Sleep and emotion modulate memory consolidation, and the role of sleep in memory consolidation is supported by multitude studies. Similar effects of sleep on consolidation have been found in most memory systems, including emotional memory. For example, Wagner et al. identified a selective benefit on retrieval of content of an emotional story when compared to retrieval of content of a neutral story, if sleep was in the retention interval. Payne et al. showed that a retention interval spent asleep led to selective consolidation of the negative objects at the cost of memory of the neutral ones. In summary, there are several studies in support of the hypotheses that emotional aspects of memory relative to neutral aspects of memory are selectively enhanced by sleep.
On the other hand, Lewis et al. found contradicting results. The authors showed their participants pictures with either negative or neutral contexts and compared the recognition performance after wake and sleep retention intervals. Their data showed a smaller decay of context memory in the course of sleep than during wakefulness; however emotional and neutral context memories were protected to the same extent during sleep.
Some studies discuss the function of different sleep phases in memory consolidation. Specifically, it is hypothesized that slow-wave sleep primarily enhances the consolidation of declarative memories. In addition, REM sleep is believed to selectively enhance the consolidation of emotionally tagged memory content while attenuating the emotion itself. In line with the earlier study by Wagner and colleagues, Groch et al. showed better recognition of emotional stimuli than of neutral stimuli after late, REM-rich nocturnal sleep compared to early, slow-wave rich sleep (SWS).
Furthermore, correlations between REM-sleep during a short nap or overnight sleep and the consolidation of emotional memory have been reported. In contrast, some studies failed to replicate any selectively enhancing effect of REM-sleep on the consolidation of emotional memory. In summary, there is some evidence in favor of the hypothesis that REM sleep fosters the consolidation of emotional memory. It is unclear, however, whether REM sleep is necessary for the consolidation of emotional memory.
According to Morgenthaler et al. study in 2013, the reduction of REM-sleep had no influence on memory consolidation. Participants (N = 29 healthy medical students) were separated into two groups: undisturbed sleep and selective REM-sleep deprived. Both groups also worked on the memory task in a wake condition.
The participants spent two nights in the sleep laboratory. The first night's purpose was to exclude severe sleep disorders like sleep apnea syndrome and to adapt the participants to the conditions in the sleep laboratory. At 10:00 p.m. before the second night in the laboratory, the participants were shown the stimulus pictures (two sets of 260 pictures – 130 emotional and 130 neutral), and then immediately tested with a small subsample of pictures afterwards to ensure sufficient encoding.
Sleep was recorded between lights off (approx. 10:40 p.m.) and lights on in the morning (approx. 6:45 a.m.) by standard procedures using a digital electroencephalogram (EEG), electromyogram (EMG), electrooculogram (EOG), and electrocardiogram (EKG). REM-sleep was classified according to the standard criteria of rapid eye movements, low muscle tone, and rapid low-voltage EEG and was monitored online by two trained psychologists. As soon as the first 30-s REM-sleep epoch was identified, participants were awakened. At the beginning of the night, it was usually sufficient to awaken the participants using an intercom. If participants did not wake up completely this way, one of the investigators went to the participant's room and addressed him or her personally. If this was not sufficient either, the participants were asked to sit up for a moment before being allowed to sleep again.
The participants performed the recognition test at 8:00 a.m. The results replicated the effect that sleep fosters the consolidation of declarative memories, regardless of the emotional arousal attached to the stimuli. After an 8 h retention interval of a night of sleep, the participants performed better in the recognition test than after a wake retention interval of the same length. In both conditions (wake and sleep), emotional pictures were better recognized than neutral pictures.
There was, however, no difference in the recognition accuracy (neutral and emotional) between the groups. In summary, the received data suggest, that sleep-dependent emotional memory consolidation does not solely rely on intact amounts of REM-sleep throughout a night of sleep.