If you keep ending up in the same relationship with different people, it is not bad luck, and it is not poor judgment. Repeating relationship patterns usually come from a mental pattern formed early in life: a set of expectations about what closeness is like and what you have to do to keep it. That pattern operates outside awareness, which is why it can keep repeating long after you have seen it clearly.
In my work with adults who carry complex childhood trauma, this is one of the most discouraging things people bring. They chose carefully this time. The person seemed different. And a year later they are standing in the same place, feeling the same thing. This article explains what the mental pattern is, how it recreates the same relationship, and what it takes to change it.
What a mental pattern is
In attachment theory, the technical name for this pattern is an internal working model. John Bowlby (1973), who founded attachment theory, used the term to describe what a child learns from repeated interactions with caregivers. The model contains two expectations. The first is about other people: when I need someone, how do they respond? The second is about the self: am I someone others want to respond to?
The model exists because it is useful. It lets a child predict what will happen and adjust. For example, if warmth mostly came after achievement, the child learns that love depends on performance. If a parent withdrew when the child was upset, the child learns that showing needs pushes people away. These conclusions were accurate for that home. The problem is that the model is later applied to everyone, as if it described how all relationships work.
It also matters how the model is stored. It is not stored as a belief that you can put into words and examine. Karlen Lyons-Ruth and her colleagues (1998) called it implicit relational knowing, meaning procedural knowledge about how to be with another person, similar to knowing how to ride a bicycle. It belongs to the implicit memory system, the same one described in the article on trauma flashbacks. This explains why a person can describe their pattern accurately and still repeat it: describing uses explicit knowledge, and the pattern runs on implicit knowledge.
How the pattern recreates the same relationship
The model does more than describe relationships. It shapes them, in three ways.
Attraction. The model affects who feels attractive. Jeffrey Young and colleagues (2003), who developed schema therapy, observed that people often feel the strongest attraction to partners who activate their early schemas, and called this schema chemistry. For example, someone who grew up trying to reach an unavailable parent will feel strong longing around an emotionally distant partner. That longing is familiar, so it is experienced as chemistry. A steady and available person produces much less of that feeling, and so can seem less interesting.
Interpretation. The model affects how ambiguous events are read. A slow reply or a quiet evening could mean many things. The model picks the meaning it already expects, such as “they are losing interest” or “I did something wrong.” Each time this happens, the model appears to be confirmed.
Behavior. The model affects what you do, and your behavior affects how the other person responds. If you expect to be left, you may pursue harder, go silent, test the other person, or give more than you can afford, which is where people-pleasing often appears. These behaviors tend to produce the reaction you were afraid of. A related finding comes from William Swann and colleagues (1992): married people with negative views of themselves were more committed to spouses who also viewed them unfavorably. In other words, being seen in a way that matches your self-image can feel more trustworthy than being treated well, even when that self-image is painful.
Together, these three processes explain how the partner can change while the relationship stays the same.
Why seeing the pattern is not enough
There are two reasons. The first is the one above: insight is explicit knowledge and the pattern is implicit knowledge, so understanding it in words does not change the system that runs it.
The second reason is that the pattern was originally a solution. It was the best way available to stay connected to the people you depended on as a child. Part of you still treats it as necessary for safety, so it does not switch off just because you now know it is outdated. This is why repeating a pattern is better understood as an old form of protection than as self-sabotage.
How therapy changes a mental pattern
A working model changes through experience that contradicts it, not through argument. The neuroscience behind this is called memory reconsolidation. Karim Nader and colleagues (2000) showed in animal studies that when a stored emotional memory is reactivated, it returns to an unstable state for a few hours and then has to be stored again. During that window the memory can be modified. This overturned the older view that emotional memories are permanent once they have formed.
Later research identified what opens the window. The memory has to be active, and at the same time something has to happen that the memory did not predict. This is called a prediction error, or mismatch. Richard Lane and colleagues (2015) proposed that this is the shared mechanism of lasting change across different forms of psychotherapy: an old emotional memory is activated, a new emotional experience contradicts it, and the memory is stored again in an updated form. This is still a model under study, but it explains why insight is not enough. Talking about a pattern does not reactivate the emotional memory, and without reactivation and mismatch together, nothing is rewritten.
There is also long-term evidence that early models are not fixed. Glenn Roisman and colleagues (2002) followed people from infancy into adulthood and found that individuals who were insecurely attached as infants could still go on to have secure and successful close relationships as adults.
In my practice I draw on AEDP, EMDR, and psychodynamic therapy. With relationship patterns, each part of the work supplies one of the conditions that reconsolidation requires.
The pattern shows up in the therapy relationship. At some point, a client will expect the therapist to be disappointed, bored, or overwhelmed, because that is what the model predicts. This means the old memory is active. If the therapist then responds differently from the prediction, that is the mismatch. Both conditions are present in the same moment, which is what allows the model to be updated and not only discussed.
The feelings under the pattern are felt, not only described. Lane and colleagues emphasize that emotional arousal is necessary: a memory has to be emotionally active before it can be modified. Under the pattern there is usually longing, and grief about what the child needed and did not get. Feeling these with another person present is what brings the old memory fully online.
The memories that built the model are reprocessed. For some people, specific scenes taught the expectation. EMDR reactivates those scenes under safe conditions so that they can be stored again as past events and stop guiding present reactions.
For people who are currently in a relationship, the pattern can also be addressed in couples therapy. There the focus is on the cycle between the two partners, in which each person’s model triggers the other’s.
What changes
What changes is the prediction itself, and this is different from self-control. Before the model is updated, the only way to act differently is to override it. You feel the alarm when a partner goes quiet, and you use effort to stop yourself from pursuing or withdrawing. That kind of override depends on the prefrontal cortex, and prefrontal function is weakened by stress (Arnsten, 2009). Relationship patterns are triggered under stress, which is why willpower tends to fail at exactly the moment it is needed.
After the model has been updated, the old reaction is no longer generated in the first place. Bruce Ecker and colleagues (2012) describe three markers of this kind of change: the cue that used to set off the emotional reaction no longer does, the behaviors that came from that reaction stop, and the change holds without effort. In practice, a partner’s slow reply is registered as a slow reply and not as a threat. Closeness with a steady person feels safe, because the brain is no longer predicting that closeness will be withdrawn.
The memory of what happened in childhood stays. What is rewritten is the expectation that was built from it.
Work with a trauma therapist in New York City
I am Alisa Wu, a therapist based in NYC working with adults who carry complex childhood trauma and the marks it leaves: patterns that leave you stuck in relationships, a relentless inner critic, the reflex to please, difficulty trusting, and recurring emotional pain. I offer individual and couples therapy in English and Chinese, in person in Manhattan and online across New York State.
If you are ready to understand why the same relationship keeps finding you, I would be glad to talk. Schedule a free consultation to see whether we are a good fit.
References
Arnsten, A. F. T. (2009). Stress signalling pathways that impair prefrontal cortex structure and function. Nature Reviews Neuroscience, 10(6), 410–422.
Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation: Anxiety and anger. Basic Books.
Ecker, B., Ticic, R., & Hulley, L. (2012). Unlocking the emotional brain: Eliminating symptoms at their roots using memory reconsolidation. Routledge.
Lane, R. D., Ryan, L., Nadel, L., & Greenberg, L. (2015). Memory reconsolidation, emotional arousal, and the process of change in psychotherapy: New insights from brain science. Behavioral and Brain Sciences, 38, e1.
Lyons-Ruth, K. (1998). Implicit relational knowing: Its role in development and psychoanalytic treatment. Infant Mental Health Journal, 19(3), 282–289.
Nader, K., Schafe, G. E., & Le Doux, J. E. (2000). Fear memories require protein synthesis in the amygdala for reconsolidation after retrieval. Nature, 406(6797), 722–726.
Roisman, G. I., Padrón, E., Sroufe, L. A., & Egeland, B. (2002). Earned-secure attachment status in retrospect and prospect. Child Development, 73(4), 1204–1219.
Swann, W. B., Jr., Hixon, J. G., & De La Ronde, C. (1992). Embracing the bitter “truth”: Negative self-concepts and marital commitment. Psychological Science, 3(2), 118–121.
Young, J. E., Klosko, J. S., & Weishaar, M. E. (2003). Schema therapy: A practitioner’s guide. Guilford Press.
