Sensitivity and Attachment: Next Steps for Developmentalists

by Dymphna C. van den Boom
Sensitivity and Attachment: Next Steps for Developmentalists
Dymphna C. van den Boom
Child Development
Start Page: 
End Page: 
Select license: 
Select License

Child Development, August 1997, Volume 64, Number 4, Pages 592-594

Sensitivity and Attachment: Next Steps for Developmentalists

Dymphna C. van den Boom

This commentary focuses on rethinking the definition of sensitivity and addresses the issue of developmental change in sensitivity. Both topics will be discussed with regard to the clinical implications of attachment.


The relation between sensitivity and attachment is a crucial phenomenon for Bowlby's (1969) account of the development of attachment. He argued that ba- bies gradually develop an attachment relationship with a principal attachment figure and suggested that the attachment figure's sensitivity might be critical for the establishment of a secure attachment. It is not surprising, therefore, that investigators have studied this topic intensively in the last 25 years, initially to replicate Bowlby's theoretical accounts, but more re- cently to substantiate alternative interpretations (e.g., Goldsmith & Alansky, 1987). Lately, the sheer vol- ume of the research has defied easy synthesis, and investigators have tended to advance interpretations around their own findings rather than offering an ex- haustive account. Consequently, there has been a danger that research would peter out in the wake of an unwieldy and apparently contradictory set of data.


The great virtue of the study by De Wolff and van IJzendoorn (1997) is that they have restored order to the field. Moreover, the order is simple and compel- ling. In normal settings, sensitivity is an important but not exclusive condition of attachment security. The authors arrive at this conclusion by making clever and sensible cuts in a potentially overwhelm- ing data set.

We gain this order at some cost. Let me add that I know- that the authors were probably under pres- sure to condense the enormous amount of data to manageable proportions and that all possible ways to analyze the data could not be explored. Although attachment theory is developmental in nature, devel- opmental change is dealt with in a nondevelopmental way. Whenever the same maternal behavior was assessed at multiple points in time, the mean

Commentary on De Wolff and van IJzendoorn, "Sensitivity and Attachment: A Meta-Analysis on Parental Antecedents of Infant Attachment."

outcome across the times of assessment was included. Nevertheless, the authors included several studies in their meta-analysis that both support and fail to support the sensitivity-security linkage de- pending on the time of assessment (e.g., Belsky, Rov- ine, & Taylor, 1984; Egeland & Farber, 1984; Miyake, Chen, & Campos, 1985). Such findings may indicate that this relation is significant for only part of the range. Or they may be due to attempts to impose an analysis of linear trends on nonlinear re1ations:Rob- erts (1986) demonstrated, for instance, the existence of a threshold effect in parental warmth, that is, very low levels of warmth are associated with a deficit in child competence. Crossing the threshold results in a rapid rise in levels of competence. This is followed by a plateau in which further increases in parental warmth have little effect on child competence. It sug- gests that an almost exclusive reliance upon linear relations that presume that more of a good thing is always in the best interest of the child (e.g., Belsky et al., 1984; van den Boom, 1994) does not seem to be justified in the case of parental warmth. The few studies that document curvilinear relations between indices of the parent-child relationship and child de- velopment suggest that there may be a point of di- minishing returns. The findings of this meta-analysis have demonstrated that sensitivity is indeed an im- portant aspect of parenting. However, its exact rela- tion to attachment security still needs further consid- eration. To address this topic in future research, rethinking the definition of sensitivity and a develop- mental approach seem to be required.


The first study of the meta-analysis shows what a highly complex phenomenon sensitivity is. Numer- ous operationalizations exist. Some define sensitivity as a unitary characteristic, whereas others conceive of it as an umbrella term for a number of only loosely

O 1997 by the Society for Research in Child Development, Inc. All rights reserved. 0009-3920 /97/6804-0012$01.00

associated aspects of parenting. Some consider sensi- tivity to be a disposition of the parent, others concep- tualize sensitivity as a characteristic that pertains to the dyad. Findings from several diverse research set- tings demonstrate that the individual scores of two persons in a dyad reflect both the child's skill in sig- naling and parental readiness to respond (e.g., Schaf- fer & Crook, 1980; van den Boom & Hoeksma, 1994). Thus, sensitivity is basically a statement about the in- teraction and, hence, is meaningless without refer- ence to both partners. The fact that differences in associations between parental antecedents and attachment are obtained in normal versus special samples further corroborates this conclusion. The implications of such a conceptualization are twofold. In- stead of parental behavior being examined in its own right, it needs to be studied as part of a dyadic unit. This conceptualization also implies that it would be more fruitful to think of sensitivity not as a parenting dimension that exists apart from other dimensions but, rather, as permeating all interactive behavior. The sensitive parent must package her social interac- tive behavior in such a way in the interaction flow that it will promote rather than interrupt the ex- change. This holds equally for stimulation, play, in- struction, control, and so on. For instance, controlling a child's behavior through threat would be insensi- tive, whereas control consisting of limit setting ac- companied by a reasonable explanation to the child can considered to be sensitive. Promptness, consis- tency, and appropriateness are the main constituents that can be used to arrive at a definition of sensitivity across parenting dimensions. That reliable measures can be obtained that way has been demonstrated in previous research (e.g., Riksen-Walraven, 1978; van den Boom & Hoeksma, 1994). Such a conceptualiza- tion is also a good approximation of Ainsworth's original definition. Instead of focusing future efforts on a multidimensional approach of attachment ante- cedents where sensitivity is only one of the dimen- sions, as De Wolff and van IJzendoorn propose, I sug- gest that sensitivity be conceptualized as outlined above. It seems likely that this will enhance the asso- ciation between parenting antecedents and attach- ment such that it could perhaps approach the values obtained in Ainsworth's pioneering Baltimore study.


Another implication of the dependence of parent and child behavior is that it makes the study of infant antecedents to attachment equally important. If parent and infant both contribute to the quality of attach- ment, the sorts of questions that arise concern the way in which the parent interweaves her behavior

Dymphna C. van den Boom 593

with that of the child's. This does not imply, of course, that the two are in any sense to be seen as equal partners. Inevitably the parent compensates for the child's initial state of immaturity. Nevertheless, that characteristics of the child play a role in the inter- action has been shown repeatedly, despite the fact that untangling the causes from the effects may be impossible. The inclusion of infant antecedents in in- quiries on the origins of attachment necessitates the study of developmental change. The advantage of a developmental approach is that it leads to a specifi- cation of behavioral manifestations of (in)sensitivity. The concept as such is too abstract and too coarse to capture the subtle nuances of parental behavior, and it lacks sufficient contextual anchorage. Then, it be- comes possible to examine differences in behavioral manifestations of (in)sensitivity across (risk) groups, to examine change in behavioral manifestations de- pending on the child's developmental stage, and to examine the different routes that lead to broadly sim- ilar endpoints within and across groups.

A focus on developmental change not only leads to a better understanding of the nature of the parent- child relationship but may also yield valuable infor- mation for practical efforts made to bring about im- provements in that relationship. In the majority of studies that were included in the meta-analysis, the temporal order of the measurements is fixed by de- sign, that is, sensitivity is measured prior to attach- ment (van den Boom & Hoeksma, 1997). Because sen- sitivity is generally considered to be a pretty stable characteristic of the parent, whereas the child's at- tachment behavior is supposed to change with age, fewer measurements would be required to capture changes in parental compared with child behavior. Paradoxically, many studies on attachment anteced- ents offer the opposite picture, that is, sensitivity is observed more frequently than the child's attachment behavior (Hoeksma, van den Boom, Koomen, & Koops, 1997). If the mother-child dyad is perceived as a dynamically changing dyadic unit, restoring the balance is required in this regard. By pinpointing down the behavioral manifestations of insensitivity across development in case of attachment difficulties, it may become clear how attachment concepts should shape treatment. In another meta-analysis van IJzen- doorn, Juffer, and Duyvesteyn (1995) showed that be- haviorally oriented, short-term preventive interven- tions in a variety of risk populations are more effective in changing parental insensitivity and chil- dren's attachment insecurity than longer, more inten- sive, and therapeutic approaches. In fact, those inter- vention programs that were preceded by a detailed observational study to determine age-related behav- ioral manifestations of insensitivity, timing, and mo-

594 Child Development

tivation were most effective (e.g., van den Boom, 1994). That meta-analysis also shows the limited gen- eralizability of intervention programs aimed at en- hancing sensitivity. Implementation of the same in- tervention in different samples yielded no or only modest effects (e.g., Juffer, 1993; Meij, 1992). This sug- gests that in different samples different develop- mental processes occur. The study of developmental change in maternal and infantile attachment anteced- ents may shed light on this issue. In addition, it may help answering the question, How sensitive is sensi- tive enough to foster a secure attachment relation- ship? Although the results of this meta-analysis dem- onstrate that sensitivity is an important aspect of parenting, there is nothing to indicate that the level of sensitivity required for a secure attachment must be 100%. We do not know yet what the clinical cutoff point is in different (risk) groups.

Few researchers have the courage to embark on an enterprise of the scope that this meta-analysis repre- sents. It is a major achievement, and it will provide, I hope, a stable backdrop against which investigators can continue to debate.


Corresponding author: Dymphna C. van den Boom, University of Amsterdam, Department of Education, Wibautstraat 4, 1091 GM Amsterdam, The Nether- lands.


Belsky, J., Rovine, M., & Taylor, D. G. (1984). The Pennsyl-

vania Infant and Family Development Project, In: The

origins of individual differences in infant-mother at-

tachment: Maternal and infant contributions. Child

Development, 55, 718-728. Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. New York: Basic Books. De Wolff, M. S., &van IJzendoorn, M. H. (1997). Sensitivity

and attachment: A meta-analysis on parental anteced-

ents of infant attachment. Child Development, 68, 571-


Egeland, B., & Farber, E. A. (1984). Infant-mother attach- ment: Factors related to its development and changes over time. Child Development, 55, 753-771.

Goldsmith, H. H., & Alansky, J. A. (1987). Maternal and infant temperamental predictors of attachment: A meta- analytic review. Journal ofconsulting and Clinical Psychol- ogy, 55, 805-816.

Hoeksma, J. B., van den Boom, D. C., Koomen, H. M. Y., & Koops, W. (1997). Modeling the sensitivity-attachment hypothesis. In W. Koops, J. B. Hoeksma, & D. C. van den Boom (Eds.), Development of interaction and attach- ment: Traditional and non-traditional approaches. Amsterdam: North-Holland.

Juffer, F. (1993). Verbonden door adoptie: Een experimenteel onderzoek naar hechting en competentie in gezinnen met een adoptiebaby [Attached through adoption: An experimen- tal study on attachment and competence in families with an adopted baby]. Amersfoort: Academische uitgeverij.

Meij, J. Th.H. (1992). Sociale ondersteuning, gehechtheidskwal- iteit en vroegkinderlijke competentie-ontwikkeling [Social support, attachment, and early competence]. Unpub- lished doctoral dissertation, Catholic University, Nij- megen.

Miyake, K., Chen, S. J., & Campos, J. J. (1985). Infant tem- perament, mother's mode of interaction, and attachment in Japan: An interim report. In I. Bretherton & E. Waters (Eds.), Growing points in attachment theory and re- search (pp. 276-297). Monographs of the Societyfor Child Development, 50(1-2, Serial No. 209).

Riksen-Walraven, J. M. A. (1978). Effects of caregiver be- havior on habituation rate and self-efficacy in infants. International Journal of Behavioural Development, 1, 105-


Roberts, W. L. (1986). Nonlinear models of development: An example from the socialization of competence. Child Development, 57,1166-1178.

Schaffer, H. R., & Crook, C. K. (1980). Child compliance and maternal control techniques. Developmental Psychol- ogy, 16,54-61.

van den Boom, D. C. (1994). 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, 65, 1449-1469.

van den Boom, D. C., & Hoeksma, J. B. (1994). The effect of infant irritability on mother-infant interaction: A growth-curve analysis. Developmental Psychology, 30, 581-590.

van den Boom, D. C., & Hoeksma, J. B. (1997). How is de- velopment conceptualized in mother-child interaction research? In W. Koops, J. B. Hoeksma, & D. C. van den Boom (Eds.), Development of interaction and attachment: Traditional and non-traditional approaches. Amsterdam: North-Holland.

van IJzendoorn, M. H., Juffer, F., & Duyvesteyn, M. G. C. (1995). Breaking the intergenerational cycle of insecure attachment: A review of the effects of attachment-based interventions on maternal sensitivity and infant secu- rity. Journal of Child Psychology and Psychiatry, 36, 225-


  • Recommend Us