Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Sep 1.
Published in final edited form as: J Fam Psychol. 2018 May 28;32(6):753–761. doi: 10.1037/fam0000420

Maternal Sensitivity to Distress and Attachment Outcomes: Interactions with Sensitivity to Non- Distress and Infant Temperament

Esther M Leerkes 1, Nan Zhou 2
PMCID: PMC6126976  NIHMSID: NIHMS949417  PMID: 29809018

Abstract

The extent to which maternal sensitivity to infant distress predicts specific attachment outcomes independent of and in conjunction with maternal sensitivity to infant non-distress and in conjunction with infant negative emotionality was examined in a sample of 259 mother-infant dyads. Maternal sensitivity to infant distress and non-distress cues were observed in a series of distress-eliciting tasks when infants were 6 months and 1 year old. Mothers rated infant negative emotionality at 6 months. Infant-mother attachment was observed during the Strange Situation at 1 year. Four attachment outcomes were considered: the dichotomous security/insecurity classification, avoidant and resistant behaviors across both reunion episodes, and the single rating for disorganized behavior. Maternal sensitivity to distress and non-distress at 1 year were treated as covariates. Sensitivity to distress and non-distress at 6 months and at 1 year did not predict more adaptive attachment outcomes as main effects. However, sensitivity to distress and non-distress at 6 months interacted significantly in relation to avoidance such that sensitivity to non-distress was significantly associated with higher avoidance when sensitivity to distress was low, but not when sensitivity to distress was high. Furthermore, sensitivity to distress at 6 months interacted with infant negative emotionality in relation to security and both resistant and disorganized behaviors such that sensitivity to distress was positively associated with security and negatively associated with resistant and disorganized behaviors only among infants who were high on mother-reported negative emotionality. Implications for future research and intervention are discussed.

Keywords: maternal sensitivity, infant distress, infant temperament, attachment

The quality of the mother-infant attachment relationship is a consistent predictor of infants’ subsequent well-being. Specifically, secure infants generally have healthier developmental outcomes than infants with an insecure attachment, avoidant infants are particularly likely to experience subsequent internalizing symptoms, and both avoidant and disorganized infants are particularly likely to experience subsequent externalizing symptoms (Groh, Fearon, van IJzendoorn, Bakermans-Kranenburg, & Roisman, 2017). Given the significance of early infant-mother attachment for subsequent well-being, efforts to identify predictors of specific attachment outcomes are important. Some research demonstrates that mothers’ sensitive responses (i.e., quick, contingent, and appropriate) to infant distress signals are a unique predictor of infant attachment security over and above mothers’ sensitive responses to infant non-distress signals (Del Carmen, Pederson, Huffman & Bryan, 1993; Leerkes, 2011; McElwain & Booth La Force, 2006). To date, this nuanced approach has not been taken in relation to specific types of insecurity and disorganization likely because these groups tend to be small. Thus, in the current paper, we examine the extent to which maternal sensitivity to distress cues and maternal sensitivity to non-distress cues at 6 months predict continuous measures of infant attachment avoidance, resistance, and disorganization independent of one another and in conjunction with one another. In addition, we examine infant negative emotionality as a moderator of associations between dimensions of maternal sensitivity and attachment outcomes.

Maternal Sensitivity and Attachment Classifications

According to Bowlby (1969/1982), the quality of caregiving infants receive from their mothers influences their internal working model, or sense of self in relation to the world. Infants whose needs are met consistently and appropriately develop a secure working model characterized by feelings of self-worth and trust in others. In contrast, infants whose needs are ignored, rejected, or responded to inconsistently or harshly develop an insecure working model characterized by feelings of unworthiness and a sense that they cannot trust others to meet their needs. Bowlby asserted that infants’ internal working models influenced their subsequent expectations and interpretations of self and other as well as their own behavior in subsequent social interactions leading to more adaptive outcomes among secure relative to insecure children.

Subsequently, Ainsworth and colleagues provided an empirical test of Bowlby’s assertions by conducting extensive home observations of maternal behavior in infants’ first year and rating infants’ attachment-related behavior in a laboratory paradigm known as the Strange Situation (Ainsworth, Blehar, Waters and Wall, 1978). Infants classified as secure in the Strange Situation, which consisted of a series of progressively more stressful separations from and reunions with the mother, characteristically initiated and maintained proximity to their mothers during the reunions whether or not they had been distressed during the separations. Infants in the insecure-avoidant group showed minimal distress during the separation episodes, and actively avoided being near and interacting with their mothers upon reunion. Finally, infants in the insecure-resistant group became quite upset during the separation, and although they sought proximity to the mother upon reunion, they also became angry and resistant towards their mothers by hitting or pushing their mothers away. During the first year, secure infants had mothers who consistently responded to them in a prompt, warm, and sensitive manner. In contrast, avoidant infants had mothers who responded slowly if at all to their infants’ signals, even showing rejection at times. Finally, resistant infants had mothers who were inconsistently responsive to their signals, and engaged in a mix of sensitive and negative responses. Notably, both resistant and avoidant infants had mothers who were less responsive to their distress signals than secure infants (Ainsworth et al., 1978).

Subsequently, Main and Solomon (1990) identified a fourth group of infants who engaged in behaviors during the Strange Situation that were not clearly organized toward achieving a specific goal (e.g., freezing in place when their mothers returned, contradictory behaviors such as approaching mothers backwards). These infants were classified as disorganized, and many were maltreated or had caregivers with a history of trauma or mental illness suggesting that compromised parenting was likely a key factor. In fact, highly egregious, anomalous parenting behaviors, often described as frightened or frightening, are consistently predictive of attachment disorganization, whereas more typical insensitive behavior is not (Lyons-Ruth & Jacobvitz, 2016; Out, Bakermans-Kranenburg & Van IJzendoorn, 2009).

In the last several decades, numerous studies have been conducted on the association between maternal behavior and attachment outcomes. Meta-analyses of these studies have demonstrated moderate associations between sensitivity and attachment security (De Wolff & van IJzendoorn, 1997; Verhage et al., 2016). The weaker than expected associations between sensitivity and attachment security have led some to call for a more nuanced approach to examining the sensitivity-attachment connection.

Maternal Sensitivity to Distress and Attachment

Drawing from Bowlby’s (1969/1982) view that the core function of attachment is to promote infants’ survival, theorists have argued that parental responses to infant distress cues or safety bids may be particularly relevant to the development of attachment security (Goldberg, Grusec, & Jenkins, 1999; Thompson, 1997). Sensitivity to distress serves the socialization goal of providing comfort and safety, which is likely highly salient to the development of a sense of trust in the caregiver, whereas sensitivity to non-distress often serves other socialization goals such as cognitive stimulation (Davidov & Grusec, 2010). To date, this proposition has been directly tested in two studies in which sensitivity to distress cues and to non-distress cues were rated separately (del Carmen, Pederson, Huffman, & Bryan, 1993; McElwain & Booth-LaForce, 2006), and in a third study in which sensitivity was rated separately during a non-stressful free play period and during a series of distress-eliciting tasks (Leerkes, 2011). In all three studies, sensitivity to distress or in distressing contexts was a significant predictor of attachment security over and above sensitivity to non-distress cues or in a non-distressing context, whereas the reverse was not the case. These results support the view that sensitivity to distress is of paramount importance for the development attachment security.

More recently, Gedaly and Leerkes (2016) posited that insensitivity during distress-eliciting tasks could also be a stronger predictor of attachment disorganization than insensitivity in non-distressing tasks given this is highly egregious and perhaps frightening to the infant. However, this association was only observed among infants from low SES households. To date no study has simultaneously examined associations between sensitivity to distress cues and to non-distress cues in relation to attachment avoidance, resistance, or disorganization. Thus, additional research on the possible role of sensitivity to distress cues, as a main effect or in conjunction with other factors, in relation to specific attachment dimensions beyond the secure/insecure dichotomy is needed.

Patterns of Sensitivity in Response to Distinct Infant Cues

Main (1990) postulated that infants’ behavior in the Strange Situation reflects conditional strategies that evolve in response to the nature of caregiving they receive, and hence are adaptive in that context as these strategies should increase their mothers’ investment in them. For example, based on their experience of maternal rejection, avoidant infants may maintain distance from the mother during times of stress to suit their mothers’ preference for independence. In contrast, based on their experiences of mixed caregiving, resistant infants may heighten their distress to even mild stressors to suite their mother’s preference for dependence. Subsequently, Cassidy (1994) described these conditional strategies as emotion regulation strategies that evolve in response to the messages mothers convey about their preferences regarding the expression and regulation of emotion. In particular, maternal rejection promotes infants’ minimization of negative emotion which contributes to infants’ use of avoidant behaviors, whereas inconsistent or nonresponsive care promotes infants’ heightening or maximization of negative affect which contributes to infants’ use of resistant behaviors (Cassidy, 1994).

Although ratings of sensitivity do not specify mothers’ use of specific types of insensitive behaviors (i.e., rejection, nonresponsiveness), it can be argued that specific patterns of sensitivity to distress and non-distress may convey clear messages to infants about the extent to which their mothers tolerate distress which may influence infants’ attachment-related behavior. That is, mothers who are sensitive when their infants are playful or interested, but insensitive when their infants are sad, afraid or frustrated communicate a preference for non-distress which may promote infant minimization of negative affect resulting in avoidant attachment behaviors. In contrast, mothers who are highly sensitive to distress cues but insensitive when their infants are positive or neutral communicate a preference for infant distress which may promote the maximization of infant negative affect and resistant attachment behaviors. To our knowledge, the proposition that sensitivity to distress cues and to non-distress cues may interact to predict infant attachment outcomes has yet to be tested. However, such an approach has proven useful in relation to predicting toddlers’ affect regulation (Leerkes, Blankson & O’Brien, 2009). That is, maternal sensitivity to non-distress was associated with subsequent adaptive emotion regulation when maternal sensitivity to distress was high, but was linked with less optimal affect regulation when maternal sensitivity to distress was low perhaps because infants learned to upregulate distress given inferences about their mothers’ emotion preferences. Next, we turn to the role of infant negative emotionality in the development of attachment outcomes.

The Moderating Role of Infant Negative Emotionality

Infant negative emotionality, a temperamental trait reflecting infants’ frequent and intense expression of fear, sadness, and anger (Rothbart & Bates, 2006), plays a minor role in the development of attachment as a main effect (Groh et al., 2017). Nevertheless, there is evidence that infant negative emotionality may moderate the extent to which sensitivity is associated with attachment security. That is, maternal sensitivity appears to predict attachment security more strongly for infants high versus low on negative emotionality (see Vaughn & Bost, 2016 for a review). Perhaps infants who are frequently and intensely distressed are particularly dependent on external regulation from caregivers to instill a sense of calm and safety and hence encounter dual risk for maladjustment when cared for in an insensitive manner (Leerkes et al., 2009). Alternatively, this pattern may reflect differential susceptibility (Elliss, Boyce, Belsky, Bakermans-Kranenbur & van IJzendoorn, 2011) whereby infants high in negative emotionality are more susceptible to the effects of sensitive caregiving on attachment outcomes for both better and worse (Groh et al., 2017). Given our premise that sensitivity to distress is of particular significance in relation to attachment outcomes, we hypothesize that insensitive responses to distress cues may be particularly deleterious in relation to attachment outcomes for infants high on negative infant emotionality. This was not the case in the single prior study that tested the extent to which infant temperament moderated the effects of sensitivity in distress and non-distress contexts in relation to attachment security (Leerkes, 2011), but this proposition has yet to be tested in relation to disorganization and specific dimensions of insecurity.

The Current Study

In sum, the goal of this study is to address three primary gaps in the literature related to the predictors of specific dimensions of attachment insecurity: the extent to which maternal sensitivity to distress when infants are 6 months old (1) plays a unique role in relation to specific dimensions of insecurity, over and above sensitivity to non-distress, (2) interacts with sensitivity to non-distress, particularly in relation to avoidant and resistant behavior, and (3) interacts with infant negative emotionality in relation to each attachment outcome. Specifically, we hypothesized that maternal sensitivity to distress cues but not to non-distress cues would be positively associated with attachment security and negatively associated with infant avoidant, resistant, and disorganized behaviors as a main effect. Additionally, we anticipated that sensitivity to distress but not to non-distress would interact with infant negative emotionality such that sensitivity to distress would be more strongly related to each attachment outcome among infants who were high on temperamental negative emotionality relative to infants who were low on temperamental negative emotionality. Further, we predicted that maternal sensitivity to distress and non-distress would interact with one another in relation to attachment avoidance and resistance, but the pattern would be unique for each. Specifically, maternal sensitivity to non-distress cues would be positively associated with infant avoidance when maternal sensitivity to distress cues is low because this pattern of maternal sensitivity conveys a preference for distress minimization. In contrast, maternal sensitivity to distress cues would be positively associated with resistance when maternal sensitivity to non-distress cues is low because this pattern of sensitivity conveys a preference for distress maximization. Although we did not anticipate that sensitivity to distress and non-distress would interact in relation to security or disorganized behavior, we test this possibility so all models are comparable and to determine the discriminant validity of predictors.

In order to address these questions, we coded maternal sensitivity to distress cues and to non-distress cues at 6 months observed in contexts that were designed to elicit infant distress. Maternal sensitivity to distress and non-distress at 1 year were treated as covariates to rule out the possibility that any observed association between sensitivity at 6 months and attachment outcomes was an artifact of concurrent sensitivity. Finally, given characteristics of our sample, we examined race as a possible moderator of all effects, but we did not anticipate significant effects based on prior research (Mesman, van IJzendoorn, & Bakermans-Kranenburg, 2012).

Method

Participants

Participants were 259 primiparous mothers (128 European American, 123 African American, 8 multiracial) and their infants from the southeastern United States. Mothers ranged in age from 18 to 44 years (M = 25.1) at recruitment. Twenty-seven percent had a high school diploma or less, 27% had attended but not completed college, and 46% had a 4- year college degree. The majority (57%) of mothers were married or living with their child’s father, 24% were dating but not living with their child’s father, and 19% were single. Annual family income ranged from less than $2,000 to over $100,000; median = $35,000. Participating infants were full term; 51% were female. Of the initial participants, 226 provided data at 6 months on maternal sensitivity and/or infant temperament, and 203 provided infant attachment outcome data. The primary reasons for missing data were inability to locate or contact mothers, moving from the area or being too busy.

Procedures

Expectant mothers were recruited at childbirth education classes, obstetric practices, and by word of mouth. Upon enrollment in the study, women were mailed consent forms and a packet of questionnaires including measures of demographics. Mothers were mailed a measure of infant negative emotionality prior to the 6 month visit. Mothers and infants visited our laboratory for a videotaped observation of mother-infant interaction when infants were 6-months (M = 6.39, SD = .72) and 1-year old (M = 13.9, SD = .98). The Strange Situation was conducted at the beginning of the 1 year visit. Mothers received $50 and a gift at the completion of the prenatal and 6 month visit and $100 after the 1 year visit. All procedures were approved by the Institutional Review Board.

Measures

Maternal sensitivity at 6 months and 1 year

Mothers and infants participated in a series of brief distress eliciting tasks during the 6-month (arm restraint, novel toy approach, and still face) and 1-year (attractive toy in a jar and novel character approach) laboratory visits as described in Leerkes, Su, Calkins, Supple, and O'Brien (2016). Mothers were seated near their infants and within reach of a toy basket. They were instructed to interact with their infants as they wished. Infant affect and maternal behavior were continuously rated/coded from digital media files using INTERACT 9 (Mangold, Arnstorf, Germany) by different teams of coders. Infant affect was rated on a 7-point scale ranging from (1) high positive affect (intense smile, laughing or squealing) to (7) high negative affect (screams, wails, sobs intensely). Inter-rater reliability was good at 6 months and 1 year: weighted kappa = .76 and .75 based on 34 and 30 double-coded cases respectively. At 6 months, 96% of infants became distressed, and the average duration of distress across the tasks was 2 minutes (range = 0 to 7.75 minutes out of 8 minutes). At 1 year, 91% percent of infants became distressed, and the average duration of distress was 1 minute (range = 0 to 4.45 minutes out of 6 minutes).

Maternal behaviors during the distress-eliciting tasks at 6 months and 1 year were continuously coded using 12 mutually exclusive categories (negative, intrusive, mismatched affect, withdraw, distracted, persistent ineffective, monitor, task focused, calming, supportive, non-task focused engagement, routine care) described in Leerkes (2010). Thirty cases were double-coded for reliability at 6 months (kappa = .77) and 27 cases at 1 year (kappa = .80). Then, the mother behavior code file and infant affect code file were merged in Interact and new codes were created reflecting the co-occurrence of infant affect and maternal behavior. Then, each of these co-occurrences was assigned an a priori sensitivity rating on a 3 point scale (1 = insensitive, 2 = moderately sensitive; 3 = sensitive) using a syntax-based automated process. For example, monitoring (i.e., watching but not interacting) a neutral infant was assigned a rating of 3 (sensitive) because the infant was not signaling a need. Monitoring when an infant was distressed was assigned a rating of 1 (insensitive) because the infant was signaling a clear need to which the mother did not respond. Sensitivity ratings for each discrete maternal behavior during infant positive, neutral, and negative affect are described in [author reference omitted for blind review]. At both time points, maternal sensitivity to distress, is the mean rating across all moments an infant was distressed (an affect rating of 5, 6, or 7) across tasks, and maternal sensitivity to non-distress is the mean rating across all moments the infant was positive or neutral (an affect rating of 1–4) across tasks.

Infant temperament at 6 months

Mothers completed the Infant Behavior Questionnaire - Revised Very Short Form (Putnam, Helbig, Gartstein, Rothbart & Leerkes, 2014). The negative affect subscale was used in this study (12 items; e.g., When tired, how often did your baby show distress). Items were rated on a scale from 1 (Never) to 7 (Always). Internal consistency was .74. This measure correlated positively with average observed infant affect during the distress tasks at 6 months, r (224) = .19; p < .01, demonstrating convergent validity.

Infant Attachment Outcomes

Upon arrival at the 1 year visit, mothers and their infants were videotaped during the Strange Situation Procedure (Ainsworth et al., 1978). E. Carlson coded the videos of the Strange Situation using standard procedures. First, using the original 3 category system, 168 infants were secure, 16 avoidant, and 19 resistant; the low frequencies for insecure groups underscore the utility of a dimensional approach. Then, in a second coding pass, D behaviors were rated on a scale ranging from 1 to 9, and children (n = 44) with a score of 5 or higher were classified as disorganized (Main & Solomon, 1990). Of the 44 children classified as disorganized, 23 had a primary classification of secure, 8 avoidant, and 13 resistant. In the primary analyses, the secure versus insecure dichotomy served as one primary dependent variable, and we included the 23 infants that we secure/disorganized in the secure group. We also ran follow-up analyses to determine if results varied if these infants were classified as insecure. Avoidance and resistance scores, respectively, were averaged across reunion episodes 5 and 8. Such an approach is justified given ratings correlated significantly across the episodes: r = .49 for avoidance and .47 for resistance, both p < .01. Inter-rater reliability was calculated based on 30 double coded cases. For the secure vs. insecure dichotomy, Kappa = .85. For the continuous measures intraclass correlations coefficients were .78 for avoidance, .91 for resistance, and .60 for disorganization. Notably, the reliability coder available on our staff had not been certified reliable in coding disorganized behavior at the time, but E. Carlson’s ratings were used in analyses, and she has demonstrated reliability in coding disorganized behavior with other raters in other studies (e.g., Bernard et al., 2012).

Results

Preliminary Analyses

Mothers for whom both sensitivity and infant attachment data were available are not different on maternal age, education, family income, infant gender, and race from those for whom sensitivity and/or infant attachment were missing. Thus, analyses were run on the entire initial sample (N = 259) with missing data handled in primary analyses via full information maximum likelihood (FIML) which takes all available data into account. FIML is considered superior to both case deletion and various imputation methods as it leads to less biased estimates (Schafer & Graham, 2002; von Hippel, 2016).

Descriptive statistics and zero-order correlations among possible sociodemographic covariates and primary variables appear in Table 1. Mothers who were more highly educated, who reported higher family income, and who were European American (relative to African American) were rated as more sensitive at both 6 months and 1 year and reported lower infant negative affect. European American infants were rated as more resistant/disorganized relative to African American infants, and infants whose mothers were more educated and had higher income to needs ratios were rated as less avoidant. Infant gender was not associated with any primary variables and was considered no further. Thus, maternal education, income-to-needs ratio, and race were included as covariates. Also included as covariates were maternal sensitivity to distress and to nondistress at 1 year to control for possible concurrent associations between maternal sensitivity and infant attachment. Sensitivity to distress and non-distress at 1 year were negatively associated with avoidance as expected; however, contrary to expectation, sensitivity to non-distress at 1 year was positively associated with resistance/disorganization. Maternal sensitivity to infant distress and non-distress both correlated positively with dichotomous attachment security, although these associations were small in magnitude.

Table 1.

Means, Standard Deviations, and Zero-Order Bivariate Correlations among Variables (N = 259)

1 2 3 4 5 6 7 8 9 10 11 12
1. Race (European American = 1)1 --
2. Maternal education Prenatal .34 --
3. Income-to-needs ratio Prenatal .47 .55 --
4. Sensitivity to distress 1Y .23 .26 .18 --
5. Sensitivity to non-distress 1Y .25 .32 .27 .42 --
6. Sensitivity to distress 6M .17 .32 .32 .29 .18 --
7. Sensitivity to non-distress 6M .13 .30 .25 .30 .28 .41 --
8. Infant negative affect 6M −.27 −.19 −.21 .01 −.04 −.05 −.08 --
9. Attachment Security Y(sec = 1)1 .03 .12 .07 .07 −.01 .16 .15 −.05 --
10. Avoidant behavior 1Y −.03 −.20 −.13 −.12 −.05 .12 −.01 .15 −.35 --
11. Resistant behavior 1Y .17 .21 .22 .11 .25 −.02 .06 −.02 −.31 −.29 --
12. Disorganized behavior 1Y .05 −.10 −.03 −.03 .08 −.11 .00 .25 −.40 .46 .27 --
M 49.8 3.84 2.93 1.94 2.43 2.55 2.73 3.46 .83 2.13 1.83 3.00
Standard deviation -- 1.80 2.11 .62 .27 .34 .17 .93 NA 1.18 1.10 1.85
n 259 257 242 173 207 203 211 226 203 203 203 203
Open in a new tab

Note. Prenatal = prenatal phase; 6M = 6 months; 1Y = 1 year.

Bold coefficients are statistically significant at p < .05.

1

The percent of European American mothers and secure infants are reported in the mean column and point biserial correlations are reported.

Hypothesis Testing

Hypotheses were evaluated by conducting logistic and linear regressions with Mplus version 7.4 (Muthén & Muthén, 2012). Logistic regression was used in the model predicting the secure/insecure dichotomy and linear regressions were conducted separately for avoidant, resistant, and disorganized behaviors respectively (Table 2). Maternal sensitivity to distress and to nondistress at 6 months, and infant negative affect at 6 months were specified as predictors, as were the three product terms representing the interactions between maternal sensitivity to distress and to nondistress, maternal sensitivity to distress and infant negative affect, and maternal sensitivity to nondistress and infant negative affect. Race, education, income-to-needs ratio, maternal sensitivity to distress and to nondistress at 1 year were specified as control variables. Race was also tested as a possible moderator of all proposed main and interactive effects in initial models. None of the two-way or three-way interactions with race were statistically significant. Thus, these are discussed no further.

Table 2.

Standardized coefficients for models predicting disorganization, attachment security, avoidance, and resistance (N = 259)

Security Dichotomous Avoidance Resistance Disorganization
β R2 β R2 β R2 β R2
Block 1: Covariates .05 .05 .09* .03
 Race −.03 .09 .07 .09
 Maternal education .19 −.19* .08 −.15┼
 Income-to-needs ratio .04 −.08 .09 −.02
 Sensitivity to distress 1Y .11 −.06 −.05 −.06
 Sensitivity to non-distress 1Y −.13 .04 .20** .13
Block 2: Primary Variables .05 .03 .01 .08*
 Sensitivity to distress 6M .13 −.08 −.10 −.12
 Sensitivity to non-distress 6M .16 .11 −.01 .04
 Infant negative affect 6M (IBQ) −.05 .15┼ .03 .27**
Block 3: Two-way interactions .06* .04* .07* .02*
 Sensitivity to distress 6M × IBQ .27* −.08 −.17* −.17┼
 Sensitivity to non-distress 6M × IBQ −.01 .14 −.11 .08
 Sen distress × Sen non-distress 6M .09 −.18* .12 −.11
Block 4: Three-way interactions .00 .00 .00 .01
 Sen distress × Sen non-distress 6M × IBQ −.02 .00 −.01 −.08
Total R2 .16* .12** .17** .14**
Open in a new tab

Note. 6M = 6 months; 1Y = 1 year.

┼

p < .10.

*

p < .05.

**

p < .01.

Security

Contrary to hypotheses, maternal sensitivity to distress at 6 months was not significantly associated with security as a main effect, nor was sensitivity to non-distress when considered simultaneously. However, sensitivity to distress interacted with infant negative emotionality in relation to attachment security as predicted. As illustrated in Figure 1, sensitivity to distress was positively associated with security among infants high on negative emotionality (+1 SD, b = .21, β = .07, p = .07; this simple slope was significant at p < .05 when negative emotionality was 1.1 SD or more above the mean), but not among infants low on negative emotionality (−1 SD, b = .02, β = .01, p = .52). The parallel interaction between sensitivity to non-distress and infant negative emotionality was not statistically significant. In follow-up analyses, this pattern of findings was replicated with the alternative security dichotomous score in which secure/disorganized infants were categorized as insecure.

Figure 1.

Figure 1

Open in a new tab

Maternal sensitivity to distress and infant negative emotionality interact to predict attachment security classification.

Avoidance

Contrary to hypotheses, maternal sensitivity to distress at 6 months was not significantly associated with avoidance, nor was sensitivity to non-distress. Neither maternal sensitivity to distress nor to nondistress interacted significantly with infant negative emotionality in relation to avoidance. Consistent with prediction, the interaction between maternal sensitivity to distress and to nondistress was statistically significantly related to avoidance. As illustrated in Figure 2, maternal sensitivity to nondistress was associated positively with avoidance when maternal sensitivity to distress was low (−1SD, b = 1.90, β = .27, p = .04) but was not significantly associated with avoidance when maternal sensitivity to distress was high (+1SD, b = −.52, β = −.08, p = .58).

Figure 2.

Figure 2

Open in a new tab

Maternal sensitivity to distress and to non-distress interact to predict avoidant behavior.

Resistance

Maternal sensitivity to distress at 6 months was not significantly associated with resistance as a main effect, nor was sensitivity to nondistress. Consistent with prediction, sensitivity to distress interacted with negative emotionality in relation to resistance. As illustrated in Figure 3, maternal sensitivity to distress at 6 months was associated negatively with resistance when mother-reported infant negative emotionality was high (+1 SD, b = −.45, β = −.41, p = .007) but was not associated with resistance when infant negative emotionality was low (−1 SD, b = .08, β = .07, p = .70). The interactions between maternal sensitivity to nondistress and infant negative emotionality and between maternal sensitivity to distress and to nondistress were not significantly associated with resistance. Counter to expectation, maternal sensitivity to nondistress at 1 year was positively associated with resistance.

Figure 3.

Figure 3

Open in a new tab

Maternal sensitivity to distress and infant negative emotionality interact to predict resistant behavior.

Disorganization

Maternal sensitivity to distress at 6 months was not significantly associated with disorganization as a main effect, nor was sensitivity to nondistress. However, infant negative emotionality was positively associated with disorganization. Sensitivity to distress interacted with negative emotionality in relation to disorganization, although the association was marginal. Probing indicated that this marginal interaction operated similar to the interaction illustrated in Figure 3. That is maternal sensitivity to distress at 6 months was associated negatively with disorganization when mother-reported infant negative emotionality was high (+1 SD, b = −1.75, β = −.32, p = .006) but was not associated with disorganization when infant negative emotionality was low (−1 SD, b = −.11, β = −.02, p = .87). The interactions between maternal sensitivity to nondistress and infant negative affect, and between maternal sensitivity to distress and to nondistress were not significantly associated with disorganization.

Discussion

The goal of the current paper was to extend prior research by (1) examining the unique role of sensitivity to infant distress cues versus to non-distress cues in relation to dimensions of attachment insecurity and disorganization, (2) directly testing Main (1990) and Cassidy’s (1994) proposition that specific patterns of maternal affective communication may lead to attachment avoidance versus resistance via the development of infants’ conditional strategies, and (3) examining infant negative emotionality as a moderator of associations between maternal sensitivity and attachment outcomes. Although no main effects were apparent for sensitivity to distress (or non-distress) in relation to attachment dimensions, sensitivity to distress played a significant role in predicting attachment outcomes when considered in conjunction with sensitivity to non-distress and infant negative emotionality.

Patterns of Sensitivity

Consistent with prediction, infants whose mothers engaged in high levels of sensitivity to non-distress coupled with relatively low levels of sensitivity to distress at 6 months tended to avoid contact with their mothers during the strange situation. This pattern is highly consistent with Cassidy’s (1994) notion that maternal minimization of negative infant emotion may underlie the development of infant attachment avoidance. That is, infants who encounter this pattern of selective sensitive responding primarily when they are in a positive or neutral state may learn that they cannot count on their mothers to help them when they are distressed. To our knowledge, this is the second study to demonstrate that this pattern of high sensitivity to non-distress coupled with low sensitivity to distress is linked with negative infant outcomes (Leerkes et al., 2009). In contrast, the proposition that high sensitivity to distress coupled with low sensitivity to non-distress would predict infant attachment resistance was not supported. Thus, there was more evidence for one type of conditional strategy than the other (Main, 1990) in this sample.

The Role of Infant Negative Emotionality

Consistent with prediction, associations between maternal sensitivity to infant distress and three of the four attachment outcomes varied as a function of infant negative emotionality. Maternal sensitivity to distress was associated with greater odds of infant attachment security and lower ratings of infant resistance and disorganization only among infants who were high in negative emotionality. This is consistent with the view that infants who are high in negative emotionality may be more strongly influenced by maternal sensitivity to distress because they are highly dependent on their caregivers for external assistance to make them feel calm and safe or because they are distressed more frequently and hence exposed to this dimension of sensitivity more frequently in daily life than less easily distressed infants (Leerkes et al., 2009). Alternatively, this cross-over pattern may reflect differential susceptibility to the effects of sensitivity to distress as suggested by Groh, Narayan et al. (2017). For example, high sensitivity to distress was linked with higher odds of security and low sensitivity was linked with lower odds of security for infants high in negative emotionality relative to infants low in negative emotionality demonstrating the for better and for worse pattern that typifies differential susceptibility (Ellis et al., 2011). That this interaction effect was not replicated with sensitivity to non-distress provides support for a unique role of sensitivity to distress in relation to infant attachment outcomes, at least among infants high on negative emotionality.

The Absence of Sensitivity Main Effects

The lack of a main effects of sensitivity at either time point in relation to attachment security warrants discussion given well-documented longitudinal and concurrent associations between sensitivity and attachment in prior research (De Wolff & van IJzendoorn, 1997). First, that sensitivity to distress cues at 6 months was not associated with security as a main effect appears counter to prior research (del Carmen et al., 1993; Leerkes, 2011; McElwain & Booth-LaForce, 2006) but may reflect un-tested moderators or methodological differences. Notably, the subsample used in the report by McElwain and Booth-La Force was somewhat select in that the relatively small group of infants (357 out of 1,364) who became distressed during the play episodes were in fact higher in mother reported temperamental reactivity than those who did not. Thus, the reported main effect of sensitivity to distress in a group of temperamentally negative infants is in fact fairly consistent with the interaction effect in the current study. The current study is the only one that focused on ratings of sensitivity to distress and non-distress cues within tasks designed to elicit distress which may also explain discrepancies. In addition, the 3-point a prior sensitivity rating applied the current study, in contrast to global 5- (McElwain & Booth- LaForce, 2006) and 9- (Ainsworth et al., 1978) point rating systems may obscure meaningful variation in sensitivity. For example, in the current coding system all instances of intrusive behavior would receive a score of 1, but some may be relatively mild (e.g., drawing an infants’ attention to the frightening toy via words and gestures when they are otherwise engaged) and others may be quite egregious (e.g., physically forcing an infant to touch the frightening toy).

The null association between both aspects of maternal sensitivity assessed at 1 year and attachment security is surprising as is the positive association between maternal sensitivity to non-distress at 1 year and resistant behavior. These unexpected results may be a function of that fact that maternal sensitivity was observed in a series of tasks that occurred immediately after the Strange Situation. It may be that some mothers of insecure infants, particularly resistant infants, behaved more sensitively than typical in an effort to overcompensate for their infants’ negative behavior toward them during the Strange Situation. That is, they may have attempted to appear more sensitive and to elicit positive infant behavior to convey a positive parent-child relationship because of social desirability demands.

Applied Implications

The role of sensitivity to distress in relation to attachment outcomes in this and other studies (Leerkes, 2011; McElwain & Booth La Force, 2006) may have implications for attachment-based interventions. That is, they support the continued emphasis on in the moment coaching of mothers’ ability to read infant cues and sensitive responses to infant distress as is the case in the Attachment and Biobehavioral Catch Up (Roben, Dozier, Caron, & Bernard, 2017) and Video-feedback Intervention to Enhance Positive Parenting (Juffer, Bakermans-Kranenburg, & van IJzendoorn, 2008) interventions. Given that naturally occurring opportunities to coach maternal sensitivity to distress may be relatively rare (Bernard, Meade & Dozier, 2013), employing brief distress-eliciting tasks such as those used in the current report may be useful in intervention work designed to enhance parents’ sensitive responding to infant distress and hence infant attachment outcomes by increasing the likelihood that parent coaches can focus on one of the target maternal behaviors more frequently. The results of the current study also suggest that such interventions may be particularly effective for infants high in negative emotionality consistent with prior intervention research (Cassidy, Woodhouse, Sherman, Stupica & Lejuez, 2011; Klein Velderman, Bakermans-Kranenburg, Juffer, van IJzendoorn, 2006; van den Boom, 1994).

Limitations, Strengths, and Conclusions

The relatively brief observations of maternal sensitivity is a limitation; however, this is necessitated by the focus on distress-eliciting tasks which must be brief for ethical reasons. It could be argued that a focus on distress-eliciting tasks may be too narrow; however, the distress-eliciting tasks were effective at eliciting infant distress and non-distress from nearly all infants, in contrast to prior research based on free play tasks in which a minority of infants displayed distress (McElwain & Booth LaForce, 2006). The low rate of attachment insecurity in this community sample is also a limitation. Although taking a dimensional perspective is appropriate to address this, our results primarily reflect the extent to which the focal constructs predict variability in relatively low levels of avoidant, resistant, and disorganized behaviors in a community sample. Thus, replication in more at-risk samples with higher rates of insecurity and disorganization is needed. Strengths of the current study that set it apart from prior studies addressing similar questions (e.g., Leerkes, 2011; McElwain & Booth LaForce, 2006) include: (a) the diversity of the sample with respect to race, maternal education, and income, which increases generalizability; (b) the relatively conservative analytic approach in which concurrent maternal sensitivity was controlled to ensure the association between sensitivity at 6 months and attachment outcomes reflected longitudinal predictions; and (c) the focus on specific dimensions of attachment insecurity rather than solely the secure/insecure dichotomy.

To conclude, the results of this study extend prior research by clarifying the role of maternal sensitivity to infant distress cues in the development of specific attachment outcomes. Insensitive responses to infant distress are particularly predictive of attachment insecurity, resistance, and disorganization among infants high in negative emotionality. In contrast, attachment avoidance is particularly likely when mothers engage in high sensitivity to non-distress coupled with low sensitivity to distress which may convey mothers’ preference for the minimization of negative emotion.

Acknowledgments

This project was supported by R01HD058578. The contents of this manuscript are the sole responsibility of the authors and do not necessarily reflect the views of the Eunice Kennedy Shriver National Institute for Child Health and Human Development. We are grateful to the participants for their time and Dr. Regan Burney and project staff for their dedication. Portions of this article were presented at the April 2017 Biennial Meeting of the Society for Research in Child Development in Austin, TX.

References

  1. Ainsworth MDS, Blehar MC, Waters E, Wall S. Patterns of attachment: Assessed in the strange situation and at home. Hillsdale, NJ: Erlbaum; 1978. [Google Scholar]
  2. Bernard K, Dozier M, Bick J, Lewis-Morrarty E, Lindhiem O, Carlson E. Enhancing attachment organization among maltreated children: Results of a randomized clinical trial. Child Development. 2012;83:623–636. doi: 10.1111/j.1467-8624.2011.01712.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bernard K, Meade EB, Dozier M. Parental synchrony and nurturance as targets in an attachment based intervention: Building upon Mary Ainsworth’s insights about mother–infant interaction. Attachment & Human Development. 2013;15:507–523. doi: 10.1080/14616734.2013.820920. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bowlby J. Attachment and loss: Vol. 1: Attachment. New York: Basic Books; 1969. [Google Scholar]
  5. Cassidy J. Emotion regulation: Influences of attachment relationships. Monographs of the Society for Research in Child Development. 1994;59:228–283. doi: 10.2307/1166148. [DOI] [PubMed] [Google Scholar]
  6. Cassidy J, Woodhouse SS, Sherman LJ, Stupica B, Lejuez CW. Enhancing infant attachment security: An examination of treatment efficacy and differential susceptibility. Development and Psychopathology. 2011;23:131–148. doi: 10.1017/S0954579410000696. [DOI] [PubMed] [Google Scholar]
  7. del Carmen R, Pedersen FA, Huffman LC, Bryan YE. Dyadic distress management predicts subsequent security of attachment. Infant Behavior & Development. 1993;16:131–147. doi: 10.1016/0163-6383(93)80014-Y. [DOI] [Google Scholar]
  8. De Wolff M, van IJzendoorn MH. Sensitivity and attachment: A meta-analysis on parental antecedents of infant attachment. Child Development. 1997;68:571–591. doi: 10.2307/1132107. [DOI] [PubMed] [Google Scholar]
  9. Ellis BJ, Boyce WT, Belsky J, Bakermans-Kranenburg MJ, Van Ijzendoorn MH. Differential susceptibility to the environment: An evolutionary–neurodevelopmental theory. Development and Psychopathology. 2011;23:7–28. doi: 10.1017/S0954579410000611. [DOI] [PubMed] [Google Scholar]
  10. Fox NA. The role of individual differences in infant personality in the formation of attachment relationships. In: Susman EJ, Feagans LV, Ray WJ, Susman EJ, Feagans LV, Ray WJ, editors. Emotion, cognition, health, and development in children and adolescents. Hillsdale, NJ, US: Lawrence Erlbaum Associates, Inc; 1992. pp. 31–52. [Google Scholar]
  11. Gedaly LR, Leerkes EM. The role of sociodemographic risk and maternal behavior in the prediction of infant attachment disorganization. Attachment & Human Development. 2016;18:554–569. doi: 10.1080/14616734.2016.1213306. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Goldberg S, Grusec JE, Jenkins JM. Confidence in protection: Arguments for a narrow definition of attachment. Journal of Family Psychology. 1999;13:475–483. doi: 10.1037/0893-3200.13.4.475. [DOI] [Google Scholar]
  13. Groh AM, Narayan AJ, Bakermans-Kranenburg MJ, Roisman GI, Vaughn BE, Fearon RMP, Van IJzendoorn MH. Attachment and temperament in the early life course: A meta-analytic review. Child Development. 2017;88:770–795. doi: 10.1111/cdev.12677. [DOI] [PubMed] [Google Scholar]
  14. Groh AM, Fearon RP, IJzendoorn MH, Bakermans-Kranenburg MJ, Roisman GI. Attachment in the early life course: Meta-analytic evidence for its role in socioemotional development. Child Development Perspectives. 2017;11:70–76. doi: 10.1111/cdep.12213. [DOI] [Google Scholar]
  15. Grusec JE, Davidov M. Integrating different perspectives on socialization theory and research: A domain-specific approach. Child Development. 2010;81:687–709. doi: 10.1111/j.1467-8624.2010.01426.x. [DOI] [PubMed] [Google Scholar]
  16. Juffer F, Bakermans-Kranenburg MJ, van IJzendoorn MH. Promoting positive parenting: An attachment-based intervention. New York, NY: Taylor & Francis Group/Lawrence Erlbaum Associates; 2008. [Google Scholar]
  17. Klein Velderman M, Bakermans-Kranenburg MJ, Juffer F, van IJzendoorn MH. Effects of attachment-based interventions on maternal sensitivity and infant attachment: Differential susceptibility of highly reactive infants. Journal of Family Psychology. 2006;20:266–274. doi: 10.1037/0893-3200.20.2.266. [DOI] [PubMed] [Google Scholar]
  18. Leerkes EM. Predictors of maternal sensitivity to infant distress. Parenting: Science And Practice. 2010;10:219–239. doi: 10.1080/15295190903290840. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Leerkes EM. Maternal sensitivity during distressing tasks: A unique predictor of attachment security. Infant Behavior & Development. 2011;34:443–446. doi: 10.1016/j.infbeh.2011.04.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Leerkes EM, Blankson AN, O’Brien M. Differential effects of maternal sensitivity to infant distress and nondistress on social-emotional functioning. Child Development. 2009;80:762–775. doi: 10.1111/j.1467-8624.2009.01296.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Leerkes EM, Su J, Calkins SD, Supple AJ, O'Brien M. Pathways by which mothers’ physiological arousal and regulation while caregiving predict sensitivity to infant distress. Journal of Family Psychology. 2016;30:769–779. doi: 10.1037/fam0000185. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lyons-Ruth K, Jacobvitz D. Attachment disorganization from infancy to adulthood: Neurobiological correlates, parenting contexts, and pathways to disorder. In: Cassidy J, Shaver PR, editors. Handbook of Attachment: Theory, research, and Clinical Applications. 3. New York, NY: Guilford Press; 2016. pp. 667–695. [Google Scholar]
  23. Main M, Kaplan N, Cassidy J. Security in infancy, childhood, and adulthood: A move to the level of representation. In: Bretherton I, Waters E, editors. Growing points in attachment theory and research. Monographs of the Society for Research in Child Development. Vol. 50. 1985. pp. 66–106. [DOI] [Google Scholar]
  24. Main M, Solomon J. Procedures for identifying infants as disorganized/disoriented during the Ainsworth Strange Situation. In: Greenberg MT, Cicchetti D, Cummings E, editors. Attachment in the preschool years: Theory, research, and intervention. Chicago, IL US: University of Chicago Press; 1990. pp. 121–160. [Google Scholar]
  25. McElwain NL, Booth-LaForce C. Maternal sensitivity to infant distress and nondistress as predictors of infant-mother attachment security. Journal of Family Psychology. 2006;20:247–255. doi: 10.1037/0893-3200.20.2.247. [DOI] [PubMed] [Google Scholar]
  26. Mesman J, van IJzendoorn MH, Bakermans-Kranenburg MJ. Unequal in opportunity, equal in process: Parental sensitivity promotes positive child development in ethnic minority families. Child Development Perspectives. 2012;6:239–250. doi: 10.1111/j.1750-8606.2011.00223.x. [DOI] [Google Scholar]
  27. Muthen K, Muthen BO. Mplus user’s guide. 7. Los Angeles, CA: Author; 1998–2012. [Google Scholar]
  28. Out D, Bakermans-Kranenburg MJ, Van IJzendoorn MH. The role of disconnected and extremely insensitive parenting in the development of disorganized attachment: Validation of a new measure. Attachment & Human Development. 2009;11:419–443. doi: 10.1080/14616730903132289. [DOI] [PubMed] [Google Scholar]
  29. Putnam SP, Helbig AL, Gartstein MA, Rothbart MK, Leerkes E. Development and assessment of Short and Very Short Forms of the Infant Behavior Questionnaire Revised. Journal of Personality Assessment. 2014;96:445–458. doi: 10.1080/00223891.2013.841171. [DOI] [PubMed] [Google Scholar]
  30. Roben CP, Dozier M, Caron EB, Bernard K. Moving an evidence-based parenting program into the community. Child Development. 2017;88:1447–1452. doi: 10.1111/cdev.12898. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Rothbart MK, Bates JE. Temperament. In: Eisenberg N, Damon W, Lerner RM, Eisenberg N, Damon W, Lerner RM, editors. Handbook of child psychology: Social, emotional, and personality development. 6. Vol. 3. Hoboken, NJ, US: John Wiley & Sons Inc; 2006. pp. 99–166. [Google Scholar]
  32. Schafer JL, Graham JW. Missing data: Our view of the state of the art. Psychological Methods. 2002;7:147–177. doi: 10.1037/1082-989X.7.2.147. [DOI] [PubMed] [Google Scholar]
  33. Thompson RA. Sensitivity and security: New questions to ponder. Child Development. 1997;68:595–597. doi: 10.2307/1132109. [DOI] [Google Scholar]
  34. van den Boom DC. The influence of temperament and mothering on attachment and exploration: An experimental manipulation of sensitive responsiveness among lower-class mothers with irritable infants. Child Development. 1994;65:1457–1477. doi: 10.2307/1131511. [DOI] [PubMed] [Google Scholar]
  35. Vaughn BE, Bost KK, van IJzendoorn MH. Attachment and temperament as intersecting developmental products and interacting developmental contexts throughout infancy and childhood. In: Cassidy J, Shaver PR, editors. Handbook of attachment: Theory, research, and clinical applications. 3. New York, NY, US: Guilford Press; 2016. pp. 202–222. [Google Scholar]
  36. Verhage ML, Schuengel C, Madigan S, Fearon RP, Oosterman M, Cassibba R, … van IJzendoorn MH. Narrowing the transmission gap: A synthesis of three decades of research on intergenerational transmission of attachment. Psychological Bulletin. 2016;142:337–366. doi: 10.1037/bul0000038. [DOI] [PubMed] [Google Scholar]
  37. von Hippel PT. New confidence intervals and bias comparisons show that maximum likelihood can beat multiple imputation in small samples. Structural Equation Modeling: A Multidisciplinary Journal. 2016;23:422–437. doi: 10.1080/10705511.2015.1047931. [DOI] [Google Scholar]

RESOURCES