There is a kind of familiarity that happens when we meet someone new.
A person says something in a particular way, responds to a text a little later than expected, becomes quiet during a disagreement, or offers reassurance at exactly the right moment. Before I know very much about them, I already have a feeling about what their behavior means. They are pulling away. They are dependable. I need to be careful. I can trust them.
But where do these interpretations come from?
When we enter a new relationship, we might like to imagine that we are starting from the beginning. However, we bring memories of parents, friends, previous partners, and other important relationships with us. These experiences can shape not only what we expect from other people, but also what we notice, how we interpret ambiguous behavior, and even which version of ourselves emerges around someone new.
Attachment Theory
Attachment theory proposes one way for understanding how our first relationships in childhood influence later ones. Bowlby proposed that repeated experiences with important caregivers contribute to internal working models: mental representations involving expectations about ourselves, other people, and relationships (Bowlby, 1969).
These models may be more than just a set “attachment style.” Pietromonaco and Feldman Barrett (2000) described working models as representations that can contain memories of interpersonal experiences, the emotions connected to them, and beliefs about how others are likely to respond. They may guide what information we pay attention to, how we interpret interpersonal situations, and what we remember. For example, a person who has experienced others around them becoming unavailable during conflicts may interpret a new partner becoming quiet after an argument as: they are withdrawing, something is wrong, I might be abandoned. Another person, with a different interpersonal history, may interpret the same event as their partner needing some time alone.
Longitudinal Findings
There is some evidence that attachment-related representations formed earlier in development can show associations with romantic relationships many years later.
Dagan and colleagues (2024) examined data from a large longitudinal study in which participants completed the Adult Attachment Interview around age 18 and were followed into adulthood. The researchers were particularly interested in secure base script knowledge: knowledge of a relational sequence in which distress leads someone to seek support and receive comfort. Among 479 participants who were romantically partnered around age 30, greater secure base script knowledge at approximately age 18 was associated with better self-reported romantic relationship adjustment roughly a decade later.
However, it is important to note that early attachment styles are not deterministic. Earlier relationships and representations of them seem to matter, but they explain only a part of the large variability in how adult relationships develop.
New People, Old Representations
It is even more interesting when we move from general attachment representations to specific people. Brumbaugh and Fraley (2006) tested whether attachment patterns associated with a previous romantic partner could transfer to a novel potential partner. Participants learned about two potential dating partners: one designed to resemble a previous romantic partner and another resembling somebody else’s former partner. Participants applied attachment representations from previous partners to both novel targets, but the effect was stronger when the new person resembled their own past partner. They also reported more attachment anxiety and less avoidance toward the ex-partner-like targets.
Gillath and colleagues’ (2025) recent review brings these findings together and argues that attachment representations can transfer in different ways depending on the relationship. Some effects appear relatively general, whereas others become especially strong when a new person resembles a particular significant other.
We Do Not Only Carry Other People with Us
We may not only carry representations of other people from one relationship into another. We may also carry representations of ourselves in relation to them.
The social-cognitive model of the relational self proposes that different relationships can activate different versions of the self (Andersen & Chen, 2002). Different relationships bring different emotions, goals, behaviors, and beliefs about us. Gillath and colleagues (2025) integrate this idea with attachment theory. They describe our relational representations as existing at multiple levels: from broad attachment orientations to relationship domains, to models connected to specific people. This means that when a new partner reminds us of someone from our past, it may also activate patterns of thought and behavior from a previous relationship. Thus, relationships may not only recreate familiar others but familiar selves.
Relationship Patterns Are Not Inevitable
Although past relationships can shape how we understand new ones, these patterns are not necessarily fixed. Gillath and colleagues (2025) emphasize that attachment involves both relatively stable tendencies and more context-dependent processes that can shift depending on the relationships and situations we encounter. In other words, the people we have known before may influence what we expect, but they do not completely determine how we will experience every new relationship.
The goal is probably not to enter relationships without a past. That would be impossible. Our earlier relationships help shape the expectations, memories, and versions of ourselves that we carry forward. But a new relationship also offers new information. The people we have loved, trusted, feared, or lost may continue to influence the way we approach closeness. Still, the person in front of us is not the person who came before them. Perhaps one of the challenges of intimacy is allowing someone new enough space to actually be new.
Kaynakça
- References
- Andersen, S. M., & Chen, S. (2002). The relational self: An interpersonal social-cognitive theory. Psychological Review, 109(4), 619–645.
- Bowlby, J. (1969). Attachment and loss (No. 79). random house.
- Brumbaugh, C. C., & Fraley, R. C. (2006). Transference and attachment: How do attachment patterns get carried forward from one relationship to the next? Personality and Social Psychology Bulletin, 32(4), 552–560.
- Dagan, O., Nivison, M. D., Bleil, M. E., Booth-LaForce, C., Waters, T. E. A., & Roisman, G. I. (2024). Longitudinal associations between attachment representations coded in the Adult Attachment Interview in late adolescence and perceptions of romantic relationship adjustment in adulthood. Infant and Child Development, 33(4), e2512.
- Gillath, O., Manchanda, N., Brumbaugh, C. C., McIntyre, K. P., & Mattingly, B. A. (2025). Attachment theory and the relational self. Self and Identity. Advance online publication.
- Pietromonaco, P. R., & Feldman Barrett, L. (2000). The internal working models concept: What do we really know about the self in relation to others? Review of General Psychology, 4(2), 155–175.