Skip to main content

Verified by Psychology Today


What Leads People Back to an Ex

Breakups can make someone's self-concept less clear.

Key points

  • Relationship dissolution can often lead to rekindling desires.
  • People with insecure attachment styles might be particularly susceptible to trying (or wanting) to rekindle old relationships.
  • Attachment issues contribute to problems with self-concept, which in turn can lead to a desire to rekindle.
 Annie Spratt/Unsplash
Source: Annie Spratt/Unsplash

Research published in the peer-reviewed Journal of Social and Personal Relationships has investigated why some people might want to rekindle their past relationships after a break-up.

In two studies, Morgan Cope (Florida Atlantic University) and Brent Mattingly (Ursinus College) explored the effects of attachment style and self-concept on motivations to get back with ex-partners after a romantic relationship had ended. This is a particularly important topic given the number of relationships that fail, and the potential effects on mental health that can follow from such a change to a person’s life circumstances.

Attachment theory is widely studied in psychology as a cornerstone of developmentally-focused research. The theory suggests that a number of different attachment styles might drive our behavior in relation to other people based on the configuration of our internal working model. This sets out how positive/negative we feel about ourselves and others, with the resultant judgments leading to the attachment style that becomes dominant for each person.

Three major attachment styles tend to be considered within the literature:

  1. People with a secure attachment style tend to be confident and self-assured. They form bonds with others that are mutually beneficial. They can be in relationships or have ties with people without needing to always be around them, but are safe in the knowledge that their attachment figure will be available to them if needed.
  2. People with an anxious-insecure attachment style tend to be clingy with their attachment targets. They are quick to form attachments and are sensitive to criticism. Relationship behaviors for people with this attachment style tend to suggest a firm fixation on their partner, with jealousy being a prominent feature of their experience.
  3. People with an anxious-avoidant attachment style tend to be averse to forming close intimate bonds with others. They are independent and often behave in a way that suggests a non-trusting nature. Self-sufficiency is something that characterizes their relationships to an extreme degree, which often leads intimate partners to be driven away.

The core motivation underpinning Cope and Mattingly’s research was a desire to understand how people might begin to make sense of (and cope with) relationship break-ups. Quoted in PsyPost, Morgan Cope suggested:

Although we may not want them to, romantic relationships often end, eliciting a range of emotions. Commonly these emotions are negative and range from moderate emotional discomfort to severe distress… Our study considered relationship rekindling—wanting to get back together with an ex—as one avenue people may consider for relieving breakup-related distress.

What Did the Researchers Do?

Cope and Mattingly report two studies in their paper, with 181 and 180 participants, respectively. Both followed a similar method, using self-reported survey designs to measure the following concepts:

  • Attachment anxiety (e.g., “I need a lot of reassurance and love from my partner”).
  • Attachment avoidance (e.g., “I am nervous when my partner gets too close to me”).
  • Self-concept clarity (e.g., “My beliefs about myself often conflicted with one another”).

The outcome in both studies was "rekindling desire," which was defined as a wanting to rekindle or restart their last relationship, or to seek out more contact with their last intimate partner. The only difference was that Study 1 looked at how much participants wanted to rekindle a previous relationship directly after the break-up (i.e., participants were recalling their feelings at that particular point in time), and Study 2 looked at how much they wanted to rekindle when completing the survey. All relationships under consideration had ended within the previous two years.

The analyses of both studies revealed a simple and consistent mediated relationship, where having higher levels of attachment anxiety predicted a greater desire to rekindle a past relationship, but this happens because the break-ups cause a reduction in the clarity of somebody’s self-concept.

In other words, being anxious in your attachment style during a break-up seems to cause you to doubt your sense of self, which motivates you to try to rekindle and restart the relationship.

Going Back Isn't (Always) the Answer

Although this study was principally correlational in nature (which means that causal arguments technically cannot be made), the results do seem to suggest that self-concept issues may drive motivations to get back with ex-partners. To properly test the causal nature of this, it would be necessary to follow couples over time to track changes in self-concept and attachment style in order to assess whether these are causally related to rekindling motivations after a break-up.

Cope and Mattingly close their paper by stating that while "rekindling may temporarily restore individuals’ identities, there could be deleterious long-term consequences that make them more vulnerable to future self-concept disruptions." This is because, in the long term, rekindled relationships appear to be of lower quality and are more likely to end in breakups than first-time bonds. With this in mind, those who rekindle past relationships to resolve issues with their sense of self may ultimately be setting themselves up for bigger issues down the road.

Facebook image:


Cope, M. A., & Mattingly, B.A. (2021). Putting me back together by getting back together: Post-dissolution self-concept confusion predicts rekindling desire among anxiously attached individuals. Journal of Social and Personal Relationships, 38(1), 384-392.