Tell me all about it: Narrated memories are less emotional than imagined memories.

People often re-live memories by talking about them. Verbal thinking is usually less emotive than imagery-based thinking but it is not known if this finding generalises to recollection. We tested if narrating memories aloud reduces their affective charge compared with recollecting them using imagery. Participants were randomised to two conditions: imagery (recalling the memory silently as vividly as possible) or narration (describing the memory out loud as clearly as possible). After practicing with a neutral topic, they recalled three aversive (Experiments 1 and 2) or three happy (Experiment 3) memories using narration or imagery, and rated emotionality of the memory after each recall. Before and after the procedure, they completed the PANAS to measure effects on mood. Experiments 2 and 3 included a 24 h follow-up. Emotionality was consistently lower following narrated recollection than imaginal recollection: narrated  = 5.3,  = 2.5; imaginal  = 7.2,  = 2.0; effect size (difference in mean values divided by overall ) = 0.78. Negative affect increased after recollection of aversive memories and positive affect decreased, but there were no effects of condition upon mood. Recalling a positive memory had no effect on mood. Follow-up data showed no lasting effects of recall mode on availability of memories or mood. We conclude that narration of emotional autobiographical memories reduces the emotionality of the recollection, but does not differentially change mood compared with image-based recall.