Abstract
Emotion exerts great impacts on memory. For instance, negative emotions can enhance memory encoding or reduce retrieval-induced forgetting. In addition, congruent mood states between encoding and retrieval (e.g., negative emotions at both stages) can also improve memory performance relative to incongruent mood states (e.g., negative emotion in encoding, but positive encoding in retrieval). However, it is unclear whether mood congruency between memory retrieval and encoding increases the probability of successful retrieval or enhance the quality of retrieved memory representations. The current study directly tested these two hypotheses using a visual long-term memory recall task. In the study phase, participants remembered a sequence of colorful objects. Each object was presented on top of a gray-scale negative image from the International Affective Picture System (IAPS). Participants gave a valence rating for each IAPS image for emotion induction. In the test phase, participants reconstructed the colors of the previously studied objects by continuously adjusting their colors under negative or positive emotional contexts using the same emotion induction procedure as that from the study phase. Negative IAPS images used for the negative emotional context were different from, but with matched valence as, those in the study phase. Positive images for the positive emotional context had comparable arousal levels as negative images. We found that memories encoded in negative context but retrieved in positive context (the incongruent condition) were less precise than memories encoded and retrieved both in negative context (the congruent condition). In contrast, no significant difference in the probability of successful retrieval was found between the two conditions. Follow-up experiments ruled out alternative interpretations that either negative emotion or positive emotion at the retrieval alone could account for the memory quality effects observed in the first experiment. Taken together, the present results support a resolution account for mood-congruency memory effect.
Meeting abstract presented at VSS 2015