Fading Affect Bias: Why Bad Feelings Fade Faster Than Good Ones

The bias is a property of the feeling, not of the memory. In the diary studies that established it, people remembered their pleasant and unpleasant events about equally well; what changed was how much emotion the memories carried when they came back. Unpleasant events lost intensity quickly, pleasant ones slowly, and the gap widened with time. Walker, Skowronski and Thompson summarized the result in 2003 under a title that states the claim: life is pleasant, and memory helps to keep it that way. They also noted that it runs against the general rule that bad is stronger than good; in autobiographical memory the bad loses its grip first.

Several processes contribute. Talking about an event changes it: Skowronski, Gibbons, Vogl and Walker found in 2004 that disclosing a memory to others speeded the fading of negative affect and slowed the fading of positive affect, and that events disclosed more often showed the stronger effect. Rehearsal that reframes a bad event, finding what was learned or how it was survived, does similar work. A self-protective motive is usually proposed as well: holding onto the sting of every failure would make action harder, and Walker and Skowronski argued in 2009 that the bias helps maintain the optimism needed to face new difficulties. Ritchie and colleagues' typology of how individual memories change, reported in that review, is useful here. Across the events in their 2009 data, fading was the most common pattern, at 47 percent; for negative events it was 51 percent, while positive events most often kept their intensity, at 49 percent. The bias is a difference in averages built from many memories that behave differently.

The effect is robust across methods and samples. It appears with diary methods and retrospective ratings, with bipolar and unipolar scales, and across varied samples, including British and Native American participants. It also has limits that make it informative. Dysphoria disrupts it: people with depressive symptoms show a weaker bias or none. The bias is also smaller for events that are rarely rehearsed or shared. Both patterns point the same way, toward the social and reflective processing of memories as the engine of the effect rather than simple decay.

Early evidence came from David Holmes in 1970, who asked 26 people to record events in a diary with the intensity of the emotion each prompted, and found a week later that the emotion attached to unpleasant events had faded more than the emotion attached to pleasant ones. The modern line of work begins with W. Richard Walker, Rodney Vogl and Charles Thompson's 1997 paper in Applied Cognitive Psychology, "Autobiographical memory: Unpleasantness fades faster than pleasantness over time."

Their method avoided the problem of asking people to remember how they once felt. Participants kept diaries of unique events and rated each one, when it happened, on a 7-point scale from extremely unpleasant through neutral to extremely pleasant. The diaries were collected weekly. At the end of the recording period participants returned to the laboratory, were shown the events they had written down, rated how pleasant or unpleasant each felt now, and rated how well they remembered it. Three studies covered three retention intervals, 3.5 months, a year and 4.5 years, with samples shrinking from dozens to six to one.

Affect faded at every interval, and the longer the interval the more it faded, but the fading was unequal. The drop in intensity for unpleasant events was larger than the drop for pleasant events in all three studies. Memory for the events themselves showed the ordinary pattern of forgetting, with only a slight advantage for pleasant events: people remembered the good and the bad with almost equal clarity while the bad lost its emotional charge faster. Walker and colleagues later confirmed the pattern with other samples, and Ritchie, Skowronski, Wood and colleagues reported in 2006 that the bias was smaller for events rated as more important to the self, and larger for events that had been talked about with others or privately savored or worked through, while other kinds of private rehearsal did not. The 2014 review by Skowronski, Walker, Henderson and Bond in Advances in Experimental Social Psychology collects the work to date.

The rows below are illustrations, not findings from the studies cited on this page.

For most purposes the bias does not need countering; it is part of what lets people carry a past full of ordinary setbacks. It needs countering only where the fading of a bad feeling would lead to a bad decision, and there the remedy is a record.

• Rate at the time. A diary line with a number attached, as in the original studies, preserves the intensity that memory will not. The Mood Tracker linked below is one way to do this.
• Before repeating a choice, read the record rather than consulting the feeling. The feeling will have faded on the unpleasant side; the record has not.
• Separate the event from the emotion. Memory for what happened stays about as accurate for bad events as for good ones, so ask yourself what happened, in order, before asking how it felt.
• Notice the role of retelling. Talking about an unpleasant event speeds the fading of its emotion, which is often what you want. If the aim is to keep a lesson alive, write it down as well as talking about it.
• Know the boundary of the effect. The diary research finds that people experiencing depressive symptoms often show no fading affect bias, so the pattern described on this page is a feature of ordinary memory rather than a rule everyone can count on.

Several biases make the past look better than it was, and they work on different parts of memory.

• It is not rosy retrospection, which is a change in how an experience is evaluated: a vacation rated higher afterward than it was rated during. The fading affect bias is a change in how much emotion a memory carries, and it is measured by comparing feelings about pleasant and unpleasant events over the same interval.
• It is not the positivity effect, the finding that older adults attend to and remember relatively more positive than negative material. That is a difference between age groups in what gets in; the fading affect bias is a difference within one person in how feelings change after the fact.
• It is not the opposite of negativity bias, the greater weight of negative information in attention and judgment. The two coexist: bad events hit harder at the time, and their emotional charge fades faster afterward.
• It is not the peak-end rule, which concerns how an episode is summarized from its most intense moment and its ending rather than how the summary changes over months.