Rosy Retrospection
Rosy retrospection is the tendency to remember an experience more favourably than it was rated while it was happening. Terence Mitchell, Leigh Thompson, Erika Peterson and Randy Cronk established it in 1997 by tracking people through three real events, taking ratings in anticipation, during, and afterwards. Across all three, expectation ran more positive than the experience and recollection ran more positive than the experience. The event itself was the low point of its own life cycle, every time.
What it is
If memory assembles rather than retrieves, the next question is whether the assembly has a systematic direction or merely adds noise. Rosy retrospection is the answer for emotionally toned autobiographical material, and the reason it is worth more than a laboratory demonstration is the design that established it.
Mitchell, Thompson, Peterson and Cronk published "Temporal Adjustments in the Evaluation of Events: The 'Rosy View'" in the Journal of Experimental Social Psychology in 1997. They did not stage anything. They followed people through three real events: a holiday in Europe, a Thanksgiving break, and a three-week cycling trip in California. Ratings were taken before, in anticipation of the event; during it, as it was happening; and afterwards, in recollection.
The pattern was the same in all three. Anticipation ran more positive than the experience. Recollection ran more positive than the experience. Which puts the event itself at the bottom of its own life cycle, with the version imagined beforehand and the version remembered afterwards both sitting above the version that was actually lived.
That structure is what makes the finding useful rather than merely amusing. The comparison is not between a memory and some external record of what happened; it is between a memory and the person's own rating, given at the time, of how they were finding it. The same person, the same event, two moments. Whatever else is uncertain, the drift is internal and its direction is consistent.
The consequences run in two directions. Forward, any decision made by consulting memory about how a comparable event went is being made on an inflated estimate, which affects holidays, jobs, relationships and anything else people evaluate by remembering. Backward, it supplies the direction of travel that Reconstructive Memory leaves open. Loftus's work establishes that the assembly can be influenced; Mitchell and colleagues establish which way it tends to go when the material is a person's own emotionally weighted experience.
And it constrains the nostalgia literature in a specific way. The functions nostalgia reliably serves, measured across thousands of participants, are functions performed by a record that has been drifting upward since it was laid down. Whether the functions depend on the drift or merely coexist with it has never been tested, and that question is the argument of Your Nostalgia Is Lying to You.
The primary-cited status on this page is precise. The paper has been identified and its design and direction verified. Sample sizes for the three studies have not been verified, and no figure from them appears anywhere on this site.