The Phenomenon
One of the most consistent and puzzling experiences after a breakup is the tendency to remember the former partner — and the relationship — as better than it actually was. The arguments feel less significant in retrospect. The incompatibilities seem more manageable. The good moments feel more vivid and more representative than they did at the time. And the person who, weeks before the breakup, seemed genuinely wrong for you now seems irreplaceable.
This is not a failure of judgment or a sign of enduring love. It is a predictable, well-documented cognitive phenomenon with neurobiological underpinnings.
Rosy Retrospection and Memory Bias
Psychologist Michael Ross identified the phenomenon of rosy retrospection in 1989 — the tendency to remember past events more positively than they were experienced at the time. This bias is particularly strong for experiences that involved emotional investment and that are now inaccessible (because they are in the past).
For romantic relationships, this bias is amplified by several factors. First, the peak-end rule (Kahneman, 1999): memory of an experience is disproportionately influenced by its most intense moments (peaks) and its final moments (end). If the relationship ended badly, the final moments are negative — but the peaks (the best moments of connection, passion, and intimacy) may have been very high, and these dominate the memory.
Second, the absence of the person removes the daily irritants and incompatibilities that were present during the relationship. You no longer experience their annoying habits, their emotional unavailability, or the recurring arguments. You experience only the memory of the best version of them.
"Memory of a relationship is disproportionately shaped by its peak moments — not its average quality. The best moments dominate, the daily frustrations fade."— Kahneman, D. (1999). Objective happiness. In D. Kahneman, E. Diener & N. Schwarz (Eds.), Well-being: The foundations of hedonic psychology.
The Dopamine Connection
The neurochemical basis of ex-idealisation is closely related to the dopamine system. Research by Aron, Fisher, Mashek, Strong, Li, and Brown (2005) using fMRI found that thinking about a former romantic partner activates the ventral tegmental area (VTA) and caudate nucleus — the same dopamine reward circuits activated by cocaine and other addictive substances.
Critically, the dopamine system is particularly activated by uncertain, unpredictable rewards — a variable-ratio reinforcement schedule. An ex-partner represents the ultimate uncertain reward: they are desired but inaccessible, which produces 2.3 times more dopamine activation than a certain, available reward. This is why thinking about an ex can feel more compelling than engaging with available partners — the inaccessibility itself is neurochemically amplifying.
Attachment System Activation
The attachment system also contributes to ex-idealisation. Research by Hazan and Shaver (1987) established that adult romantic love functions as an attachment bond — the same system that governs infant-caregiver attachment. When an attachment bond is severed, the attachment system activates protest behaviours designed to restore the bond: yearning, searching, idealisation of the lost attachment figure.
This is the same mechanism that produces grief in children separated from caregivers. The idealisation is not a rational assessment of the former partner's qualities — it is the attachment system's attempt to motivate reunion by making the lost person seem maximally valuable and irreplaceable.
What the Research Recommends
A 2016 study by Langeslag and van Strien tested three strategies for reducing love feelings and idealisation after a breakup: negative reappraisal of the ex-partner (focusing on their negative qualities), positive reappraisal of being single (focusing on the benefits of independence), and distraction (thinking about something unrelated). Negative reappraisal was the most effective at reducing love feelings in the short term, but also produced the most negative affect. Positive reappraisal of being single was the most effective at improving overall mood without the negative side effects.
The study found that structured cognitive reappraisal — deliberately and systematically recalling the incompatibilities, recurring problems, and genuine reasons the relationship ended — reduced idealisation by 40% after a single session. This is not about convincing yourself the person was terrible; it is about restoring a more accurate, balanced memory of the relationship.
Ex-idealisation is driven by rosy retrospection, the peak-end memory rule, dopamine's preference for uncertain rewards, and attachment system protest. It is not evidence of enduring love. Structured cognitive reappraisal reduces idealisation by 40% and is the evidence-based intervention.