“Behavioral and Autonomic Antecedents of Attachment in High-Risk Infants”

by Nathan A. Fox

found in The Psychobiology of Attachment and Separation, edited by Martin Reite and Tiffany Field

Academic Press, Inc, 1985 – Orlando

ISBN 0125867808

Pp 389 – 414



“The evidence from the high-risk sample, and the basis for this speculative model, centers on the measurement of heart rate variability (HRV).  The assumption underlying the model is that the patterning of heart rate (HR) or HRV may be used as an estimate of neural control over the heart and thus, indirectly, as a measure of central nervous system (CNS) integrity.  (Fox/hr/390)”

“These data, although preliminary in nature, seem to indicate that individual differences in facial expressiveness are associated with variations in autonomic pattern.  (Fox/hr/390)”


“Much of the work involving measurement of autonomic activity in infants is the result of the theoretical interpretation of phasic autonomic changes articulated by the Laceys (Lacey & Lacey, 1958).  Lacey and Lacey (1958) proposed that changes in the direction of heart rate may mediate cognitive and emotional responses.  Subsequently, Graham and Clifton (1963) adopted these notions and those of Sokolov (1963) and proposed a response dichotomy based on deceleration or acceleration interpreted as orienting and attention or a defensive response.  (Fox/hr/391)”

“…the patterning of HR or HRV has been used as an estimate of neural control over the heart and thus indirectly as a measure of CNS integrity…(Fox/hr/392)”

“Heart rate and HRV are a result of a complex interaction between parasympathetic and sympathetic divisions of the autonomic nervous system….  Both long- and short-term variability may indicate phasic alterations in parasympathetic activity….  Short-term variability, the mean successive difference score of beat-to-beat variability, is considered the best noninvasive measure of vagal tone or parasympathetic activity (Porges, 1974; Yeh, Forsythe, & Hon, 1973).  (Fox/hr/392)”

“The primary neural contribution to HRV is through the vagus.  Thus, HRV has been considered a noninvasive estimate of vagal tone.  ….the balance of parasympathetic and sympathetic influences on the heart may serve as a measure of individual differences in attention style and may accurately reflect CNS integrity…..  Data from Fox and Porges (in press) indicte that a measure of neonatal HRV variability accurately predicts positive developmental outcome in a high-risk sample.  Of interest, then, is the possibility that HRV may also affect socioemotional development and personality style.  In partic- (Fox/hr/392) ular, differences in HRV may be related to certain aspects of sociability that affect the development of a secure attachment bond.  (Fox/hr/393)”


“Notarious and Levenson (1979), for example, exposed two groups of adults, preselected as being facially expressive or not, to a threatening situation while recording HR, respiration rate, skin conductance, and facial expression.  They found that inhibitors were less facially expressive and more physiologically reactive than were expressers.  These data were interpreted along a “discharge” model of emotion.  (Fox/hr/397)”

“Others (Lanzetta & Kleck, 1970) have hypothesized that facially expressive adults are externalizers, whereas nonexpressive subjects fit more along the internalizer dimension (hence, their greater physiological response while lack of facially expressive response).  Interestingly, physiological reactivity in all these studies among nonexpressive adults was sympathetic in nature (tonic HR acceleration and increased skin conductance….).  (Fox/hr/397)”


“Short-term variability has been used as a measure of respiratory sinus arrhythmia and is thought to indicate indirectly the degree of parasympathetic influence on vagal tone of the heart (Katona & Jih, 1975).

“While these data are only a preliminary exploration of individual differ- (Fox/hr/399) erences in physiological activity and their relationship to expressive behavior, the associations uncovered in this small sample seem to indicate that a relationship between a measure of neural control over the heart may be significantly related to measures of facial behavior.  (Fox/hr/400)”

“High LTV [long-term variability] may indicate greater sympathetic dominance (Baust & Bohnert, 1969; Chess et al., 1975) and may be associated with less attentive behavior – hence, its correlation with mothers turning from interaction.  Similarly, attentive infants (indicated by their high vagal tone or high STV [short-term variability] will have mothers who turn toward them and are affectionate to them.  (Fox/hr/401)”

“The emerging pattern from this study is that individual differences in an infant’s autonomic responsiveness seem to be closely linked to differences in infant facial expression and indirectly to the mother’s responsiveness to her infant.  The data indicate that differences in vagal tone affect facial behavior.  It is possible that those centers in the brain stem responsible for modulation of vagal tone have intimate connections with the cranial nerve system responsible for facial behavior (Porges, 1983, personal communication).  Thus, oscillation in vagal tone would lead to more attentive facial expression, whereas decreases in these oscillations (or increases in sympathetic influence) would decrease the incidence of facial expressions.  (Fox/hr/401)”


“The data relating HRV and facial expression point to important physiological differences in infants that may affect the mother-infant interaction and, perhaps, the infant’s positive socioemotional development.  (Fox/hr/401)”

“In order to investigate this possibility, we examined the relationship between HRV and the child’s socioemotional development at 12 months of age.  (Fox/hr/401)”

“..the very fact that a significant relationship exists between a physiological behavior measured in early infancy and the behavioral output of a complexly determined system such as attachment may indicate that measurement of HRV may have tapped into a fundamental dimension of individual differences in sociability.  (Fox/hr/408)”

“The data relating HRV to facial expression data indicate that there may be a physiological basis for variations in individual infant expressiveness.  That variation, is based on differences in parasympathetic influence or vagal tone.  Infants with high vagal tone (high STV) exhibited facial expressions of interest more often than did infants with low vagal tone (low STV).  (Fox/hr/408)”

“…differences in expressiveness might influence the responsiveness and sensitivity of the mother to the infant.  Given the importance of both maternal responsiveness and sensitivity in face-to-face (Blehar, Lieberman, & Ainsworth, 1977) and other early interactive situations (Ainsworth, Bell, & Stayton, 1974) in determining quality of attachment, one may speculate that the avoidant behavior of infants in the current study was the result of early differences in facial expressivity  which, in turn, affected maternal behavior.  (Fox/hr/409)”

“…mothers respond, match, and even imitate their infants’ facial behavior as early as 3 months…)  (Fox/hr/409)”


“…differences in HRV may reflect a fundamental individual difference dimension that may affect cognitive as well as affective style….  Porges (1976)…has speculated that autonomic balance, the relative influence of both sympathetic and parasympathetic activity, may reflect central nervous system integrity and has articulated a model relating autonomic balance to conditions of psychopathy, autism, and hyperactivity (Porges, 1976).  (Fox/hr/409)”


“More recently, Kagan (1982) has presented data on individual differences in HR range and temperamental style in infants and toddlers.  In a series of studies, Kagan found that toddlers who were rated as more inhibited and fearful in a number of social settings displayed less HRV and faster HRs than infants rated as less fearful.  Most interestingly, infants who showed frequent separation anxiety between the ages of 7 and 11 months displayed higher HRs as well.  In addition, children showed lawful changes in rate and range during tasks that required their active assimilation of events.  That is, children who worked to assimilate stimuli displayed de- (Fox/hr/409) creased range during th experiment.  Kagan has referred to the state children assume when they are working actively to assimilate unfamiliar events as “attentive vigilance.”  He believes that the cardiac pattern accompanying vigilance and related to inhibition and fearfulness may index an important early individual difference dimension central to personality development.  (Fox/hr/410)”

“…tonic HR accelerations are primarily the result of increased sympathetic activity (Obrist et al., 1965).  (Fox/hr/410)”

“…decelerations are under vagal control…(Fox/hr/410)”

“…securely attached infants, reliably elicited vagally mediated responses, whereas insecurely attached did not.  Indeed, avoidant-insecure infants exhibited a sympathetic response to the separation event….  These data then implicate autonomic balance (and indirectly such measures of balance as HRV) as an important index of social responsiveness in the attachment situation.  (Fox/hr/410)”

“Although there clearly seems to be a normal variation in the balance of sympathetic and parasympathetic influences on autonomic responding, these natural variations may mediate differences in both cognitive, attentional, and affective behavior.  Such behaviors early in the first year may be reflected in perceptions of infant temperament (Rothbart & Derryberry, 1981). For a discussion of autonomic indices related to dimen- (Fox/hr/410) sions of infant temperament). And these variations in temperament or personality style may be enhanced or narrowed through interaction with the environment, ultimately contributing to the formation of the attachment bond.  (Fox/hr/411)”

“In this chapter, we have outlined a model for the use of HRV as a measure of neural mediation of the heart and hence central nervous system integrity…..  Data were presented linking differences in HRV to facial expressiveness, as well as attachment behavior at 12 months….  (Fox/hr/411)”


One would have to consider the information that infants of depressed mothers do not have the same facial expressiveness – decreased because the depressed mothers do not see others “happy” expressions – do not respond to them in their infants.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s