How You Were Raised Shapes How You Parent — The Intergenerational Transmission of Attachment

Lead

"Am I going to parent the way I was parented?"

Many people ask this question for the first time the moment they become a parent. Whether the memories of your own upbringing are mostly warm or carry real wounds, those memories have a way of surfacing at unexpected moments — the late-night feed, the moment you lose your patience, the first time your child's cry sounds accusatory. Why do people so often reproduce, without consciously choosing to, the patterns in which they were raised? And can that pattern be interrupted?

This is not merely a psychoanalytic intuition. Since the 1980s it has been the subject of sustained empirical investigation.

What intergenerational transmission of attachment means

One of the central questions in attachment research is whether the patterns of attachment formed between a young child and their caregiver carry forward across generations.

In 1984, Mary Main and Ruth Goldwyn developed the Adult Attachment Interview (AAI), a structured clinical interview for assessing adults' own attachment experiences [1]. The AAI asks respondents to describe their childhood relationships with caregivers and then analyzes not so much the content of what is recalled but the narrative coherence — how fluently and consistently the speaker integrates both the emotional and factual dimensions of memory. The resulting classifications — secure-autonomous, dismissing, preoccupied, and unresolved — map onto the infant Strange Situation categories (secure, avoidant, ambivalent, and disorganized) with notable correspondence.

The quantitative case for this correspondence was made by Marinus van IJzendoorn's 1995 meta-analysis [2]. Across 18 samples (N = 854), van IJzendoorn found a statistically moderate effect size (r ≈ .47) between parents' AAI classification and their infants' attachment classification, replicated independently across multiple research groups. A parent's own mental representation of attachment, it turned out, significantly predicted the attachment relationship that parent formed with their child.

The meta-analysis also identified a problem that would come to be called the "transmission gap." Parental sensitivity — the appropriate, timely response to a child's signals — was assumed to be the main mediating pathway. But in the data, sensitivity alone explained only about half of the transmission effect [2]. The other half remained unexplained, presumably mediated by variables the studies had not measured.

Thirty years of data recalibrate the picture

In 2016, Verhage and colleagues published a large-scale meta-analysis in Psychological Bulletin integrating three decades of research [3]. Across 95 samples (N = 4,819), the effect size for intergenerational transmission of secure attachment was r = .31; for unresolved transmission it was r = .21 — both smaller than the 1995 figures. This reduction reflects the greater methodological rigor of later studies, not a reversal of the finding itself.

The honest summary of where the evidence stands: intergenerational transmission of attachment is real, but its magnitude is moderate.

A separate meta-analysis by Fearon and colleagues (2010) found that insecure and disorganized attachment is associated with children's externalizing behavior problems [4]. Attachment quality does have downstream effects on behavioral development, but the causal pathways are complex, and the relationship is not a simple deterministic one.

The empirical case for breaking the pattern

The most practically important implication of this research for parents is that you are not obligated to parent the way you were parented.

The evidence for this lies partly in the effect sizes themselves. As noted above, the correlation between a parent's AAI classification and their child's attachment classification is in the range of r ≈ .31 — which accounts for roughly 10% of the variance. The remaining 90% is determined by other factors: genetic influences, environmental variables, relationships outside the primary caregiver pair, and chance experiences.

The more important point involves the nature of the AAI itself. The interview evaluates narrative coherence, not the valence of the experiences recalled. A person who had a difficult childhood but can speak about it with consistency and emotional integration — neither dismissing the pain nor being overwhelmed by it — is classified as secure-autonomous, and this classification predicts secure attachment in their children. In other words, what predicts the next generation's attachment pattern is not whether hard things happened but how those experiences have been processed and made sense of. The chain of transmission is not fated.

Clinically, interventions that support parents in reflecting on and making sense of their own attachment history — attachment-based parenting programs — have been shown in multiple studies to increase rates of secure attachment in children [3]. The act of reflecting, articulating, and reconsidering one's relational history is itself a mechanism that can shift the pattern.

What this looks like in daily life

Research on intergenerational transmission can be brought into daily life through questions, not diagnoses.

"When I was a child, who could I turn to?" "When I was frightened, who responded?" "What habits did I develop from the times when no one responded?" These are not invitations for psychoanalytic excavation — they are reference points for observing your own reaction patterns in the present.

Some parents notice, for instance, that when their child cries inconsolably, a memory surfaces of being told to stop crying, and that an impulse arises to repeat exactly that. The first step available is simply noticing that this reaction is automatic — that it is coming from somewhere outside the present moment. That noticing alone creates space for a different choice.

Keeping a record — of daily interactions, of your own responses to your child's distress, of the patterns you observe in yourself — can support this kind of self-observation. A log entry is not therapy, but it can serve as a mirror.

Summary

The intergenerational transmission of attachment is empirically established. Its effect size is moderate, and the chain is not deterministic. Difficult experiences from the past, when they can be narrated and understood, become a point where transmission can be interrupted rather than continued.

You do not have to parent the way you were parented. But the ways in which you were parented are worth examining — not to assign blame, but because understanding them is one of the clearest paths available.

References

1. Main M, Goldwyn R. Adult attachment scoring and classification system. Unpublished manuscript, University of California at Berkeley; 1984/1998. [For published version, see: Main M, Hesse E, Goldwyn R. In: Steele H, Steele M, eds. Clinical Applications of the Adult Attachment Interview. New York: Guilford Press; 2008.]
2. van IJzendoorn MH. Adult attachment representations, parental responsiveness, and infant attachment: a meta-analysis on the predictive validity of the Adult Attachment Interview. Psychol Bull. 1995;117(3):387–403. PMID: 7777645. doi:10.1037/0033-2909.117.3.387
3. Verhage ML, Schuengel C, Madigan S, et al. Narrowing the transmission gap: a synthesis of three decades of research on intergenerational transmission of attachment. Psychol Bull. 2016;142(4):337–366. PMID: 26653864. doi:10.1037/bul0000038
4. Fearon RP, Bakermans-Kranenburg MJ, van IJzendoorn MH, Lapsley A-M, Roisman GI. The significance of insecure attachment and disorganization in the development of children's externalizing behavior: a meta-analytic study. Child Dev. 2010;81(2):435–456. PMID: 20438450. doi:10.1111/j.1467-8624.2009.01405.x
5. George C, Kaplan N, Main M. Adult Attachment Interview. Unpublished manuscript, University of California at Berkeley; 1984. [Protocol source document; same context as Main & Goldwyn 1984.]