Paternal socio-psychological factors and infant attachment: The mediating role of synchrony in father–infant interactions

Abstract

The present research explored the role of synchrony as a mediator of the relationship between marital satisfaction and infant attachment, and between depressive symptoms and infant attachment. Fifteen father–infant and mother–infant dyads were videotaped in face-to-face interactions. Videotaped interactions were coded for frequency of synchronized interactions. Quality of attachment was assessed several months later using the Waters’ Attachment Q-sort (AQS). Tests of mediation revealed that synchrony accounts for the relation between marital quality and infant attachment to fathers, and between depressive symptoms and level of attachment to mothers. Finally, demographic variables associated with work and family role strain were explored as potential correlates of marital dissatisfaction and depressive symptoms.

Introduction

The formation of secure attachments during infancy has been associated with a broad range of positive developmental outcomes, including increased social competency (Bohlin, Hagekull, & Rydell, 2000; Booth, Rose-Kraanor, & Rubin, 1991; Sroufe, 1983), intellectual curiosity (Arend, Gove, & Sroufe, 1979), persistence in problem-solving (Frankel & Bates, 1990; Matas, Arend, & Sroufe, 1978), and more symbolic play (Slade, 1987). Although most research has focused on the developmental outcomes of infant–mother relationships, secure infant–father attachments have also been linked to positive developmental outcomes (Belsky et al., 1984, Belsky et al., 1984, Belsky et al., 1984; Lamb, Hwang, Frodi, & Frodi, 1982; Main & Weston, 1981). Furthermore, infants securely attached to both parents demonstrate higher levels of competency compared to infants securely attached to only one parent (Belsky et al., 1984, Belsky et al., 1984, Belsky et al., 1984).

The positive outcomes associated with attachment security have resulted in numerous investigations into potential developmental antecedents. Maternal sensitivity, for example, has been explored extensively in relation to the formation of infant–mother attachment relationships. Infants with more sensitive mothers are more likely to develop secure attachments to their mothers (see de Wolff & van Ijzendoorn, 1997, for a recent meta-analysis). The strength of this relationship, however, has been questioned (Seifer, Schiller, Sameroff, Resnick, & Riordan, 1996). Unfortunately, the maternal sensitivity construct has been operationalized and measured in various ways by different investigators. For the purpose of the present research, sensitivity was defined in terms of interactional synchrony, or the “extent to which an interaction appears to be reciprocal and mutually rewarding” (Isabella, Belsky, & von Eye, 1989, p. 13). Isabella et al. (1989) reported that, while the interactions between mothers and infants with secure attachments, were characterized by disproportionately more frequent synchronous exchanges, those of insecurely attached infants were disproportionately asynchronous (i.e., one-sided, unresponsive, or intrusive). However, whether the frequency of synchronized interactions is similarly related to the formation of secure attachments to fathers has not yet been investigated.

Most infants do form attachments to their fathers (see review by Lamb, 1976b). Early research on father–infant relations focused primarily on the interaction styles of fathers compared to mothers (Lamb, 1975, Lamb, 1977, Nash, 1965, Parke & Tinsley, 1987), and infants’ preferences for one parental gender over another (Lamb, 1976a, Lamb, 1976c; Spelke, Zelazo, Kagan, & Kotelchuck, 1973). However, interest subsequently turned to the importance of infant–father attachment security in relation to early developmental outcomes (Belsky et al., 1984, Belsky et al., 1984, Belsky et al., 1984, Lamb et al., 1982, Main & Weston, 1981). The positive outcomes have resulted in a few investigations into potential antecedents. Similar to the maternal research, one such antecedent has included parental sensitivity (see van Ijzendoorn & de Wolff, 1997, for a recent meta-analysis).

Although a weak association has been found between paternal sensitivity and infant attachment security, the findings have been inconsistent and the measures of sensitivity have differed across studies (Cox, Owen, Henderson, & Margand, 1992; Easterbrooks & Goldberg, 1984, Notaro & Volling, 1999, Volling & Belsky, 1992). When assessed in terms of emotional supportiveness and quality of assistance during a problem-solving task, a relationship between sensitivity and infant attachment was not demonstrated (Easterbrooks & Goldberg, 1984). However, when assessed within the context of play, a composite variable that included global ratings of sensitivity, warmth, stimulation and activity level was found to significantly predict infant–father attachment security (Cox et al., 1992). In addition to the inconsistent findings and measures, only a few studies conducted with fathers have included comparable data from mothers (Cox et al., 1992, Notaro & Volling, 1999).

The maternal research indicates that certain socio-psychological factors can interfere with a mother’s ability to engage in synchronous interactions. Depressed mothers, for example, tend to exhibit either intrusive (i.e., overstimulating) or withdrawn (i.e., understimulating) styles of interaction (Cohn, Campbell, Matias, & Hopkins, 1990; Field, Healy, Goldstein, & Guthertz, 1990; Jones et al., 1997; Jones, Field, Hart, Lundy, & Davalos, 2001). These styles of interaction have been associated with delays in early socioemotional and cognitive development (Field, 1992, Gelfand & Teti, 1990; Zahn-Waxler, Iannotti, Cummings, & Denham, 1990), and are similar to the types of interactions associated with the development of insecure attachments (Lyons-Ruth et al., 1990, Lyons-Ruth et al., 1986; Radke-Yarrow, Cummings, Kuczynski, & Chapman, 1985; Teti, Gelfand, Messinger, & Isabella, 1995).

Furthermore, the depression does not have to be clinical depression. Maternal reports of depressive symptoms have also been found to influence maternal sensitivity (Bettes, 1988; Donovan, Leavitt, & Walsh, 1998), synchrony in mother–infant behavioral states (Field, 1992), and early socioemotional and cognitive development (Field, 1992). Maternal reports of depressive symptoms have also been linked to the development of insecure infant attachments (Dawson et al., 2001). Even depression that has been experimentally induced has been found reduce the mothers’ ability to elicit positive responses from their infants (Zekowski, O’Hara, & Wills, 1986).

The depression is more likely to have a negative impact on the mother–infant relationship when combined with additional socio-psychological stressors (Lyons-Ruth et al., 1990; National Institute of Child Health and Human Development, NICHD, 1999; Sameroff, Seifer, Baldwin, & Baldwin, 1993; Seifer, 1995, Shaw & Vondra, 1993). Maternal depression has been associated with stressors such as conflict regarding whether or not to return to work (Donovan et al., 1998). In addition, a common correlate of maternal depression is marital dissatisfaction (e.g., Coyne, Burchill, & Stiles, 1991; Downey & Coyne, 1990, Gotlib & Whiffen, 1989, Lyons-Ruth et al., 1990), which has also been associated with reduced maternal sensitivity (Cox, Owen, Lewis, & Henderson, 1989; Donovan et al., 1998, Goldberg & Easterbrooks, 1984, Howes & Markman, 1989, Isabella & Belsky, 1985). Thus, maternal depression and marital dissatisfaction may each contribute to reduced levels of maternal sensitivity.

The influence of socio-psychological factors, such as depression and marital quality, on fathers’ interactions with infants has received less attention. However, the quality of parental relations has been suggested to have a greater impact on fathers’, compared to mothers’, relationships with infants (Belsky et al., 1984, Belsky et al., 1984, Belsky et al., 1984; Belsky, Youngblade, Rovine, & Volling, 1991; Cox et al., 1989, Goldberg & Easterbrooks, 1984, Lamb & Elster, 1985, Volling & Belsky, 1992). Higher levels of marital satisfaction in fathers has been associated with more positive thoughts and feelings toward their infants (Belsky et al., 1984, Belsky et al., 1984, Belsky et al., 1984, Corwyn & Bradley, 1999). Furthermore, because the quality of the marital relationship appears to be so strongly related to the nature of the father–infant relationship, some researchers have proposed that fathers may view marriage and fatherhood as one and the same (Corwyn & Bradley, 1999, Furstenberg & Cherlin, 1991).

Marital dissatisfaction in fathers has been associated with stressors such as work-family role conflict (Barling, 1986) and financial strain (Elder, Conger, Foster, & Ardelt, 1992). The role of fatherhood is still stereotyped as the primary economic provider (Cowan & Cowan, 1987). At the time many fathers are becoming parents, they are also engaged in establishing new careers (Cowan & Cowan, 1987). Compared to mothers, fathers still receive little support from employers with regard to family caregiving responsibilities, and paternity leave is still rarely available (Cowan & Cowan, 1987). Thus, the processes that underlie work-family role conflict for fathers may differ from those of mothers (Corwyn & Bradley, 1999). In general, fathers who find themselves with greater caregiving responsibilities that conflict substantially with their responsibilities related to work, or with their role as economic provider, may experience increased levels of marital dissatisfaction (Corwyn & Bradley, 1999). Financial strain may also have a more negative affect on the fathers’ level of marital quality, compared to the mothers’, and may affect the father–infant relationship more adversely than the mother–infant relationship (Elder et al., 1992).

The role of depression on fathers’ interactions with their infants has received much less attention. However, the effects of depression may be greatest when combined with poor marital quality. In one study, fathers with mild depression touched and interacted with their infants less frequently and provided their infants with less affection compared to nondepressed fathers (Zaslow, Pederson, Cain, Suwalsky, & Kramer, 1985). In addition, paternal depression was found to be inter-correlated with marital conflict (Zaslow et al., 1985). In another study, depressed fathers interacted with their infants similarly to nondepressed fathers (Field, Hossain, & Malphurs, 1999). However, none of the fathers were legally married, although the majority resided with their infants’ mothers (Field et al., 1999). It is possible that conflict within a marriage may be more distressing, since the relationship between the parents is likely to be more significant, than in a nonmarital relationship.

The influence of socio-psychological factors, such as depression and marital quality, on fathers’ interactions with infants has received less attention compared to the maternal research. However, marital quality has been suggested to have a greater impact on fathers’, compared to mothers’, relationships with their infants (Belsky et al., 1984, Belsky et al., 1984, Belsky et al., 1984, Belsky et al., 1991, Cox et al., 1989, Goldberg & Easterbrooks, 1984, Volling & Belsky, 1992). Furthermore, based on the limited research available, while depression and marital dissatisfaction may co-occur in fathers (Zaslow et al., 1985), it is unclear whether paternal depression per se, may have the same negative effect on father–infant interactions (Field et al., 1999) as maternal depression has on mother–infant interactions (Cohn et al., 1990, Field et al., 1990, Jones et al., 1997, Jones et al., 2001).