Are real moods required to reveal mood-congruent and mood-dependent memory?

While simulating, or acting as if, they were either happy or sad, university students recounted emotionally positive, neutral, or negative events from their personal past. Two days later, subjects were asked to freely recall the gist of all of these events, and they did so while simulating a mood that either did or did not match the one they had feigned before. By comparing the present results with those of a previous study, in which affectively realistic and subjectively convincing states of happiness and sadness had been engendered experimentally, we searched for--and found--striking differences between simulated and actual moods in their impact an autobiographical memory. In particular, it appears that the mood-congruent effects elicited by simulated moods are qualitatively different from those evoked by induced moods, and that only authentic affects have the power to produce mood-dependent effects.