Emotions are central in dreams, especially those occurring during REM sleep. The emotions in dreams play a key role in several theories of dreaming, such as those of Ernest Hartmann or Rosalind Cartwright, which suggest that dreams function to process and regulate emotions.

However, there is disagreement by researchers on the frequency and prevalence of emotions in dreams.  Are they more frequently emotional or neutral? Are they more negative or positive, or balanced? While some authors suggest that fear and negative emotion are more prevalent, others disagree.

These inconsistencies are likely due to differences in the methods that researchers use to collect and analyze dream content. Dreams can be recorded in a sleep laboratory or in a home setting; they could be rated by the dreamer themselves or rated by external judges using a specified scoring system; emotions may be qualified simply by their valence or intensity, or they could be categorized into happy, sad, angry, afraid, and so on.

Given these variations across studies, researchers at Turku University in Finland set out to investigate how different methods of dream collection and analysis might influence results on the frequency, valence, and content of emotions in dream reports.

Over the course of three research studies, the researchers compared the emotion content of dreams based on two factors: first, who is rating the dream’s emotion, the dreamer or a judge?; and second, in which environment were dream reports collected, at home or in the lab?

Regarding the first factor, in the majority of studies dream reports are collected and then analyzed by independent judges, termed external rating. Nevertheless, some researchers prefer to have participants themselves rate their dreams, termed self-rating. Regarding the second factor, typically dreams are written down in the morning at the participant’s home upon awakening. Lab studies are also done in order to wake participants up specifically from REM sleep and ask them to immediately report their dream upon awakening.

In the first study, published in Consciousness and Cognition in 2014, the researchers compared self and external ratings of emotion for laboratory-collected REM sleep dreams. For this study, 17 participants spent two nights in the sleep lab, and were awoken after five minutes of continuous REM sleep. Upon awakening, they measured the emotions they experienced using a scale, the Differential Emotions Scale, which asked them to rate the presence of 10 positive and 10 negative emotions. These same dream reports were later analyzed by judges using the same scale, so that the self-ratings could be compared with external-ratings.

For this study, all the participants self-rated their dreams as having some emotional content, whereas only a third of the dreams were rated as emotional by external raters. Moreover, self-raters scored a majority of their dreams as positive, whereas external raters only scored 10 percent of dreams as positive. In other words, when scored by external raters, the dreams were scored as less emotional overall, and particularly less positive.

The authors then conducted a similar study in the home setting, in the forthcoming September 2017 issue of the American Journal of Psychology. In this study, dream reports were collected from 44 participants who kept a three week home dream diary, and then rated the emotions of their dreams using the Differential Emotions Scale. Again, the same dream reports were later analyzed by two independent judges. The authors found that, with self-ratings, significantly more dreams were rated as emotional (both positive and negative) than with external ratings, similar to the previous study. Further, self-raters scored significantly more of their dreams as having positive emotion compared to negative emotion, whereas external raters scored more dreams as being negative than positive.

In the third study, published in the Journal of Sleep Research this year, the authors compared home dream reports with laboratory dream reports. Eighteen subjects participated. The participants kept a one week home dream diary and reported their dreams in the morning upon spontaneous awakening. They also spent two nights in the sleep laboratory, during which they were awakened once early in the night and another time later in the morning, both times during REM sleep. Results showed that dreams reported in the home settings were rated as more emotional, and specifically more negative, than dreams reported in the laboratory. This difference was mainly due to the lab reports collected earlier in the night, which were less emotional than the lab reports collected in the morning, although home dream reports were still more negative than both early and late REM dreams. In general, this study confirmed that the time of night effects emotionality of dreams, and that setting may as well, with home reports being more emotional than those collected in the lab.

To summarize: 1) Dreams are mostly emotional when scored by self-raters both in the lab and at home, whereas they seem to be less emotional when scored by external raters, especially when reported in the lab; 2) Dreams are more positive than negative when scored by self-raters, but more negative than positive when scored by external raters.

In general, it is likely that external raters do not score emotion as high as self-raters, because the participants have a memory for the actual experience, its mood and general feelings, whereas external raters can only look for specifically articulated emotion in the dream report. This may be particularly relevant for rating positive emotions and general moods which may not come across in a dream report. For instance, it may be easier to express negative emotion than to express positive emotion (e.g. I was scared vs. it’s a beautiful day). Further, in many dream reports participants will focus mainly on the sensory environment, such as the visual setting of the dream, without giving much description of the mood and emotion. Overall, it may be easier for external raters to accurately judge negative emotion, while self-ratings seem to be more accurate for positive emotion. 


Sikka, P., Revonsuo, A., Sandman, N., Tuominen, J., & Valli, K. (2017). Dream emotions: a comparison of home dream reports with laboratory early and late REM dream reports. Journal of Sleep Research.

Sikka, P., Valli, K., Virta, T., & Revonsuo, A. (2014). I know how you felt last night, or do I? Self-and external ratings of emotions in REM sleep dreams. Consciousness and cognition, 25, 51-66.


You are reading

Dream Factory

How to Have Lucid Dreams

A recently published study compares techniques for inducing lucid dreams.

Animating Dreams and the Future of Dream Recording

Recent research uses body signals during sleep to power an animated avatar.

Dream Deprived: A Modern Epidemic?

A recent paper reviews how poor sleep habits are robbing us of our dreams.