Rosy retrospection

Rosy retrospection is a proposed psychological phenomenon of recalling the past more positively than it was actually experienced.

The highly unreliable nature of human memory is well documented and accepted amongst psychologists. Some research suggests that both negative and positive emotions are exaggerated in memory.

Despite being a cognitive bias w that distorts a one's view of reality, it is suggested that rosy retrospection serves a useful purpose in increasing self-esteem and sense of well-being.

Simplifications and exaggerations of memories such as occur in rosy retrospection may make it easier for the brain to store long-term memories, as removing details may reduce the burden of those memories by requiring the generation and maintenance of fewer neural connections.

Declinism, the predisposition to view the past more favourably and the future more negatively, may be related to cognitive biases like rosy retrospection.

Rosy retrospection is very closely related to the concept of nostalgia, though the broader phenomenon of nostalgia is not usually seen as based on a biased perspective.

The English idiom "rose-colored glasses" or "rose-tinted glasses" refers to perceiving something more positively than it is in reality.

The Romans occasionally referred to this phenomenon with the Latin phrase "memoria praeteritorum bonorum", which translates into English roughly as "memory of good past", or more idiomatically as "good old days".

Research
In one group of experiments, three groups going on different vacations were interviewed before, during, and after their vacations. Most followed the pattern of initially positive anticipation, followed by mild disappointment thereafter. Generally, most subjects reviewed the events more favorably some time after the events had occurred than they did while experiencing them.

A 2003 pair of studies tracked 68 and 117 undergraduates, suggesting rosy retrospection is caused by high self-esteem. Participants journaled the day's events and associated emotions each night for seven nights. They later recalled their emotions when asked about said events. Those with higher self-esteem recalled their positive emotions being stronger than they journaled. They did not also recall their negative emotions more strongly. However, this result varied in its strength and did not occur consistently.

A 1995 study tracked 30 employed adults over 2 working weeks, having them report their mood every 2 hours during their waking day, as well as end-of-day and end-of-week reflection. It suggests a rosy bias which grows with time. For positive emotions, it found that end-of-day reflections were stronger than an average of the 2-hourly ratings of that day; likewise end-of-week reflections were stronger than an average of the end-of-days. But for negative emotions, there was no such significant difference neither between the averaged hourly and daily ratings nor the averaged daily and weekly ratings.

Exaggeration of both negative and positive emotions
Some studies have found evidence of a bias to exaggerating negative emotions - a.k.a. a 'blue' retrospective - as well as positive ones.

A 2016 study of 179 adults tracked their emotional state at regular intervals over 10 days, upon reflection after one day, and again after 1-2 months. It found that for both positive and negative emotions, stronger peak emotions (the strongest rating of the day) were more likely to result in exaggerated recollections upon reflection. Unlike the study above, it did not find that this effect increased with time. It also found a negative correlation with the average rating and the exaggerated recollections; suggesting those who consistently experienced stronger emotion recall more accurately. Additionally, it found extraverted personalities were more likely to have ‘rosy’ positive bias whereas neurotic personalities were more likely to have negative 'blue' bias on recall.

A 2021 work studied a group of 120 Swiss children aged about 12, later repeating the study on the same group aged about 15. For a week, the children filled in short emotional questionnaires at random points during their school day. Afterwards, they were asked to recall their week’s emotions in retrospect. Note they were asked only about the preceding week: the 15-year-olds were not asked to recall their emotions at age 12. It found evidence of a ‘rosy’ positive bias for the 12-year-olds. But this was the opposite for the 15-year-olds, who showed a 'blue' negative bias instead.

A 2003 study surveyed 41 participants around the time of their vacations. Subjects predicted their emotions before vacating, reported their emotions during (in-situ), and recalled their emotions after. It indeed found a rosy effect as subjects recalled (and predicted) their positive emotions being stronger than they actually were. But it also found recollections of negative emotions were recalled and predicted more intensely than was reported at the time (as an aside, the only significant predictor of a desire to repeat a holiday was the recalled emotions, but not the predicted nor in-situ reports).

Limitations
Relying on subjective ratings, the above studies could suffer from demand characteristics: participants may guess the study's goals and expected results, unconsciously changing their answers thinking they are 'supposed' to recall their emotions inaccurately.

These studies may be vulnerable to sample biases. They rely on small samples which are more likely to be unrepresentative of the general population by unlucky chance. Many of the samples are homogenous to varying extents, with subjects often being relatively young, in education, and western. A similar sample bias may occur in the way researchers find subjects. Potentially those who come across and are appealed to join and remain in studies will be those with relatively more free-time, better education, higher wealth and income, etc. Though the studies took varying efforts to reduce this by ensuring a balance of ages, ethnicities, sexes, etc.

These studies typically asked subjects to recall their emotions only days or weeks after an event. Thus they may predict little for rosy retrospection on the scale of months, years, and decades.

A suggested cause of such findings may be in the social and linguistic norms of the subjects, rather than their actual emotions. Especially if a subject fails to fully recall their emotions, social convention may bias them to more positive terms in an attempt to answer. Though this raises the question as to if evidence exists or can be found of such a norm and bias.