What memory does to what we felt
The judgement passed on an episode depends on its most intense moment and on its ending, hardly at all on its duration. That holds for holidays as much as for questionnaires.
Two evaluations that diverge
One must distinguish the experience lived moment by moment, which can be collected by questioning someone while it unfolds, from the retrospective judgement passed afterwards. The work of Daniel Kahneman and colleagues has shown that the two do not coincide.
The retrospective judgement organises itself around two moments: the peak of intensity and the ending. The duration of the episode, for its part, weighs surprisingly little, a phenomenon called duration neglect.
The experiment that establishes it
The clearest demonstration compares two unpleasant medical examinations. The second is identical to the first, extended by a few less painful minutes. It therefore contains strictly more total discomfort, and yet it is retrospectively judged less painful.
The result has been recovered in other protocols, notably with cold water. It is not a curiosity: it indicates that memory is not an instrument for measuring experience, and that one cannot use it as an average.
What this does to questionnaires
An item asking how you have felt over the past month does not obtain an average. It obtains a judgement built from a few salient moments and from the way the period ended.
The best-built instruments take this into account by asking about typical situations rather than about periods to be evaluated globally. It is a detail of phrasing with real consequences for what the score means.
In-the-moment methods
To get around the problem, people are questioned at the moment itself. Experience sampling prompts participants at random moments of the day; day reconstruction, developed in 2004, asks them to cut the day into episodes and then evaluate each one.
These methods are expensive, suffer from significant drop-out, and are reactive: being questioned several times a day alters the attention paid to one's own state. They remain the reference for everything concerning daily experience.
A consequence for oneself
The same mechanism applies to what one recounts of one's own week, one's job or a relationship. A striking episode occurring at the end of a period colours the whole judgement passed on it, and that colouring then appears well founded.
This does not make memory a liar. It indicates that it answers a different question from the one we think we are asking: not how much discomfort there was, but what trace the period left.
Key points
The retrospective judgement of an episode depends on its peak and its ending, hardly at all on its duration.
A longer but better-ending examination is judged less painful than a shorter one: the total does not count.
Questionnaires bearing on a period inherit this distortion; in-the-moment methods partly get around it.
This article describes documented mechanisms and disagreements. It recommends no position, no party and no vote, and commits to no reading other than that of the work it cites.
And you, where do you stand?
The Civix test situates your convictions on six dimensions and matches you with one of the twelve profiles, without ever telling you what to think.
Start my test