awake Active Awake Boys M SD n 13.00 6.11 26 16.15 5.44 26 11.42 5.37 12 14.92 6.8 13 14.14 G i r l s M SD n 11.30 5.58 23 12.16 5.87 19 12.37 4.41 8 13.77 7.46 13 12.21 Cry Latency* Boys SD n 2.26 1.17 26 1.49 1.41 26 2.29 1.25 12 2.29 1.12 13 2.01 G i r l s M SD n 2.77 1.01 23 2.44 1.22 19 1.97 1.21 8 2.06 1.24 13 2.43 Cry Duration* Boys M^  SD n 3.29 .45 26 3.25 .51 26 3.12 .55 12 3.31 .41 13 3.25 G i r l s M SD n 3.33 .43 23 3.34 .42 19 3.31 .62 8 3.34 .45 13 3.33 * Log 10 (msec) 70 interactions. However, Kruskal-Wallis ANOVA showed a significant sex 2 difference with boys crying sooner than girls X (4) = 12.39, p - .02. Log-linear analysis of cry latency (Tables 19 and 20) confirmed the sex finding and suggested a possible difference across states, with infants in quiet/sleep crying 2 less quickly than the other states X (3) = 8.18, p = .04. Fundamental Frequency Mean fundamental frequency of the first post heel-lance cry, 1 2 measured at the first pitch-point (fQ ), at 100 msec, (f ), and at 200 msec. 3 (fQ ), is displayed graphically by state in Figure 7. Three-way 4 (state) x 2 (sex) x 3 (time) univariate analysis of variance with repeated measures on the time factor was carried out. Means and standard deviations for each of the fundamental frequency measures by state are shown in Table 12, and ANOVA summary tables in Appendix 3. There were no main effects for state or sex and no interaction. The time factor however, was significant, F (2, 234.0= 14.26, p=.0000, with a rising pattern evident over the 300 msec, for boys and girls in each of the behavioral states. Pitch-Plot Pattern A pitch-plot was computer generated for the entire first cry to heel-lance for each infant. Each was classified into one of six categories based on the overall pattern. Frequency of pitch-plot pattern 71 5 0 0 4 0 0 . / ><— — State 4 State 3 D State 2 'State 1 ~I— 1 s t I J 0 0 msec 20Tj m s e c F i g u r e 7 , Mean F u n d a m e n t a l F r e q u e n c y o f t h e F i r s t C r y t o H e e l L a n c e O v e r T i m e by S t a t e 72 Table 12 Means and Standard Deviations of Fundamental Frequency (bz) of Fi r s t Cry to Heel Lance Quiet Active Quiet Active Sleep Sleep Awake Awake f Q 1 M 458.8 476.9 489.9 480.0 472.6 SD 129.5 197.1 195.5 162.7 f Q 2 M 491.3 505.9 520.1 535.9 507.9 SD 146.0 205.7 170.7 165.9 f Q 3 M 514.5 578.7 527.9 564.2 546.8 SD 157.3 231.2 169.1 172.0 M 488.2 520.5 512.6 526.7 73 Table 13 Frequencies and Percentages'*" of O v e r a l l Cry Patterns by State P i t c h Plot R i s i n g / F a l l i n g / State Rising F a l l i n g F a l l i n g Rising F l a t No Pattern Quiet Sleep 2 (4%) 7 (14%) 15 (31%) 12 (24%) 11 (22%) 2 ( 4 % ) Active Sleep 3 (7%) 5 (11%) 15 (33%) 12 (27%) 7 (16%) 3 (7%) Quiet Awake 0 5 (26%) 6 (32%) 3 (16%) 2 (10%) 3 (16%) Activ e Awake 0 2 (8%) 10 (38%) 4 (15%) 6 (23%) 4 (15%) TOTALS 5 19 46 31 26 12 N = 139* + percentage of occurence i n each behavioural state * p i t c h p l o t missing for one case 74 by sleep/wake state in shown in Table 13. Approximately one-third of all the infants showed a rising-falling pitch pattern, regardless of the prior state. There were no overall significant differences between the states. It was interesting to note, however, that of the few infants (5 out of 139) who showed only a rising pitch pattern, all had been asleep prior to intervention. F a c i a l A c t i o n s a n d C r y M e a s u r e s I n t e r r e l a t i o n s h i p s A m o n g F a c i a l A c t i o n s It was of interest to examine the extent to which the facial actions were related to each other, to see to what degree any different parts of the face were moving independently. Each face action was summed across the seven heel-rub and heel-lance segments combined, and Spearman correlations between face actions calculated (Table 14). Eye squeeze, naso-labial furrow and brow action tended to occur simultaneously forming a cluster of facial actions. Mouth movement (lip part, vertical stretch and taut tongue) were significantly related to upper face action and naso-labial furrow, but only to a moderate degree. All infants showing vertical stretch mouth also had naso-labial furrow, but the correlation between the two was not high as there were many infants with naso-labial furrow whose mouths were open but not stretched. Lip part was the most common initial face action to disturbance, and whereas all infants with taut tongue had their lips apart, the converse was certainly not true, thus no significant correlation. 75 Table 14 Spearman Cor r e l a t i o n s Between F a c i a l Actions Summed Over Heel Rub and Heel Lance Segments Eye Naso-labial L ip V e r t i c a l Taut Squeeze Furrow Part Stetch Mouth Tongue Brow Eye Squeeze Naso-labial Furrow Li p Part V e r t i c a l S tretch Mouth *.95 **.78 **.50 **.27 **.84 **.46 *.24 *.24 **.49 *.23 .14 .28 *.24 *.23 **.27 * p 6 .01 ** p £ .001 76 Table 15 Pearson Correlations Between Total Facial Action to Heel Lance and Latency to Facial Action to Heel Lance Latency to Facial Movement Total Face Action -.52** ** p x^.OOl 77 Discrete face actions for segments 3 to 7 were summed to provide an overall measure of facial movement to heel-lance. Latency to facial movement was significantly negatively correlated with total facial action (Table 15). This indicated that faster speed of initial face movement following heel-lance was significantly related to a greater amount of facial action overall following heel-lance. I n t e r r e l a t i o n s h i p s A m o n g C r y M e a s u r e s Pearson correlations between the cry measures are given in Table 16. The number of cry cyles following heel-lance was unrelated to either the speed of cry reaction or to the duration of the first cry. The only significant correlation at p^.01 was between fundamental frequency at 300 msec, and speed of cry response. R e l a t i o n s B e t w e e n C r y M e a s u r e s a n d F a c e M o v e m e n t As shown in Table 17, speed of facial movement to heel-lance was significantly correlated with speed of cry response. Total face action following heel-lance was negatively correlated with latency to cry, indicating faster speed to cry was related to amount of face action. Also, higher number of cry cycles following heel-lance was significantly correlated with amount of face action, but only to a rather moderate degree (r_ = .24). This suggested that many babies who cried only a few cycles moved their faces a good deal. Fundamental frequency showed no relation to facial movement. 78 Table 16 Pearson C o r r e l a t i o n s Between Cry Measures Number of Cry Cry Cycles Latency Duration f Q 1 .06 .03 .08 f Q 2 -04 .09 .04 f Q 3 .05 .26* -.07 Number of Cycles -.17 -.10 Cry Latency -.12 * p 4 .01 79 Table 17 Pearson C o r r e l a t i o n s Between Cry Measures and Face Actions to Heel Lance fo 1 f 2 ^o fo 3 Number of Duration Latency to Cry Cycles of 1st Cry Cry Latency to F a c i a l Movement .00 .15 .13 -.07 .13 .53* Total Face Action .10 .03 .01 ,24" ,00 -.34" * p .01 ** p ^  .001 80 E x p l o r a t o r y A n a l y s e s R e l a t i o n s h i p o f P e r i n a t a l a n d M a t e r n a l  V a r i a b l e s t o P a i n E x p r e s s i o n Stepwise discriminant analyses were used to examine the relationship of perinatal and maternal variables (see page 43) to each dependent measure. A median split was used to form low and high groups for each variable (except fundamental frequency). This appeared preferable to extreme groups as this provides a far more conservative test of the relationship between the set of predictor variables and the response measure. Stepwise discriminant analysis reduced the initial predictor variables to an optimum set. The classification tables show the degree of success in identifying the actual group membership using the reduced set of predictor variables. C r y L a t e n c y Stepwise discriminant analysis of fast and slow cry response resulted in three significant variables : sex, F (1,133) = 6.80, p==.01, state, F (1,133) = 6.63, p=.02, and Apgar (at 5 minutes), F (1,133) = 4.50, p= .05. Classification results are presented in Table 18. A three-way log-linear analysis of fast and slow cry response, sex and state was carried out to further explicate the interrelationship of these variables. Results of the log-linear analysis are SI shown in Table 19. Observed frequency, percent of males and females across state by speed of cry response, and the ratios of the log-linear parameter estimates to standard error are in Table 20. Significantly more boys showed a fast cry response and girls a slower cry response. Cry response speed was significantly slower for infants in quiet/sleep as compared with the other three sleep/awake states. The three-way interaction between state and sex in relation to speed of cry response just missed statistical significance (p = .06). Mean Apgar was 9.40 for the fast cry response and 9.22 for the slower cry response group. Three infants had a 5 minute Apgar of 8 ; all of these had cry latencies above 500 msec, which probably accounted for the difference between the groups. Fundamental Frequency Mean fundamental frequency of pain cry is generally reported in the literature to be around 400 to 600 hz for healthy neonates. Low and high 1 2 3 groups for f , f , and f were formed by categorizing f q below 600 hz and at or above 600 hz respectively. The discriminant functions for 2 3 f and fQ were not significant. 82 Table 18 C l a s s i f i c a t i o n Results of P e r i n a t a l Measures to Short ( ^ 499msec. ) and Long ( ^ 500msec. ) Cry Latency P r e d i c t e d Group Membership A c t u a l Group Membership Short Cry Latency Long Cry Latency Short Cry Latency n = 72 46 (64%) 26 (36%) Long Cry Latency n = 68 24 (35%) 44 (65%) o v e r a l l 64% 83 Table 19 Log-Linear Analysis of Short and Long Cry Latency (C) by State (S) and Sex (X) Models formed by d e l e t i n g terms from modex XSC Model Simple E f f e c t Chi-Square P. SC,XC,XS 3 d i f f e r e n c e due to d e l e t i n g XSC 3 7.23 .06 Step 1 Best Model found i s : SC, XC, XS sc,xc 6 7.76 .26 d i f f e r e n c e due to d e l e t i n g XS 3 .54 .91 SC,XS 4 13.16 .01 d i f f e r e n c e due to d e l e t i n g XC 1 5.93 .01 XC,XS 6 15.41 .02 d i f f e r e n c e due to d e l e t i n g SC 3 8.18 .04 84 Table 20 Observed Frequency, Percent of Column Totals and Ratio of the Log-Linear Parameter Estimates to Standard Error for Short and Long Cry Latency by State and Sex Latency to Cry State Male Female Total Fast Quiet Sleep 12 (25%) 5 (20%) 17 -2.80 Cry .86 -.86 Response Active Sleep 21 (45%) 7 (28%) 28 .99 2.17 -2.17 Quiet Awake 6 (13%) 6 (24%) 12 .92 -2.01 2.01 Active Awake 8 (17%) 7 (28%) 15 .49 - .42 .42 Sub-Total 47 (61%) 25 (40%) 72 1.51 -1.51 Slow Quiet Sleep 14 (47%) 18 (47%) 32 2.80 Cry - .86 .86 Response Active Sleep 5 (17%) 12 (32%) 17 -.99 -2.17 2.17 Quiet Sleep 6 (20%) 2 (5%) 8 -.92 2.01 -2.01 Active Awake 5 (17%) 6 (16%) 11 -.49 .42 - .42 Sub-Total 30 (39%) 38 (60%) 69 1.51 -1.51 N = 140 85 Discriminant analysis of perinatal data to low and high f showed a significant relationship with birthweight, F (1,133) = 11.68, p= .001. Classification results are provided in Table 21. Mean birthweight by low and high f * is shown in Table 22. As birthweight is known to also be related to maternal smoking during pregnancy, further analysis of the subgroups for whom smoking status was available (N = 113), using two-way analysis of variance 2 (f * level) x 2 (smoker/nonsmoker) was carried out. As shown in the ANOVA summary table (Appendix L) birthweight was significantly related to f * , F (1,112) = 7.23, p .008, and to smoking status F (1,112) = 4.48, p= .04. Mean birthweight for these groups is given in Table 23, showing the infants whose mothers had smoked, who also had a high f * to be the smallest babies. As it was of interest to examine the relationship between birthweight itself and level of fundamental frequency, the analysis was rerun omitting the infants whose mothers had smoked. Birthweight remained significant F (1,81) = 6.06, p = .025. Classification results (Table 24) improved noticeably, with correct identification of the high f * group at 85%. Latency of Facial Response to Heel Lance Discriminant analysis showed no statistically significant relationship between the perinatal variables and speed of facial response. 86 Table 21 C l a s s i f i c a t i o n Results of P e r i n a t a l Measures to Low ( 599hz) or High ( £ 600hz) f 0 1 Actual Group Membership Predicted Group Membership Low f 0 1 High f 0 1 Low f Q 1 n = 117 78 (67%) 39 (33%) High f Q 1 n = 23 10 (43%) 13 (56%) o v e r a l l c o r r e c t 65% 87 Table 22 Mean Birthweight (gm) by Level of f 0 ' LOW HIGH 4z 599hz ^ 600hz n = 117 n • 23 M 3496 3192 SD 426 324 Table 23 Mean Birthweight (gm) by Level of f g 1 f o r Smokers and Non-Smokers LOW HIGH d=599hz ^ 600hz n = 97 n = 16 SMOKER NO 3537 3233 3492 n = 88 n = 75 n = 13 YES 3336 3060 3303 n = 25 n = 22 n = 3 N = 113 3491 3201 89 Table 24 C l a s s i f i c a t i o n Results f o r Low (^599 hz) and High (^600 hz) f Q 1 Excluding Infants of Smoking Mothers Actual Group Membership Predicted Group Membership Low f Q 1 High f Q 1 Low f Q 1 High f Q 1 n = 75 n = 13 57 (76%) 18 (24%) 2 (15%) 11 (85%) Overal l c o r r e c t 77% 90 Face Movement Each facial action was summed across heel-rub segments 1 to 2 (6 seconds) to give a measure of total face movement to heel-rub. Similarly, each facial action was summed across segments 3 to 7 (15 seconds) following heel-lance to reflect total face movement to heel-lance. A median split provided two groups of low and high face movement, separately for heel-rub and heel-lance. Heel R u b Classification results of stepwise discriminant analysis of low and high face movement to heel-rub is provided in Table 25. Only state was statistically significant, F (1,133) = 10.28, p = .001. About two-thirds of the infants were correctly classified as showing low or high facial reactivity to heel-rub on the basis of their sleep/wake state alone. Heel Lance Classification results of stepwise discriminant analysis of low and high face movement to heel-lance is shown in Table 26. The following variables were significant : type of obstetric medication, F (1,130) = 14.62, p= .001 ; mode of delivery, F (1,130) = 11.60, p=.001 ; state, F (1,130) = 6.09, p = .02 ; weeks of gestation, F (1,130) = 5.93, p*.02. 91 Table 25 C l a s s i f i c a t i o n Results f o r Low and High F a c i a l Movement to Heel Rub Predicted Group Membership  Actual Group Membership Low Face Action High Face Action Low n = 72 50 (68%) 24 (32%) High n = 68 24 (36%) 42 (64%) o v e r a l l 66% 92 Table 26 C l a s s i f i c a t i o n Results of Low and High T o t a l F a c i a l Movement to Heel Lance Predicted Group Membership Actual Group Membership Low Face Action High Face Action Low High n = 52 n = 88 39 25 (75%) (28%) 13 63 (25%) (72%) Overa l l 73% 93 The relationship between state and facial action was analysed earlier. Slightly higher gestational age was associated with more facial activity. To tentatively explore relationships of medication and delivery mode to face action, mean total face movement to heel-lance for each obstetric medication category is given in Table 27, and mean face movement by delivery mode in Table 28. No medication, or epidural alone, appeared to be related to less facial action than either combinations of epidural with other drugs or general anesthetic. Spontaneous vaginal delivery and planned cesareans showed slightly less facial action than the possibly more 'stressful' forceps or emergency cesarean categories. However, medication and delivery mode were confounded to some degree as the more 'stressful' deliveries also received higher levels of medication. Type of obstetric medication was related to mode 2 of delivery, X (36) = 154.07, p= .0001. All those who received general anaesthesia were cesarean deliveries. Of 20 who received epidurals only, 10 were planned cesareans, whereas half the mid or high forceps deliveries had received nitrous oxide in oxygen by mask or a local in addition to epidural. Frequency of the ten categories of obstetric medication across mode of delivery is given in Appendix M. In order to explore these findings further the relationship between obstetric medication and sleep/wake state was examined. Although there was no significant pattern to the frequency of specific maternal medication across sleep state (Appendix N), it was noteworthy that only one of the nine babies whose mother had received a narcotic alone or with other medication was awake. A higher than expected proportion of the general anesthesia group were in active sleep. 94 Table 27 Mean Tot a l Pace Movement to Heel-Lance by Type of Obstetric Medication Obstetric Medication M SD None 17 23.47 3.14 50% N2/0 by mask 7 24.43 5.83 Local 17 25.34 4.62 N 20 + Local 27 25.85 4.52 Epidural 20 22.95 5.67 Epidural + N 20 or l o c a l 23 25.22 2.95 Epidural + N 20 + l o c a l 8 26.87 2.75 Narcotic + N2O or l o c a l or both 3 25.67 4.73 Epidural + Narcotic 6 26.67 1.86 General 12 26.17 3.13 Total 140 Table 28 Mean T o t a l Amount of Face Movement to Heel Lance by Delivery Mode Delivery Mode Planned Emergency Spontaneous Cesarian Cesarian Vaginal Forceps Section Section 24.93 25.82 24.09 25.00 4.22 3.63 5.58 2.97 72 34 21 13 96 To clarify whether the significant relationship between facial action and obstetric medication may have been due to perhaps a wider range or duration of drugs administered for forceps and emergency cesarean deliveries, or to unknown interactions with delivery mode, the discriminant analysis was repeated using data of the 71 infants of spontaneous vaginal deliveries only, none of whom had narcotics or general anesthesia. Obstetric medication remained significant, F^l,63) = 1^46, p=r.001. Since these relationships reported between perinatal variables and pain expression were based, on non-random assignment of subjects to groups, it appeared very important to consider other variables which may be related to delivery mode and obstetric medication. 2 Mode of delivery was significantly related to parity "X. (9) = 34.27, pss.0001, as shown in Table 29. There were significantly more than expected forceps deliveries and emergency cesarean deliveries for first time mothers than for women who had undergone childbirth previously. Conversely of the 24 women having their third or fourth baby, 71% had spontaneous vaginal deliveries, and none were emergency cesarean. Pearson correlations among several perinatal variables were calculated (Table 30). Type, dosage and number of drug administrations during labor and delivery were significantly related to duration of labor and parity. The infant's head circumference was significantly correlated with obstetric medication variables as well. 97 Race It had been observed during data collection that incidence of cesarean delivery mode and type of feeding may differ by race. Chi-square analyses showed high significant differences between races in type of feeding (Table 31), with 84% of whites breast feeding compared with only 39% of Orientals. Incidence of vaginal and cesarean section delivery did not differ significantly by race (Table 32) but there was a slight trend toward more caesarian deliveries for non-whites. 98 Table 29 P a r i t y and Mode of D e l i v e r y (Frequency, Column Percentage, and Adjusted Residuals) P a r i t y Spontaneous Vaginal n = 72 28 ' 40.0% -2.7 27 58.7% 1.2 14 82.4% 2.7 3 42.9% - c5 Forceps n = 34 27 38.6% 3.9 5 10.9% -2.6 0 -2.5 2 28.6% .3 Planned Cesarian Section n = 21 4 5.7% -3.1 12 26.1% 2.6 3 17.6% .3 2 28.6% 1.0 Emergency Cesarian Section n = 13 11 15.7% 2.6 2 4.3% -1.4 0 -1.4 - .9 70 46 17 99 Table 30 Co r r e l a t i o n s Between O b s t e t r i c Medication and P e r i n a t a l / I n f a n t V a r i a b l e s Duration of B i r t h Head Labour Weight Circum. Length P a r i t y Type of Maternal Medication .51**2 ,07 2 .36** 2 -.02 2 -.55** 2 Duration of Drug Admin. .59**1 •03 1 .21 1 -.02 1 -.57** 2 Number of Drugs Administered ,55** 2 ,07- .27* 2 .OO2 -.57** 2 P a r i t y -.53** 2 .06 2 -,06 2 .05' Maternal Age -.01 J .041 ,01J ,14J ,26*-* p ^  .01 ** p i .001 * Pearson product moment 2 Spearman rho 100 Table 31 Method of Feeding by Race White O r i e n t a l Other Breast 75 (84%) 15 (39%) 8 (62%) Bot t l e 14 (16%) 23 (61%) 5 (38%) Total 89 38 13 /140 •X2 = 25.93 p = .0001 101 Table 32 Incidence of Vaginal and Cesarian Section Delivery by Race Vaginal Cesarian Section Whites 71 (80%) 18 (20%) n = 89 1.5 -1.5 Non-Whites 35 (69%) 16 (31%) n - 51 -1.5 1.5 •X2 = 2.19 n.s. n = 140 102 DISCUSSION There were two aims in this study : first, to examine neonatal reaction to heel-lance in relation to behavioral state and sex ; second, to explore effects of perinatal factors on pain behavior. State From adult pain theories two alternative hypotheses regarding the relationship of behavioral state to pain expression were proposed : From specificity theory : neonatal pain behavior would reflect degree of tissue damage only, so no differences would be expected as a function of behavioral state. From gate-control theory : pain behavior would be modified by behavioral state at the time of stimulation. Facial action varied significantly in relation to state, which was consistent with gate-control theory. Following heel-lance, significantly less overall facial activity was apparent for the quiet sleep infants than for the awake/alert states. Through fine-grained analysis of individual facial components it became clear that the significant differences due to state mainly reflected variation in occurrence of vertical stretch mouth and taut tongue. Awake/alert infants showed the highest frequency for both those facial actions. 103 Of three alternative hypotheses which were proposed regarding the nature of response variation across state, there was no support for the sleep versus wake, or the basic rest activity cycle patterns. The third alternative, namely that quiet/awake infants would respond differently than the other three groups, received the most support, in that analysis of individual facial actions showed the highest frequency of response occurred in this state. Cry latency, but not speed of facial response, was related to state. The quiet/sleep infants took the longest time to respond with cry. The hypothesis that fundamental frequency of the first pain cry would vary as a function of facial reactivity was not supported. The results did not support Levine and Gordon's (1982) proposal that pain induced vocalization would be a sensitive indicator of infant pain perception, insofar as pain experience differed across sleep/waking states. Levine and Gordon's position was mainly derived from studies of vocalization in infrahuman species, in which distinct 'calls' have been identified. Fundamental frequency of pain cry has been found to be related to the health status of neonates, and has been viewed as indicative of recovery capacity following stimulation (Lester and Zeskind, 1982). It was hypothesized that in those circumstances where more facial activity was evident, fundamental frequency would vary accordingly. The results suggested that the view from studies of 'at risk' infants, suggesting fundamental frequency reflects capacity for response modulation, does not generalize to healthy infants across sleep/waking states. 104 The state dependency of facial movement to heel-rubbing was striking. The quiet/awake infants responded immediately with facial action. On every facial action (except lips apart) they were significantly above the other states in the first heel-rub segment. The quiet/sleep infants were consistently the lowest in facial response at this point. This supports Brazelton's view (1973) of quiet wakefulness as a state of readiness to receive environmental input. It was initially not surprising that face movement to heel-rub, i.e. pre-pain, was related to state, as state was operationally defined using face movement. However, it was very interesting that the most face response to initial disturbance was not by infants in states where the faces were moving already, but rather, a group whose faces had been still prior to stimulation. All babies had their heel swabbed before the prick. Behavioural changes were already taking place at this point. The taut tongue and vertical stretch mouth responses must be particularly robust to continue to reflect prior state considering the intervening event of heel-rub. There was a noticeable element of extra muscular tension that could be interpreted as stress evident when these two responses were observed. In adapting the Baby FACS concept, Oster (1978) emphasized that FACS is purely a method of facial analysis, which makes no assumptions about 'natural', 'meaningful' or 'emotional' correlates of facial muscle actions. The value of this system is that it permits an objective comprehensive description of the behavior observed. 105 It is interesting to note that the findings are in line with the view of investigators who have emphasized that sleep/awake states do not reflect an underlying continuum of arousal (Emde and Robinson, 1979 ; Prechtl, 1974). If this were the case then active/awake infants would have shown the most face movement. In fact, considering that active/awake babies were already moving their faces prior to intervention, it is amazing that the quite/awake ones significantly surpassed them and moreover did so at once. State related facial reaction was actually far more apparent to heel-rub than to heel-lance. This is consistent with the findings of Vlach, Bernuth and Prechtl (1969) in their study of state dependency of tactile reflexes in newborns. States 1, 2 and 3 (but not 4) were compared. Gentle stroking of the extremities produced the least response in quiet/sleep and the most in quiet/wakefulness, just as heel-rubbing did in the present study. In contrast, limb girdle reflexes did not show state related effects in the Vlach et al. (1969) study. They interpreted the absence of state dependency for limb girdle reflexes as having a protective role. They speculated that biological significance lies in the protective nature of these reflexes, and concluded that the fact that they occurred regardless of ongoing functional state was because it is vital for survival that they do not disappear. A parallel situation may exist between noninvasive versus invasive responses to limb extremities. In other words, heel-lance was far less state dependent than heel-rub perhaps because invasive procedure elicits a defensive reaction. 106 It was interesting that not all the babies cried to heel-lance. There were 7 out of 147 who did not do so. Although these infants were excluded from the study due to their failure to cry, this implies that not all healthy newborns react vigorously to invasive procedure. Rather than being exceptional, analysis of concomitant factors suggested that their low level of response probably reflected one end of the normal reaction distribution rather than some unusual response. Further, most of the non-crying babies had been aroused from quiet sleep. As the results showed infants in quiet/sleep displayed the least facial reactivity to heel-lance. The low responsivity of those infants who did not cry was probably a reflection to some extent of their sleep/waking state. Oral functioning has been viewed as the highest level of co-ordinated functioning of the newborn due to the fact that it is so intrinsic to survival (Stratton, 1982). It was interesting that the facial features which persisted in significantly differentiating responses between states were both oral : i.e. vertical stretch mouth and taut tongue. The significance of non-nutritive oral behavior is unclear, however it has been regarded as a sensitive indicator of arousal (Korner, 1973). It is particularly interesting to find behavioral state variation to pain in the newborn, because at this age one has the opportunity to observe patterns of reaction prior to the possibility of learned behaviors in the social milieu. At this age, in the absence of learned reaction patterns, most pain behavior would be very likely to be a direct function of degree of tissue 107 damage. However, the results of this study demonstrated that even at this very early point in development, ongoing functional activity modified pain reaction. These findings are interpreted as support for the fundamental role of subcortical modulating systems in pain experience, rather than a secondary reaction to it. According to gate control theory, pain may be modified by cortical or subcortical descending systems so higher cortical involvement is not necessarily implied. There is no doubt that multiple feedback networks integrating cortical and subcortical activity of the nervous system are the least advanced at birth. However, as Bronson (1982) points out in discussion of the psychobiology of the human newborn, seemingly similar behavioral effects may be processed quite differently at different ages. Based on these results, it seems to be fruitful to conceptualize study of neonatal pain within the framework of the adult theories. Bennett and Bowyer (1982) appeared to be implicitly assuming a specificity theory model when they maintain that due to limited cortical development, neonates cannot 'experience' pain. In other words, that a 'pain centre' in the cortex is required. Certainly examination of functional capacities needs to be carried out with reference to the changing form of the mediating networks (Bronson, 1982). However, even if neonatal response to nociception is largely mediated subcortically, this does not imply that it is therefore somehow less important. 108 Sex Early studies of sex differences in tactile responsivity either found girls more reactive, or no differences (Maccoby and 3acklin, 1974 ; Owens and Todt, 1984). Due to the lack of control of circumcision status, the studies which have found sex differences have been viewed as suspect (Richards, Bernal and Brackbill, 1980). However, Brackbill and Shroder (1980) have concluded subsequently that there is little evidence for behavioral differences between males and female neonates, regardless of circumcision status. There is one study of response to pinpricks (as part of the Brazelton scale) which reported that although boys and girls showed no differences in rates of 'expressivity1, boys showed more immediate cry response (Field et al., 1982). This is analogous to the findings in the present study. No sex differences were observed in the amount of facial activity, or in the fundamental frequency or duration of the first cry. However, boys showed faster response speed for facial movement and cry, and greater number of cry cycles than girls. Careful attention to sampling and stimulus aspects of the studies may explain some but not all of these apparent discrepancies. Sex differences, if any, appear to depend on stimulus charactertistics or other parameters not yet delineated or understood. 109 Technician Effects Highly significant differences in infant facial activity during the latter phase of blood collection, as a function of laboratory technician, were observed. Although this portion of the procedure was not used in the data analyses, it was of interest in itself as an indication of neonates surprising sensitivity to variations in handling. One difference in blood collection technique between lab technicians was observed during the squeezing phase, namely the degree to which the technician 'fought' agains the infants foot withdrawal reflex. Several heel squeezes were required to collect sufficient blood. Following each squeeze the infant would pull back the affected leg, then released it. Some technicians waited momentarily while the leg was pulled in, and squeezed when the baby extended the leg again. Others appeared to ignore the infants' cyclical response pattern and squeezed regardless of leg position. As technician techniques were not recorded, this was only an incidental observation. However, in the light of the prevalent interest in the importance of caretaker-infant transactional behavior, and contingent responding to infant cues (Goldberg, 1982 ; Tronick, 1982 ; Brazelton, 1983) the possibility that this interaction cycle may have modified neonates facial behavior certainly warrants systematic study. 110 Perinatal and Maternal Correlates of Neonatal Pain Expression Birthweight and Cry Fundamental frequency, at the beginning of the first cry to heel-lance, was significantly related to birthweight. One possible interpretation, in the light of the infant cry literature which consistently has reported higher fundamental frequency in groups of infants with wide-ranging clinical problems (reviewed by Lester and Zeskind, 1982), is that perhaps some of the lighter infants in the present study were somehow marginally different in a clinical sense from the heavier infants. Maternal smoking was shown to be a possible mediating factor. Lower birthweight has been found to be related to smoking. The U.S. Department of Health, Education, and Welfare (1979) reported 45 studies demonstrating this relationship. Placental abnormalities in smokers, which presumably result in decreased nutrition and pooer oxygen supplies to the fetus, (Meyer et al. 1976 and Naeye, 1979), and low maternal weight gain (Luke et al., 1981) have been implicated as possible causal factors. When infants of mothers who had smoked during pregnancy were excluded from analysis, birthweight remained significantly associated with level of fundamental frequency. Studies of infant cry in relation to central nervous system integrity have found higher fundamental frequencies in groups of "abnormal" infants covering a wide range of clinical problems. Formulations of possible sympathetic and central nervous system mechanisms reflected in fundamental frequency changes, via control of vocal fold Il l vibration, have been proposed (Tenold, Crowell, Jones, Daniel, McPherson and Popper, 1974 ; Golub and Corwin, 1982 ; Lester and Zeskind, 1982). Due to the results of the present study, in which fundamental frequency was related to birthweight in full term infants, the possibility arises that some factor associated with infant size, rather than 'health status' may be reflected in the cry literature to date. The cry studies have compared groups of clincially abnormal neonates with groups of clinically normal. It appears likely that, on average, infants in the abnormal groups would be smaller (Avery, 1981). This would introduce a systematic confound, namely infant size, which has not been addressed in drawing conclusions regarding the reasons for the differences in fundamental frequency. Size of the vocal cords is related to the level of fundamental frequency with shorter vocal cords resulting in a higher level (P. Lieberman, personal communication, May, 1985). Although the cry literature has produced results which appear promising in evaluation of newborns, it is premature to conclude to what extent only biobehavioral shifts in central nervous system organization are being reflected. Studies are needed in which samples of infants with specific neonatal problems are matched for birthweight with infants in the healthy sample, to evaluate cry as a clinical indicator while controlling for infant size. 112 Obstetric Medication It is essential to consider the finding that differences in amount of facial movement following heel-lance were related to delivery mode and obstetric medication, within the broader context of obstetric management. The increase during the 1970's in cesarean delivery rate is mainly attributable to the desire to improve chances for optimal outcome of the infant. Advances in anesthesia and surgical techniques made cesarean delivery extremely safe for the mother, thereby enabling physicians to elect surgical delivery readily to benefit the infant. This has been encouraged by clinical findings of better infant outcome with cesarean deliveries when there are obstetric complications, for example, abnormal presentation, fetal distress, prolonged labor. Thus a living healthy baby is produced whereas difficult vaginal delivery in similar circumstances may have resulted in death or permanent injury to the infant (Affonso, 1981). Moreover, these are correlational findings of relationship, therefore, one must be very cautious in interpretation. There are other factors associated with the delivery mode and obstetric medication namely the physician's reason for intervention, and the mother's physiologic reactions to labor pain and distress. The present finding of a significant association between length of labor and obstetric medication is consistent with other studies (Nuite, 1976). The conclusion usually has been that the longer the labor, the greater the pain and need for pain relieving drugs. Brackbill (1979) suggested that as 113 analgesics may decrease uterine activity (Petrie et al., 1976), and prolong labor (Nuite, 1976), drug administration may initiate longer labor rather than vice versa. The finding in the present study of a statistically significant association between infant head circumference and obstetric medication suggested that the more difficult labors and/or known cephalo-pelvic disproportion leading to planned cesarean delivery, were related to increased medication for this sample. Brackbill (1979) raised the problem of 'standing orders' observing for example, that in some obstetrical units particularly in the United Kingdom and Africa, without exception every entering patient received 100 mg of demerol. The present study took place in a university affiliated teaching hospital designed for tertiary care. There was a low incidence of the use of narcotics. During labor no tranquilizers were given ; nitrous oxide in oxygen was the commonly used medication for pain relief in labor. If indeed it were the mothers who were in most distress who were given most medication, as appears to likely be the case in this sample, then the mother's physiological reactions to pain may have been confounded with medication effects. The reason for examining obstetric medication and delivery mode in this study was to evaluate the need for care in sample selection for research in neonatal pain behavior. Concern that obstetric medication may be important in studying pain related behavior in the first few days of life was raised in 1959 by Lipsitt and Levy, but has received no mention in the literature since 114 that time. The few studies of the neonatal pain behavior have treated neonates as if they were an homogeneous group. In fact, the major review article in this area (Owens, 1984) does not even mention sample selection. The present findings suggest facial movement does reflect obstetric variables. Thus, for fine grained analysis of response patterns to nociceptive input, specifying sample selection is important. This study was not designed to evaluate behavoural variation in relation to maternal medication. There were a wide variety of drugs, dosages and combinations. This is a highly complex topic pharmacologically, and it has been emphasized that the degree to which the neonate may show subtle behavoural variation is linked to the expertise of the anesthetist in drug selection, administration, dosage and timing, amongst other considerations (Bonica, 1980). Even within what may appear to be a homogeneous category, namely planned cesarean deliveries with an epidural, the drug selected is important. For example, bupivicaine has been found to exert less effect on muscle tone and neurobehavior than mepivacaine or lidocaine (Amercian Academy of Pediatrics Committee on Drugs, 1978). The depressant effect of narcotics and barbituates, which may persist for as long as two to four days after birth, is well documented (Bonica, 1980). Thus, the possibility that the medication effect found on total facial movement after pain may have reflected only those infants whose mothers had received general anesthetics or narcotics was explored. It was evident, 115 however, that obstetric medication was highly related to total facial movement, even when only spontaneous vaginal deliveries were analysed, with no narcotics administered. Bonica (1980) describes circumstances in which fetal depression from local anesthetics can occur. The really surprising finding regarding medication was that total facial movement was higher rather than lower as compared with infants from nonmedicated deliveries. It was expected that pharmacologically 'depressed' infants would show less, not more reaction. Moreover, it was fascinating that these behavioral differences in facial movement were only significant to heel-lance, not heel-rub. There has been some indication in animal research (Levine and Mullins, 1968 ; Denenberg and Zarrow, 1971) that experimental animals 'stressed' in infancy were significantly less responsive than controls to mild stimuli but significantly more responsive to noxious, stressful or potentially dangerous stimuli. From a bioevolutionary perspective, ability to modulate respond to external stimulation would appear important for survival. The energy reserves of a young organism are needed for growth and temperature control. Over-exertion to external demands would appear to be energy draining. This perspective has been discussed with premature infants who in some cases hyper-react to handling. Healthy newborns show excellent adaptation and recovery to external demands. It may be that the higher reactivity of the more stressed infants reflects reduced ability to 'shut down' systems following noxious intervention. 116 Another possibility, which is high speculative, is that a self-selection process may occur. Perhaps the less reactive mothers tend to get no medication and that their children are in some way expressively less reactive newborns. This may be mediated by genetic temperamental factors or by maternal physiologic factors during labor and delivery. Another possibility is that endogenous endorphins may be produced by nonmedicated mothers, whereas in medicated mothers as their pain is controlled externally, endorphins may not be released (C. Bradley, personal communication, April, 1985). Endorphins may be passed on to the infant, thereby modulating pain reception and expression. It is not known at what point in pre- or postnatal development capacity to produce endorphins becomes present. The results of more facial movement to pain on the part of infants who had experienced more 'stressful' deliveries, often involving higher levels of maternal medication, suggested they may have been more irritable and perhaps less able to modulate responses to nociceptive stimulation. Analyses of facial action to specific drugs requires very carefully controlled studies which would necessitate separate analyses for different delivery modes, control of dosage, number of drug administrations and drug combinations as well as levels of maternal endorphins and monitoring of maternal and infant physiological as well as behavioral indices. The complex interplay of factors which potentially affect neonatal pain expression highlights the need for more interdisciplinary research in this area. 117 Race Cultural variation in pain expression has been of interest to researchers (Craig, 1980). Freedman (1976) reported differences between Chinese-American and European-American newborns in behavior relating to defensive movements and excitability. For example, when a cloth was placed on the infant's face the European-American infants immediately struggled whereas "the typical Chinese-American infant lay impassively, exhibiting few overt motor responses" (p. 38). The Chinese-American newborns were rated in general as less perturbable and habituating more readily to external change. The conclusion was drawn that these differences reflected genotypic variation. In the present study, significant differences were found in feeding practise between Oriental-Canadian and White-Canadian mothers, with more of the former group bottle feeding and the latter group breast feeding. There was also a trend toward more cesarean deliveries in the Oriental group. Thus it appears extreme caution should be exercised in drawing conclusions regarding the causes of behavioral differences between cultural groups since cross-cultural research is correlational by its nature, in that random assignment is precluded. The present findings illustrate early sources of cultural variation other than genetic variation, namely maternal variables, which may be related to apparent differences in infant behavior. 118 Conclusions Infant studies in the last two decades have shown a far greater degree of competence and behavioral organization in young infants and most recently also in newborns, than previously thought possible (Stratton, 1982). Capacities of the neonate have been demonstrated, for example, in neuro-behavoural organization (Brazelton, 1983), perceptual functioning in vision and audition, taste and smell (Rosenblith and Sims, Knight, 1985), learning (Lipsitt, 1982), and imitation (Meltzoff and Moore, 1977 ; Field et al., 1982). To the extent that tests of learning are in fact also tests of memory, short-term retention has been evident in studies of habituation and conditioning (Fagan, 1982 ; Rowee-Collier and Lipsitt, 1982). The present study demonstrates a cohesive response pattern to invasive stimulation, which is consistent with the response capacities currently recognized in the other major senses modalities. The results of this study are consistent with gate control theory rather than specificity theory of pain. This conclusion was based on the finding that facial action varied depending on the infant's prior sleep/waking state. This indicated that pain expression was a function of ongoing functional state in combination with tissue damage, rather than solely related to degree of tissue damage. Whereas within a specificity framework, the possibility for social or behavioral modifiers of pain expression exists, it is seen as a secondary psychological outcome. Due to the limited opportunity and capacity for 119 neonates to have learned or developed secondary responses to nociceptive input, the differences in facial pain expression across state were interpreted as reflecting differences in the pain sensation itself. In other words, state variation affected pain perception. Of the alternatives delineated as to how infant responsivity may vary across state, Brazelton's view (1973) of the alert/awake state as providing optimal orientation or receptivity to environmental input was supported. It was hypothesized that fundamental frequency of the first pain cry to heel-lance would be higher in whatever state or states infant's exhibited the most facial activity as an indicator of nervous system 'stress1. This was not supported. Rather, it appears that whereas unwell infants may have difficulty modulating their responses to stimulation, and this may be reflected in higher fundamental frequency of initial pain cry, healthy neonates have excellent adaptive capacities as part of their biological survival mechanisms. Moreover, differential infant size between unwell and healthy samples may be a systematic confound in the cry literature which calls into question current formulations regarding fundamental frequency of pain cry in neonates. There were very few sex differences in this study. Speed of response to both cry and facial action following heel-lance was shorter for boys than girls, and the number of cry cycles was greater for the boys than girls. There was no apparent explanation for this finding, and the existing studies have been inconclusive. 120 One of the main contributions of this study was in the area of neonatal pain measurement. Discrete facial action appeared to be a reliable and useful measure of pain expression. Moreover, two aspects were isolated as particularly sensitive to state variation and worthy of further study in their own right, namely taut tongue and vertical stretch mouth. Of course tongue action is actually quite another behavioral feature which is not 'facial'. However, the use of a coding system which did not limit the investigator to preselected emotional categories actually promoted the discovery of 'new' information, as any visually detectable category of movement on the video frame qualified for possible attention as a 'facial variable'. Perinatal events did affect pain expression to some degree. It was beyond the bounds of this study to specifically delineate these factors. However, the findings suggested sample selection for neonatal pain study should include attention to birth factors in addition to infant measures which define the neonate as 'healthy'. In conclusion, as ongoing functional state modified pain behavior, further study of pain expression using the gate control construct that ongoing events alter pain perception, may be fruitful. Within this model it would be expected that other factors either intrinsic to the individual neonate such as temperamental variation, or extrinsic such as soothing or swaddling during the stimulation, may also affect pain behavior. The complexities of the obstetric and maternal interrelationships with neonatal pain expression suggest that interdisciplinary research may be most appropirate for future progress in this area. 121 Distinguishing between 'pain' and 'reflexive' responding has been an issue in considering the newborn's capacity to respond to invasive procedures (Bennett and Bowyer, 1982). 'Pain' is a construct inferred from observation of tissue damage and behavior. Following heel-lance, in this study, neonates early in the second day of life showed a constellation of facial changes, namely eye squeeze, brow contraction, naso-labial furrow and open mouth, together with cry response, significantly different from the amount of facial action to heel-rub. This reaction pattern may be described operationally as 'pain' expression. The issue of whether newborns are capable of 'experiencing' pain is quite fruitless. Doubtless the reaction pattern observed is a complex reflex response. In bioevolutionary terms, for species survival, neonatal sensitivity to pain and a highly developed means of communicating pain to the caretaker appears fundamental. 122 R E F E R E N C E S Acredolo, L.P. and Hake, 3.L. (1982). Infant perception. In B.B. Wolman and G. Strieker (Eds.), Handbook of Developmental Psychology. Englewood Cliffs, N.3. : Prentice-Hall. Affonso, D.D. (1981). Impact or Cesarean Childbirth. Philadelphia : Davis. Anders, T.F. and Chalemain, R.3. (1974). The effects of circumcision on sleep-wake states in human neonates. Psychosomatic Medicine, 36, 174-179. Anokhin, P.K. (1964). Systemogenesis as a general regulator of brain development. Progress Brain Research, 9, 54-86. Apley, 3. (1975). The Child With Abdominal Pains. London : Blackwell. Ashby, W.R. . (1956). An Introduction to Cybernetics. Chapman and Hall : London. Ashton, R. (1971). Behavioural sleep cycles in the human newborn, Child  Development, 42, 2098-2100. Ashton, R. (1973). The state variable in neonatal research, Merrill Palmer  Quarterly, 19, 3-20. Atkinson, 3. and Braddick, O. (1982). Sensory and perceptual capacities of the neonate. In P. Stratton (Ed.), Psychobiology of the Human Newborn. Wiley : New York. Avery, G.B. (Ed.), Neonatology : Pathophysiology and Management of the  Newborn. Philadelphia : Lippencott. 123 Barr, R.G. (1983). Pain tolerance and developmental change in pain perception. In M.D. Levine et al. (Eds.), Developmental - Behavioral  Pediatrics. Philadelphia : W.B. Saunders. Barr, R.G. and Feuerstein, M. (1983). Recurrent abdominal pains in children : How appropriate are our usual clincial assumptions? In P. Firestone and P. McGrath (Eds.), Pediatric Behavioural Medicine. New York : Springer-Verlag. Bell, R.Q. and Costello, N.S. (1964). Three tests for sex differences in tactile sensitivity in the newborn. Biological Neonatorum, 7, 335-347. Bell, R.Q., Weller, G.M. and Waldrop, M.F. (1971). Newborn and preschooler : Organization of behaviour and relation between periods. Monographs of  the Society for Research in Child Development, 36, series no. 142. Bennett, E.J. and Bowyer, D.E. (1982). Principles of Pediatric Anesthesia. Springfield, 111. : C C . Thomas. Berg, W.K. and Berg, K.M. (1979). Psychophysiological development in infancy : State, sensory function, and attention. In Osofsky, J.D. Handbook of Infant Development. N.Y. : Wiley. Birns, B. (1965). Individual differences in human neonates' responses to stimulation. Child Development, 36, pp. 249-256. Blass, E.M., Ganchrow, J.R. and Steiner, J.E. (1985). Classical conditioning in newborn humans 2-48 hours of age. Infant Behavior and Development, 7, 223-235. Bondy, A.S. ( 1980 ). Infancy. In S. Gabel and M.T. Erickson (Eds.), Child  Development and Developmental Disabilities.. Boston : Little, Brown and Co. Bonica, J.J. and V. Ventafridda, (Eds.) (1979). Advances in Pain Research and  Therapy. New York : Raven Press. Bonica, J.J. (1980). Obstetric Analgesia and Anesthesia. Amsterdam : World Federation of Societies of Anaesthesiologists. Borden, G.3. and Harris, K.S. (1980). Speech Science Primer. London : Williams and Wilkins. Boring, E.G. (1942). Sensation and Perception in the History of Experimental  Psychology. N.Y. : Appleton-Century-Crofts. Brackbill, Y. (1979). Obstetrical medication and infant behaviour. In 3.D. Osofsky (Ed.), Handbook of Infant Development. Wiley and Sons : New York. Brackbill, Y. and Schroder, K. (1980). Circumcision, gender differences, and neonatal behaviour : An update. Developmental Psychobiology. 13 (6), 607-614. Brazelton, T.B. (1970). Effect of prenatal drugs on the behaviour of the neonate. American 3ournal of Psychiatry, 126, 1261-1266. Brazelton, T.B. (1973). Neonatal Behavioural Assessment Scale. Clinics in Developmental Medicine No. 50, Spastics International Medical Publications : London. Brazelton, T.B. (1981). Behavioural competence of the newborn infant. In G.B. Avery (Ed.), Neonatology : Pathosphysiology and Management of the  Newborn. Philadelphia : Lippincott. Brazelton, T.B. (1983). Precursors for the development of emotions in early infancy. In Plutchik, R. and Kellerman, H. (Eds.), Emotion : Theory,  Research and Experience. Vol. 2. Academic Press. Bronson, G.W. (1982). Structure, Status and Characteristics of the nervous system at birth. In P. Stratton (Ed.), Psychobiology of the Human  Newborn. New York : Wiley and Sons. Carmichael, L. (Ed.) (1954). Manual of Child Psychology. 2nd. Ed. New York : Wiley. Committee on Drugs, American Adcademy of Pediatrics : Effects of medication during labor and delivery on infant outcome. Pediatrics, 62, 402, 1978. Craig, K.D. (1978). Social modeling influences on pain. In R.A. Sternback (Ed.), The Psychology of Pain. New York : Raven Press. 125 Craig, K.D. (1980). Ontogenetic and cultural influences on the expression of pain in man. In H.W. Kosterlitz and L.Y. Terenius (Eds.), Pain and  Society. Weinhem : Verlag Chemie GmbH. Craig, K.D. (1983). Modeling and social learning factors in chronic pain. In J.B. Bonica et al. (Eds.), Advances in Pain Research and Therapy. New York : Raven Press. Craig, K.D., McMahon, R.J., Morison, J.D. and Zaskow, C. (1984). Developmental changes in infant pain expression during immunization injections. Social Science and Medicine, 19(12), 1331-1337. Craig, K.D. and Prkachin, K.M. (1978). Social modeling influences on sensory decision theory and psychophysiological indexes of pain. Journal of  Personality and Social Psychology, 36, 805-815. Craig, K.D. and Prkachin, K.M. (In press). Nonverbal expression of pain. In R. Melzack (Ed.), Pain Measurement and Assessment. New York : Raven Press. Crelin, E.S. (1981). Development of the musculoskeletal system. CIBA Clincial Symposia, 33 (1). Crowell, D.H., Davis, CM., Chun, B.J. and Spellacy, F.J. (1965). Galvanic skin reflex in newborn humans. Science, 148, 1108-1110. Darwin, C (1872). The Expression of The Emotions in Man and Animals. London : John Murray. Denenberg, V.H. and Zarrow, M.X. (1971). Effects of handling in infancy upon adult behavior and adrenocortical activity : Suggestions for a neuroendocrine mechanism. In D.N. Walcher and D.L. Peters (Eds.), Early Childhood : The Development of Self-Regulatory Mechanisms. New York : Academic. Dittrichova, J. and Paul, K. (1974). Responsivity in newborns during sleep. Activitas Nervosa Superior, 16, 112-113. Dixon, S., Snyder, J., Holve, R., and Bromberger, P. (1984). Behavioural effects of circumcision with and without anesthesia. Developmental and -. Behavioural Pediatrics, 5(5), 246-250. 126 Ekman, P. and Oster, H. (1979). Facial expression of emotion. Annual Review  of Psychology, 30, 527-554. Ekman, P. and Friesen, W.V. (1978). The Facial Action Coding System (FACS). Palo Alto, California : Consulting Psychologists Press. Eland, J.M. and Anderson, J.E. (1977). The experience of pain in children. In A.K. Jacox (Ed.), Pain. Boston : Little, Brown and Co. Emde, R.N., Harmon, R.J., Metcalf, D.R., Koenig, K.L., and Wagonfeld, S. (1971). Stress and neonatal sleep. Psychosomatic Medicine, 33, 491-497. Emde, R.N. and Robinson, J. (1979). The first two months : Recent research in developmental psychobiology and the changing view of the newborn. In J.D. Noshpitz (Ed.), Basic Handbook of Child Psychiatry, Vol. 1. New York : Basic Books. Fagan, J.F. Ill (1982). Infant memory. In T.M. Field, A. Huston, H.C. Quay, L. Troll, G.E. Finley (Eds.), Review of Human Development. New York : Wiley. Field, T. (1982). Individual differences in the expressivity of neonates and young infants. In R.S. Feldman (Ed.) Development of Nonverbal  Behaviour in Children. New York : Springer-Verlag. Field, T.M., Woodson, R., Greenberg, R., and Cohen, D. (1982). Discrimination and imitation of facial expressions by neonates. Science, 218, 179-181. Fisichelli, V.R., Karelitz, S., Fisichelli, R.M., and Cooper, J. (1974). The course of induced crying activity in the first year of life. Paediatric  Research, 8, 921-28 Fitzgerald, M., Beal, J.A., Gebhardt, G.F. and Semba, E. (1984). The development of pain pathways and mechanisms. Pain, Supplement No. 2, 193. 127 Freedman, D.G. (1976). Infancy, biology and culture. In Lipsitt, L.P., Developmental Psychobiology. N.Y. : Wiley. Goldman, P.S. and Nauta, W.J.H. (1977). Columnar distribution of cortico-cortical fibers in the frontal association, limbic and motor cortex of the developing rhesus monkey. Brain Research, 122, 393-413. Golub, H.L. and Corwin, MJ. (1982). Infant cry : A clue to diagnosis. Pediatrics, 69, 197-201. Gottlieb, G. (1976). Conceptions of prenatal development : Behavioural embryology. Psychological Review, 83, 215-234. Graham, F.K. and Jackson, J. (1970). Arousal systems and infant heart-rate responses. In H.W. Reese and L.P. Lipsitt (Eds.) Advances in Child  Development and Behaviour, Vol. 5, 60-111. Academic Press : New York. Gullickson, G.R. and Crowell, D.H. (1964). Neonatal habituation to electrotactual stimulation, journal of Experimental Child Psychology, 1, 388-396. Gunnar, M.R., Malone, S. and Fisch, R. (1984). Deep sleep and levels of plasma C o r t i s o l during recovery from routine circumcision in human newborns. Paper presented at the Fourth International Conference on Infant Studies. New York, April 5-8. Guyton, A.C. (1982). Human Physiology and Mechanisms of Disease. Toronto : Academic Press. Haberman, S.J. (1978). Analysis of Qualitative Data. Vol. 1. N.Y. : Academic Press. Harpin, V.A. and Rutter, N. (1982). Development of emotional sweating in the newborn infant. Archives of Disease in Childhood, 57, 691-695. Haslam, D.R. (1969). Age and the perception of pain. Psychonomic Science, 15, 86. 128 Hutt, S.J., Lenard, H.G. and Prechtl, H.F.R. (1969). Psychophysiological studies in newborn infants. In L.P. Lipsitt and H.W. Reese (Eds.), Advances in Child Development and Behaviour, Vol. 4, 127-172. Academic Press : New York. Izard, C.E., Hembree, E.A., Dougherty, L.M. and Spizzirri, C.C. (1983). Changes in facial expressions of 2 to 19 month old infants following acute pain. Developmental Psychology, 19, 418-426. Izard, C E . (1978). On the development of emotions and emotion-cognition relationship in infancy. In M. Lewis and L.A. Rosenblum (Eds.), The  Development of Affect. N.Y. : Plenum Press. Izard, C E . (1982). Measuring Emotions in Infants and Children. London : Cambridge University Press. Izard, C.E., Hembree, L.M., Dougherty, L.M., Spizziri, C.C. (1983). Changes in facial expressions of 2 to I9-month-old infants following acute pain. Developmental Psychology. 19(3), 418-426. Jeans, M.E. (1983). The measurement of pain in children. In R. Melzack (Ed.), Pain Measurement and Assessment. New York: Raven Press. Joffe, R., Bakal, D.A., and Kaganov, J. (1982). A self-observation study of headache symptoms in children. Headache, 22. Johnson, G.G., McGrath, P.J. and Schillinger, J.F. (1983). The evaluation of supplemental regional blocks in children. Unpublished manuscript. Katz, E.R., Sharp, B., Kellerman, J., Marston, A.R., Hershman, J.M., and Siegel, S.E. (1982). - endorphin immunoreactivity and acute behavioural distress in children with leukemia. The Journal of Nervous and Mental  Disease, 170, 72-77. Kauffman, R.E. (1980). Pain. In H.C Shirkey (Ed.), Pediatric Therapy. Toronto : C.V. Mosby. 129 Kleitman, N. (1973). The basic rest-activity cycle in sleep and wakefulness. In U. Juranovich (Ed.), The Nature of Sleep. Stuttgart : Verlag. Korner, A.F. (1969). Neonatal startles, smiles, erections and reflex sucks as related to state, sex and individuality. Child Development, 40, 1039-1053. Korner, A.F. (1972). State as variable, as obstacle, and as mediator of stimulation in infant research. Merrill Palmer Quarterly, j8, 77-94. Korner, A.F. (1973). Sex differences in newborns with special references to differences in the organization oral behavior. Journal of Child Psychology  and Psychiatry. 14,18-29. Korner, A.F. (1983). Individual differences in neonatal activity : Implications for the origins of different coping styles. In J. Call, E. Galenson, and R.L. Tyson, (Eds.), Frontiers of Infant Psychiatry. N.Y. : Basic Books. Korner, A.F., and Thoman, E.B. (1970). Visual alertness as evoked by maternal care. Journal of Experimental Child Psychology, 10, 67-78. Korner, A.F. and Thoman, E.B. (1972). Relative efficacy of contact and vestibular stimulation in soothing neonates. Child Development, 43, 443-453. Lamper, K. and Eisdorfer, J. (1971). Prestimulus activity level and responsivity in the neonate. Child Development, 42, 465-473. Lavori, P.W., Thomas, A.L., Baiiar, J.C. Ill and Polansky, M. (1983). Designs for experiments-parallel comparisons of treatment. The New England  Journal of Medicine, 309 (21), 1291-1299. Lester, B.M., Als, H. and Brazelton, T.B. (1982). Regional obstetric anesthesia and newborn behaviour : A reanalysis toward synergistic effects. Child  Development, 53, 687-692. 130 Lester, B.M. and Zeskind, P.S. (1982). A biobehavioural perspective on crying in early infance. In H.E. Fitzgerald, B.M. Lester and M.W. Yogman (Eds.) Theory and Research in Behavioural Pediatrics, Vol. 1. New York : Plenum. Lester, B.M. (1976). Spectrum analysis of the cry sounds of well-nourished and malnourished infants. Child Development. 47, 237-241. Lester, B.M. (1978). A synergistic process approach to the study of prenatal malnutrition. International Journal of Behavioural Development, i, 393-402. Levine, J.D. and Gordon, N.C. (1982). Pain in prelingual children and its evaluation by pain-induced vocalization. Pain, 14, 85-93. Levine, S. and Mullins, R.F., Jr. (1968). Hormones in infancy. In G. Newton and S. Levine (Eds.), Early Experience and Behavior. Springfield, 111. : Thomas, 1968. Levy, D.M. (1960). The infant's earliest memory of innoculation : A contribution to public health procedures. Journal of Genetic Psychology, 96, 3. Lipsitt, L.P. (1982). Infant learning. In T.M. Field, A. Huston, H.C. Quay, L. Troll, G.E. Finley (Eds.), Review of Human Development. New York : Wiley. Lipsitt, L.P. and Levy, N. (1959). Electrotactual threshold in the neonate. Child Development, 30, 547-554. Lipton, E.L., Steinschneider, A. and Richmond, J.B. (1966). Autonomic function in the neonate. VIII : Maturational changes in cardia control. Child Development, 37, 1-16. Luke, B., Hawkins, M.M. and Petrie, R.H. (1981). Influence of smoking, weight gain, and pregravid weight for height on intrauterine growth. American  Journal of Clinical Nutrition, 34, 1410-1417. 131 Maccoby, E.E. and Jacklin, C.N. (1974). The Psychology of Sex Differences. Stanford University Press, Stanford. Marshall, R.E., Stratton, W.C, Moore, 3. and Boxerman, S.B. (1980). Circumcision : Effects upon newborn behaviour. Infant Behaviour and  Development, 3,1-14. McGraw, M.B. (1941). Neural maturation as exemplified in the changing reaction of the infant to pinprick. Child Development, 9, 31. McKeel, N.L. and Saunders, S.H. (1984). An innovative measurement paradigm for new-born infant pain responses. Paper presented at the Annual Meeting of the Association for Advancement of Behavior Therapy. Philadelphia, November. Meltzoff, A.N. and Moore, M.K. (1977). Imitation of facial and manual gestures by human neonates. Science, 198, 75-78. Melzack, R. (1973). The Puzzle of Pain. Harmondsworth, England : Penguin. Melzack, R. and Wall, P.D. (1982). The Challenge of Pain. New York : Penguin. Melzack, R. and Wall, P.D. (1965). Pain mechanisms : A new theory. Science, 150, 971-9. Mersky, 3. (1971). On the development of pain. Headache, 10, 116-123. Mertus, 3.A. (1984). Waveform Editor and Signal Processor, Version F-4.12. Unpublished Manual. Providence, Rhode Island. Meyer, M.B., 3ones, B.S. and Tonascia, 3.A. (1976). Perinatal events associated with maternal smoking during pregnancy. American 3ournal of Epidermiology, 103, 464-476. 132 Mountcastle, V.B. (1980). Pain and temperature sensibilities. In V.B. Mountcastle (Ed.), Medical Physiology, Vol. 1, St. Louis : Mosby. Muller, E., Hollien, H. and Murry, T. (1974). Percpetual responses to infant crying : Identification of cry types. Journal of Child Language, i, 89-95. Murray, A.D. (1979). Infant crying as an elicitor of parental behavior : An examination of two models. Psychological Bulletin, 86(1), 191-215. Murry, T. and Murry, J. (Eds.) (1980). Infant Communication ; Cry and Early  Speech. Houston, Texas : College-Hill Press. Naeye, R.L. (1979). The duration of maternal cigarette smoking and fetal and placental disorders. Early Human Development, 3_» 229-237. Nuite, J.A. (1976, September). Effects of narcotic agonists on the central nervous system of rates. Paper presented at the meeting of the American Psychological Association, Washington, D.C. Oster, H. (1978). Facial expression and affect development. In M. Lewis and L.A. Rosenblum (Eds.) The Development of Affect, New York : Plenum. Ostwald, P. (1972). The sounds of infancy. Developmental Medicine and Child  Neurology. _14, 350-361. Owens, M.E. (1984). Pain in infancy : conceputal and methodological issues. Pain, 20, 213-230. Owens, M.E. and Todt, E.H. (1984). Pain in infancy : neonatal reaction to a heel-lance. Pain, 20,77-86. Parmelee, A.H., Wenner, W.H., Akiyama, Y., Schultz, M. and Stern, E. (1967). Sleep states in premature infants. Developmental Medicine Child  Neurology, 9, 70-77. 133 Petrie, R.H., Wu, R., Miller, F.C., Sacks, D.A., Sugarman, R., Paul, R.H. and Hon, E.H. (1976). The effect of drugs on uterine activity. Obstetrics and  Gynecology, 48, 431-435. Prechtl, H.F.R. (1974). The behavioural states of the newborn infant : A review. Brain Research, 76, 185-212. Prechtl, H.F.R. (1981). The study of neural development as a perspective of clinical problems. In K.J. Connolly and H.F.R. Prechtl (Eds.), Maturation  and Development : Biological and Psychological Perspectives. Clinics in Developmental Medicine No. 77/78, Spastic International Medical Publications. London : Heinemmann. Prechtl, H.F.R. and O'Brien, M.J. (1982). Behavioural states of the full-term newborn. The emergence of a concept. In P. Stratton (Ed.), Psychobiology of the Human Newborn. New York : Wiley and Sons. Prechtl, H.F.R. and O'Brien, M.J. (1981). Behavioural states of the full-term newborn. In P. Stratton (Ed.), Psychobiology of the Human Newborn. N.Y. : Wiley. Reinis, S. and Goldman, J.M. (1980). The Development of the Brain. Springfield III : C C . Thomas. Rich, E.C, Marshall, R.E. and Volpe, J.J. (1974). The normal neonatal response to pin prick. Developmental Medicine and Child Neurology, 16, 432-434. Rieser, J.J., Green, J.A., McLaughlin, F.J. and Doxsey, P.A. (1982, March). The effects of contact comfort on infant's modulation of response to  pain. Paper presented at the International Conference on Infant Studies, Austin, Texas. Rinn, W.E. (1984). The neuropsychology of facial expression : A review of the neurological and psychological mechanisms for producing facial expressions. Psychological Bulletin, 95, 52-77. 134 Ritchie, J.A. (1981). Development of body concept and concepts of illness and wellness. In M. Tudor (Ed.), Child Development, New York : McGraw-Hill. Richards, M.P.M., Bernal, J.F. and Brackbill, Y. (1976). Early behavioural differences : Gender or circumcision? Developmental Psychobiology, 9, 89-95. Robinson, R.J. (1969). Nomenclature of the stages of sleep. In R.J. Robinson (Ed.) Brain and Early Behaviour, Academic Press : New York, 173-177. Roffwarg, H.P., Muzio, I.N. and Dement, W.C. (1966). Ontogenetic development of the human sleep-dream cycle, Science, 152, 604. Rose, S.A., Schmidt, K. and Bridger, W.H. (1978). Changes in tactile responsivity during sleep on the human newborn infant. Developmental  Psychology, 14, 163-172. Rosenblith, J.F. and Sims-Knight, J.E. (1985). In the Beginning : Development  in the First Two Years of Life. Monterey, California : Brooks/Cole. Rouvee-Collier, C.K. and Lipsitt, L.P. (1982). Learning, adaptation and memory in the newborn. In P. Stratton (Ed.), Psychobiology of the Human  Newborn. N.Y. : Wiley. Sameroff, A.J. and Cavanagh, P.J. (1979). Learning in Infancy : A developmental perspective. In J.D. Osofsky (Ed.), Handbook of Infant  Development, 344-392. N.Y. : Wiley. Savedra, M., Gibbons, P., Tesler, M., Ward, J. and Wegner, C. (1982). How do children describe pain ? A tentative assessment. Pain, 14, 95-104. Schmidt, K. and Birns, B. (1971). The behavioural arousal threshold in infant sleep as a function of time and sleep state. Child Development, 42, 269-277. 135 Shephard-Look, D.L. (1982). Sex differences and the development of sex roles. In B.B. Wolman (Ed.) Handbook of Developmental Psychology. Prentice Hall : Englewood Cliffs, N.J. Sherrington, CS. (1900). Cutaneous sensations. In E.A. Schafer (Ed.), Textbook of Physiology. London : Pentland. Sirvio, P. and Michelsson, K. (1976). Sound spectographic cry analysis of normal and abnormal newborn infants. Folia. Phoniat., 28, 161-173. Spears, W.C. and Hohle, R.H. (1967). Sensory and perceptual processes in infants. In Y. Brackbill (Ed.), Infance and Early Childhood : A handbook and guide to human development. New York : Free Press. Stechler, G. and Halton, A. (1982). Prenatal influences on human development. In B.B. Wolman (Ed.), Handbook of Developmental  Psychology. Prentice Hall : N.J. Sternman, M.B. (1972). The basic rest-activity cycle and sleep : Developmental considerations in man and cats. In CB. Clemente, D.P. Purpura, F.E. Mayer, (Eds.), Sleep and the Maturing Nervous  System. N.Y. : Academic Press. Stratton, P. (1982). Rhythmic functions in the newborn. In P. Stratton (Ed.), Psychobiology of the Human Newborn. Wiley and Sons : New York. Swafford, L.I. and Allen, D. (1968). Pain relief in the pediatric patient. Medical Clinics of North America. 48(4), 131. Tenold, J.L., Crowell, D.H., Jones, R.H., Daniel, T.H., McPherson, D.F., Popper, A.N. (1974). Cepstral and stationarity analyses of full-term and premature infants' cries. Journal of theAcoustic Socity of America, 56(3), 975-980. Thoden, C. and Koivisto, M. (1980). Acoustic analysis of the normal jpain cry. In T. Murry and J. Murray (Eds.), Infant Communication ; Cry and Early  Speech. Houston Texas : College-Hill Press, 124-151. 136 Titchener, E.B. (1909-1910). A Textbook of Psychology. N.Y. : Macmillan. Trevarthen, C. (1979). Neuroembyology and the development of perception. In F. Falkner and J.M. Tanner (Eds.), Human Growth, Vol. 3, Neurobiology and Nutrition. New York : Plenum. Tronick, E.Z., (Ed.) (1982). Social Interchange in Infancy. Baltimore : University Park Press. U.S. Department of Health, Education and Welfare (1979). Smoking and  Health : A Report of the Surgeon General. CDHEW Pub. No. PNS 79-50066. Washington, D.C. : U.S.Government Printing Office. Varni, J.W., Bessman, C.A., Russo, D.C. and Cataldo, M.F. (1980). Behavioural management of chronic pain in children : A case study. Archives of Physical Medicine and Rehabilitation, 61> 375-379. Varni, J.W., Gilbert, A. and Dietrich, S.L. (1981). Behavioural medicine in pain and analgesia management for the hemophilic child with factor VIII inibitor. Pain, 11, pp. 121-126. Varni, J.W., Katz, E.R. and Dash, J. (1982). Behavioural and neurochemical aspects of pediatric pain. In D.C. Russo and J.W. Varni (Eds.), Behavioural Pediatrics. New York : Plenum Press. Volpe, J.J. and Koenigsberger, R. (1981). Neurologic disorders. In G.B. Avery (Ed.), Neonatology : Pathophysiology and Management of the Newborn. Philadelphia : Lippencott. Vlach, V., Bernuth, H. Von, and Prechtl, H.F.R. (1969). State dependency of exteroceptive skin reflexes in new-born infants. Developmental Medicine  and Child Neurology. U, 353-362. Wasz-Hockert, O., Lind, J., Vuorenkoski, V., Partanen, T. and Valanne, E. (1968). The infant cry. Clinics in Developmental Medicine, (No. 29) Spastics International Medical Publications. London Heinemann. 137 Wasz-Hockert, O., Partenen, T., Vuorenski, V., Valanne, E., and Michelsson, K. (1964). Effect of training on ability to identify perverbal vocalizations. Developmental Medicine and Child Neurology, 6, 393-396. Weiss, C. (1968). Does circumcision of the newborn require an anesthetic? Clinical Pediatrics, 7, 128-129. Weller, G.M. and Bell, R.Q. (1965). Basal skin conductance and neonatal state. Child Development, 36, 647-657. Wolff, P.H. (1959). Observations on newborn infants. Psychosomatic  Medicine, 21, 110-118. Wolff, P.H. (1966). The causes, controls and organization of behaviour in the neonate. Psychological Issues, Vol. 5 (1), Monograph 17, International University Press, New York. Wolff, P.H. (1969). The natural history of crying and other infant vocalizations in early infancy. In B.M. Foss (Ed.), Determinants of Infant  Behaviour, Vol. 4. London : Metheun. Yang, R.K. and Douthitt, T.C. (1974). Newborn responses to threshold tactile stimulation. Child Development, 45, 237-242. 138 APPENDIX A P A R E N T C O N S E N T F O R M 138a R e s e a r c h S t u d y : I n f a n t R e s p o n s e s t o H o s p i t a l P r o c e d u r e s A s t u d y o f I n d i v i d u a l d i f f e r e n c e s I n b a b i e s ' a c t i o n a n d c r y i n g , w h e n t h e y a r e I n v a r i o u s s t a t e s o f a c t i v i t y r a n g i n g f r o m s l e e p i n g t o c r y i n g w h e n t h e y a r e I n d i s c o m f o r t , i s b e i n g c a r r i e d o u t a t G r a c e H o s p 1 t a I . We w i l l b e v i d e o t a p i n g a n d r e c o r d i n g w h i l e e a c h b a b y i s u n d e r g o i n g r o u t i n e h o s p i t a l p r o c e d u e e s . I CONSENT I DO N O T C O N S E N T t o m y b a b y ' s p a r t i c i p a t i o n I n t h i s s t u d y . S i g n a t u r e o f P a r e n t N o t e : T h e s e r e c o r d i n g s w i l l b e c o n f i d e n t i a l a n d u s e d f o r r e s e a r c h p u r p o s e s o n l y . Y o u a r e f r e e t o w i t h d r a w p e r m i s s i o n a t a n y t i m e , a n d t h i s w i l l n o t a f f e c t y o u r b a b y ' s c a r e w h i l e I n t h e h o s p i t a l . F o r e n q u i r i e s c o n t a c t : D r . R u t h G r u n a u , P s y c h o l o g i s t , C h i l d r e n ' s H o s p i t a l , D r . K e n n e t h C r a i g , P s y c h o l o g i s t , U . B . C . , . 139 APPENDIX B Characteristics of Infants Who Did Not Cry to Heel Lance Apgar Maternal Delivery Hours Case Sex Birthweight (5 min.) Age Race Mode Since Feed: 40 boy 3020 9 29 Or i e n t a l Emg. C/S 1.86 69 boy 3200 9 27 East Ind. Planned C/S 5.33 89 g i r l 3870 9 32 White Sp. Vag. 1.33 99 boy 3740 8 34 White Planned C/S 1.33 104 g i r l 4050 9 36 White Sp. Vag. 1.42 114 g i r l 3320 9 26 Or i e n t a l Forcep. Vag. 1.58 115 g i r l 3560 10 33 White Sp. Vag. .72 140 APPENDIX C Inter-Rater R e l i a b i l i t y Per Subject Computed Across A l l Face Actions ( R e l i a b i l i t y Sample n = 32) Case C o e f f i c i e n t Case C o e f f i c i e n t 15 .50 128 .91 16 .97 133 .84 26 .82 149 .92 29 .63 152 .94 30 .53 156 .94 32 .84 168 .92 33 .98 34 .90 46 .97 47 .91 52 .80 54 1.00 55 .91 69 .79 77 .84 79 .90 80 .87 81 .82 92 .81 110 .95 111 .79 113 .85 123 .98 124 .99 126 .92 127 .98 APPENDIX D Summaries of Univariate ANOVA of each cry v a r i a b l e by Technician Source df SS MS F P — f o 1 Between groups 8 Within groups 131 Total 139 f 2  r o Between groups 8 Within groups 123 Total 131 f 3  r o Between groups 8 Within groups 117 Tota l 125 Number of Cry Cycles Between groups 8 Within groups 131 Total 139 157857.20 3723445.73 3881302.94 19732.15 .69 n.s 28423.25 108498.59 13562.32 .42 n.s 3990487.65 32442.99 4098986.24 199383.84 24922.98 .68 n.s 4251213.37 36335.16 4450597.21 293.23 36.65 1.00 n.s 4776.46 36.46 5069.68 Duration (1st Cry) Between groups 8 27701976.43 3462747.05 .66 n.s Within groups 131 683092564.50 5214447.06 Total 139 710794541.00 Cry Latency Between groups Within groups Total 8 56652010.91 131 767953175.30 139 824605186.20 7081501.36 1.21 n.s 5862237.98 142 APPENDIX E ANOVA Summary f o r Time Since Feeding by State Source df SS MS F p=: Between Groups 3 9.59 3.20 1.60 .19 Within Groups 136 271.72 2.00 143 APPENDIX P ANOVA Summary f o r F a c i a l Movement to Heel Eub a/Heel Lance by State and Sex Source df F Between Subjects State 226.14 3 75.38 7.29 .0001 Sex 0.74 1 0.74 .07 .79 State x Sex 27.24 3 9.08 .88 .45 Error 1365.39 136 10.34 Within Subjects Heel rub/lance 1649.53 1 16.49.53 295.35 .0000. State x rub/lance 61.85 3 20.62 3.69 .01 Sex x rub/lance 1.52 1 1.52 .27 .60 State x sex x rub/lance 25.58 3 8.53 1.53 .21 Error 737.23 132 5.58 Note: a summed over 6 seconds p r i o r to heel lance summed over 6 seconds following heel lance APPENDIX G ANOVA Summary f o r * Latency to F a c i a l Movement by State and Sex Source SS df MS State 1.40 Sex 2.78 Stat e x Sex 3.97 Re s i d u a l 77.60 3 1 3 132 .47 2.78 1.32 .59 .79 .50 4.73 .03 2.25 .08 * transformed to l o g 10 145 APPENDIX H ANOVA Summary f o r Number of Cry Cycles by state and Sex Source SS df MS State 195.48 Sex 131.98 State x Sex 9.46 Residual 4652.80 3 65.16 1 131.98 3 30.49 132 35.25 1.85 3.74 .86 .14 .05 .46 146 APPENDIX I ANOVA Summary f o r *Latency to Cry by State and Sex Source SS df MS State 8.24 Sex 5.50 State x Sex 8.54 Residual 194.40 3 1 3 132 2.75 5.50 2.85 1.47 1.86 .14 3.73 .06 1.93 .13 transformed to log 10 147 APPENDIX J ANOVA Summary f o r *Duration of Cry by State and Sex Source SS df MS F \o4= State Sex State x Sex Residual .20 .19 .09 29.34 3 1 3 132 .07 .19 .03 .22 .30 .83 .88 .35 .13 .94 log 10 transformation 148 APPENDIX K ANOVA Summary f o r Fundamental Frequency of Cry Over Time by State and Sex Source SS df MS F Between Subjects State 99359.47 3 33119.82 .42 .74 Sex 24112.18 1 24112.18 .31 .58 State x Sex 112117.60 3 37372.53 .47 .70 Error 9226751.16 117 78861.12 Within Subjects Time 249791.28 2 124895.64 14.26 .000 Time x State 59573.41 6 9928.90 1.13 .34 Time x Sex 2176.59 2 1088.30 .12 .86 Time x State x Sex 68602.839 6 11433.73 1.31 .26 Error 2049813.28 234 8759.88 149 APPENDIX L ANOVA Summary f o r Birthweight by f 0 ± Category ( 599 hz vs. 599 hz) and Smoking Status Source SS df MS F p £ Smoking Status 756158.63 1 756158.63 4.48 .04 f , category 1220528.51 1 1220528.51 7.23 .008 o 1 Smoking status x f Q ! 1568.24 1 1568.24 .009 .92 Residual 18408740.11 109 16887.52 150 APPENDIX M Frequency (and Column %) of Type of Ob s t e t r i c Medication by Mode of Delivery O b s t e t r i c Medication Spontaneous Vaginal Low Mid or High Forceps Forceps Vaginal Vaginal Planned Emergency Cesarian Cesarian None 50% N20/0 Local N 20 + Local Epidural Epidural & N 20 or l o c a l Epidural & N 20 & l o c a l Narcotic &/or N 20 &/or l o c a l Epidural & Narcotic General 17 (24%) 5 ( 7%) 15 (21%) 23 (32%) 1 ( 1%) 5 ( 7 % ) 5 ( 7%) 0 0 1 ( 8 % ) 1 ( 5 % ) 1 ( 8 % ) 1 ( 5%) 3 (25%) 1 ( 5%) 1 ( 8 % ) 6 (27%) 4 (33%) 9 (41%) 1 ( 8 % ) 2 ( 9%) 1 ( 8 % ) 2 ( 9 % ) 0 0 0 0 10 (48%) 0 0 0 0 2 (15%) 5 (38%) 1 ( 1%) 0 3 (14%) 2 (15%) 8 (38%) 4 (31%) Totals N = 140 72 12 22 21 13 151 APPENDIX N Freqency (and row %) of Type of Ob s t e t r i c Medication by State State Quiet Active Quiet Active Sleep Sleep Awake Awake Row Obs t e t r i c Totals Medication N=140 None 6 (35%) 4 (23%) 3 (18%) 4 1 (24%) 17 50% N20/0 3 (43%) 2 (29%) 0 2 (29%) 7 Local 5 (29%) 4 (24%) 3 (18%) 5 (29%) 17 N 20 + Local 7 (26%) 10 (37%) 4 (15%) 6 (22%) 27 Epidural 10 (50%) 5 (25%) 2 (10%) 3 (15%) 20 Epidural + N 20 or Local 6 (26%) 8 (35%) 6 (26%) 3 (13%) 23 Epidural + N 20 + Local 5 (63%) 1 (12%) 1 (12%) 1 (12%) 8 Narcotic &/or N 20 &/or Local 2 (67%) 1 (33%) 0 0 3 Epidural + Narcotic 2 (33%) 3 (50%) 0 1 (17%) 6 General 3 (25%) . 7 (58%) 1 ( 8%) 1 ( 8%) 12 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items