Characteristics of the memory sources of dreams: A new version of the content-matching paradigm to take mundane and remote memories into account

Abstract

Several studies have demonstrated that dream content is related to the waking life of the dreamer. However, the characteristics of the memory sources incorporated into dreams are still unclear. We designed a new protocol to investigate remote memories and memories of trivial experiences, both relatively unexplored in dream content until now. Upon awakening, for 7 days, participants identified the waking life elements (WLEs) related to their dream content and characterized them and their dream content on several scales to assess notably emotional valence. Thanks to this procedure, they could report WLEs from the whole lifespan, and mundane ones before they had been forgotten. Participants (N = 40, 14 males, age = 25.2 ± 7.6) reported 6.2 ± 2.0 dreams on average. For each participant, 83.4% ± 17.8 of the dream reports were related to one or more WLEs. Among all the WLEs incorporated into dreams dated by the participants (79.3 ± 19%), 40.2 ± 30% happened the day before the dream, 26.1 ± 26% the month before (the day before excluded), 15.8 ± 21% the year before the dream (the month before excluded), and 17.9 ± 24% happened more than one year before the dream. As could be expected from previous studies, the majority of the WLEs incorporated into dreams were scored as important by the dreamers. However, this was not true for incorporated WLEs dating from the day before the dream. In agreement with Freud's observations, the majority of the day residues were scored as mundane. Finally, for both positive and negative WLEs incorporated into dreams, the dreamt version of the WLE was rated as emotionally less intense than the original WLE. This result, showing that dreams tend to attenuate the emotional tone of waking-life memories towards a more neutral one, argues in favor of the emotional regulation hypothesis of dreaming.

Fig 1. Emotional valence of the dreamt version and the original version of the WLEs incorporated into dreams.

A) Emotional valence of the dreamt version (white bars) and the original version (black bars) of the WLEs incorporated into dreams for positive (rating > 5), neutral (rating = 5) and negative (rating < 5) elements. The dreamt version was scored as more neutral than the original version of the WLEs. B) Significant positive correlation between the emotional valence ratings of the original and the dreamt versions of the WLEs incorporated into dreams. Gray bars indicate the distribution of the ratings for the original and the dreamt version of the WLEs incorporated into dreams. Error bars represent 95% confidence intervals. *** p <.001.

Fig 2. Characters and places of the dreams.

Error bars represent standard error of the mean. **p<.01 ***p<.001.

Fig 3. Characteristics of waking life elements incorporated into dreams.

All waking life elements (black) and day residues only (grey). Error bars represent standard error of the mean. *p<.05.

Fig 4. Distribution of the dated WLEs incorporated into dreams as a function of their remoteness.

(A) Distribution of all the WLEs incorporated into dreams when 3 categories of remoteness are considered. (B) Distribution of the WLEs incorporated into dreams which happened between 1 day to 14 days when 4 categories of remoteness are considered. Error bars represent standard error of the mean. ***p <.001.

Fig 5. Emotionality of the dated WLEs incorporated into dreams as a function of their remoteness.

(A) Importance scores of the WLEs incorporated into dreams as function of their remoteness. (B) Emotional intensity as a function of remoteness. (C) Emotional tone as a function of remoteness. Error bars represent standard error of the mean. *p<.05—**p<.01—***p<.001.