Dream recall endophenotypes?

Dreams are important part of psychiatric practice.  A discussion of dreams comes up in a number of contexts ranging from diagnoses like Post Traumatic Stress Disorder to primary sleep problems like Nightmare Disorder.   Dreams can be affected by substance abuse and medications.  Some people are still interested in what a dream might mean or they have their own interpretation that they want to discuss.  Sleep is often a source of stress to people who come in to see psychiatrists and questions about dreams frequently come up in discussion about too much sleep or too little sleep.  As a result, I have done a lot of reading and study about sleep and dreams.  I have the last 5 editions of Kryger, Roth and Dement's Principles and Practice of Sleep Medicine and additional texts and journals.  Since I worked in a residential settings, I see people who have their sleep observed and can tell me if they have apneic episodes or behavioral problems associated with sleep and refer them for polysomnography.  Whenever I ask about sleep there are a significant number of people who tell me: "I never dream."

Is it possible that a person is not dreaming at night?  Since the discovery of REM sleep it is well known that this biological process and dreaming are inextricably linked.  Dream researchers have determined that dream recall is influenced by a number of factors including the setting, whether a person is awakened slowly or rapidly and the sleep stage that they are awakened from.  For example, awakenings form REM sleep can result in 4 or 5 dream narratives per night.  Writing dreams recalled the next morning is not likely to produce that amount of content.

When an article suggesting a marker for differences in dream recall showed up on my Facebook feed I was naturally interested.  The authors in this case had a pool of 1,000 people who completed questionnaires indicating an interest in the study.  They were contacted by phone and asked the question: "on the average, how many mornings in the week do you wake up with a dream in mind?"  That is an important distinction from the people I talk with because they usually say: "I dream a lot." or "I don't dream at all."  For the purpose of this study the authors defined high recallers (HR) as those who recalled dream narratives or images on three mornings per week(4.42 ± 0.25 SEM dream recalls/week).  Low recallers (LR) recalled narratives or images per month (0.25  ± 0.02).   The subjects underwent standard polysomnography and an experimental paradigm that involved presenting a recorded voice saying first names through headphones in the alert and REM state.  Event related potentials (ERPs) and alpha frequency (8-12 Hz) responses to the auditory hallucinations were recorded.        

The authors summarize their data using the above graphics.  The top graphic is a little confusing at first if you are used to seeing similar graphics from QEEG analysis.  It is only alpha spectrum and the white lines represent occurrences of the auditory stimulus.  The bottom row shows the HR - LR power and the significant difference at the Pz electrode.  The black and white graphics at the bottom show ERPs and alpha power in response to first names for HR, LR, and HR-LR.  In general the alpha power decreases during wakefulness and increases during  REM sleep on all graphics.  The HR group had a more sustained decrease in alpha power to first names at 1000 to 1200 ms during wakefulness.

The authors go on to discuss the implications of these findings including the theory that increased alpha power during REM sleep could imply microarousals without awakenings.  A second hypothesis is that increased alpha power during REM sleep implies cortical deactivation rather than microarousal that would lead to decreased processing and less likelihood of awakening.  The authors interpret the greater reactivity in ERPs and alpha activation in the HR state as indicating that alpha is associated with activation in sleep.  They point out that the increased intrasleep wakefulness being great in HR is consistent with that observation.  They go on to point out that this trait may be central to a personality organization and cognitive substrate within the brain.  They pose a larger question about moving from one phenotype to the other.  They make the important observation that a hippocampus needs to be in the loop for dream recall and that there may be a point where functional imaging will be able to provide that level of detail.

I could not help but wonder if dream recall is a possible endophenotype.  What would happen if families were studied on their ability to recall dreams?  Would there be characteristic findings on polysomnography?  What would the pattern of heritability be and what would lead to transitions between phenotypes?  Sleep medicine is one of the areas of psychiatry where there are clear and valid biomarkers and it would be interesting to look at those differences.  In the meantime, it appears that what I have been saying to people about possible REM related dreams seems to be true based on this study.  Microarousals - probably from a number of possible etiologies will probably increase dream recall of characteristic REM type dreams and you may not actually experience interrupted sleep.  There is also the interesting consideration of dreaming without the hippocampus being engaged and have no dream recall on that basis.

George Dawson, MD, DFAPA

Ruby PM, Blochet C, Eichenlaub J-B, Bertrand O, Morlet D, Bidet-Caulet A (2013) Alpha reactivity to first names differs in subjects with high and low dream recall frequency. Frontiers in Psychology 4.

All of the figures in this post are from the above reference and are produced here via Creative Commons license.  Please see the original article for all of the details.