Circadian and Sleep Modulation of Dreaming in Women with Major Depression
Sydney Aten
Round 1
Reviewer 1 Report
This paper from Brichler-Pedross and the colleagues describes a study to evaluate the relationship between dream recall and circadian rhythms in major depressive disorder (MDD) patients. The authors first hypothesized that reduced dream recall may be linked with circadian clock, as well as different sleep stages, especially rapid eye movement (REM) stage. They conducted a sleep test and a dream recall survey with a multi-nap protocol. They found unexpected high emotional load in women with MDD, which was peaked at nighttime. Moreover, this higher emotion load in patients was not associated with NREM/REM modulatio. Finally, the authors found that dream emotional load is correlated with the subjective perception of mood in both healthy and MDD women.
Overall, the scientific question this manuscript addressed is important and timely. The unexpected findings of increased dream recall and emotion load in unmedicated women with MDD is intriguing, which will be of broad interest to many readers. The experimental design is logic, and the paper is clear-structured and well-written. I think this is an interesting paper and only have a few comments to improve it prior to publication.
- Is the multiple nap protocol well controlled?
- Are the data obtained by this protocol well represented the ground truth? Is there any difference of the sleep parameters (e.g. total sleep time, REM/NREM, melatonin oscillation) between normal sleep and the protocol used in this study? disruption elevated emotion load?
- Line 51, an extra space between “dreaming” and “occurs”.
- Line 52, explain “NREM”.
- Line 204, need reference.
Author Response
- Is the multiple nap protocol well controlled?
Response: We thank the reviewer for the constructive comments and for the positive assessment of the manuscript. The multiple nap protocol is a validated well-controlled circadian laboratory protocol designed to tease apart the relative contribution of the circadian timing system on a given outcome of interest (PMID: 22390241). This is because such laboratory protocol controls for the masking effects of e.g., changes in light exposure (< 8 lx, typically 3–5 lx at the angle of gaze, during scheduled wakefulness and 0 lx during scheduled sleep). Moreover, it controlled for ambient temperature/humidity, posture (i.e., semi-recumbent posture in bed during wakefulness and supine during sleep), no time cues, and hourly isocaloric meals and water, among others (PMID: 11837947). Because participants completed ten alternating cycles of 75 min of scheduled sleep and 150 min of scheduled wakefulness, homeostatic sleep pressure was maintained low. Thus, the effects of the circadian timing system – without the effects of increased sleep pressure – can be observed on an outcome of interest. We now include this exact explanation on Page 11 (section “4.2. Study design).
- Are the data obtained by this protocol well represented the ground truth? Is there any difference of the sleep parameters (e.g. total sleep time, REM/NREM, melatonin oscillation) between normal sleep and the protocol used in this study? disruption elevated emotion load?
Response: Participants completed ten alternating cycles of 75 min of scheduled sleep and 150 min of scheduled wakefulness, as a sleep cycle below 75 min would not include sufficient REM sleep. In a previous study using the same 40-h multiple nap protocol (PMID: 12217973), the polysomnographic structure (i.e., total sleep time, NREM and REM sleep) of the baseline (normal) sleep and that of each sleep cycle within the nap were robustly associated. As such, we strongly believe the current findings represent the NREM/REM sleep modulation underpinning dreaming. Furthermore, endogenous melatonin rhythms are strongly driven by the circadian timing system, which was not disrupted by this laboratory circadian protocol. Therefore, the melatonin profiles illustrated here very likely reflect endogenous circadian melatonin rhythms.
- Line 51, an extra space between “dreaming” and “occurs”.
Response: We corrected the typo accordingly.
- Line 52, explain “NREM”.
Response: We included the term NREM (non-rapid eye movement) sleep on Page 2.
- Line 204, need reference.
Response: We added the reference accordingly.
Reviewer 2 Report
Overall, the manuscript by Pedross et al is very well designed and well written. In particular, it is important that the study was conducted in subjects who were in approximately the same menstrual (i.e. follicular) phase. I only have a minor comments (see below) that should be addressed.
-For the Figure 1 legend, please add a description for the 'PSG' abbreviation found in the schematic (not all readers will be familiar with sleep research terminology)
-For Figure 2, please put the x axis titles/labels on all graphs (not just the bottom graph). This will help with ease of viewing for readers
-For Figures 4 and 5, why are data not plotted on one (single) graph (i.e. why are they separated into 2 different graphs for each parameter measured)? If possible, all 4 groups should be plotted on one graph (and statistical measures should be reported across the 4 conditions).
-Please read through the manuscript again and make minor changes to sentence structure, etc. See the following notes that need to be addressed:
*Line 40--eliminate 'e.g.'
*Line 43--'work' not 'worker'
*Line 64--'latter' may be better here
*Line 71--should read "...which is purported to be the sleep stage most conducive..."
*Line 194-195 is not a complete sentence
*Line 334--'saliva samples were collected' (avoid redundancy)
Author Response
Overall, the manuscript by Pedross et al is very well designed and well written. In particular, it is important that the study was conducted in subjects who were in approximately the same menstrual (i.e. follicular) phase. I only have a minor comments (see below) that should be addressed.
-For the Figure 1 legend, please add a description for the 'PSG' abbreviation found in the schematic (not all readers will be familiar with sleep research terminology).
Response: We thank the reviewer for the constructive comments and for the positive assessment of the manuscript. We now included the term PSG (polysomnography) on the legend of figure 1.
-For Figure 2, please put the x axis titles/labels on all graphs (not just the bottom graph). This will help with ease of viewing for readers.
Response: We now included the x-axis titles/labels on all panels of Figure 2.
-For Figures 4 and 5, why are data not plotted on one (single) graph (i.e. why are they separated into 2 different graphs for each parameter measured)? If possible, all 4 groups should be plotted on one graph (and statistical measures should be reported across the 4 conditions).
Response: On section 4.7. Statistical analyses (Page 12), we indicate that mixed-model analyses were conducted using between-factor “Group”, within-factor “Circadian day/night” and the interaction of “Group” and “Circadian day/night”, as suggested by the reviewer. We now present Figures 4 and 5 with all four groups plotted in a single panel for dream recall, number of dreams and dream emotional composite score, as suggested by the reviewer.
-Please read through the manuscript again and make minor changes to sentence structure, etc. See the following notes that need to be addressed:
*Line 40--eliminate 'e.g.'
*Line 43--'work' not 'worker'
*Line 64--'latter' may be better here
*Line 71--should read "...which is purported to be the sleep stage most conducive..."
*Line 194-195 is not a complete sentence
*Line 334--'saliva samples were collected' (avoid redundancy)
Response: We have corrected the listed typos accordingly and performed a full proofreading of the revised manuscript.
